 BRITISH COLUMBIA AIR QUALITY
DISPERSION MODELLING GUIDELINE




   British Columbia Ministry of Environment and Parks
            Environmental Protection Division
        Environmental Policy and Initiatives Branch
                     Air Quality Unit




             Victoria, British Columbia, Canada
                                                 Obtain a Copy

Copies of this document are available from the Ministry website.




                                              Guideline Comments

Comments or questions on the Guideline can be sent to: bcdispersion.model@gov.bc.ca

Users are encouraged to check or subscribe to the Ministry website for any Guideline updates.


                                                       Air Team
                                    Environmental Policy and Initiatives Branch
                                        Environmental Protection Division
                               British Columbia Ministry of Environment and Parks
                                            PO Box 9341, Stn Prov Govt
                                         Victoria, British Columbia, Canada
                                                      V8W 9M1




                                               Document History

July 2005:                DRAFT released

November 2006:            October 2006 version released

April 2008:               March 2008 version released (minor revisions)

December 2015:            November 2015 version released (major revisions)

November 2021:            October 2021 version released (minor revisions and the release of the GUIDANCE FOR
                          NO2 DISPERSION MODELLING IN BRITISH COLUMBIA)

August 2022:              July 2022 version released (minor revisions and the release of the LOWER FRASER
                          VALLEY PHOTOCHEMICAL MODELLING GUIDELINE)

July 2026:                July 2026 version released (minor revision and the release of the GUIDANCE FOR
                          DISPERSION MODELLING OF MINING-RELATED FUGITIVE PARTICULATE MATTER
                          EMISSIONS IN BRITISH COLUMBIA)
                             British Columbia Air Quality Dispersion Modelling Guideline



                                                              TABLE OF CONTENTS

    INTRODUCTION .............................................................................................................................................. 1
         PURPOSE OF GUIDELINES ..................................................................................................................................1
         THE APPLICATION OF DISPERSION MODELS: BC’S UNIQUE SCIENTIFIC AND REGULATORY CONTEXT ................................1
         DISPERSION MODELS: SCREENING, REFINED AND ADVANCED ..................................................................................2
         DISPERSION MODEL TYPES................................................................................................................................3
         MODEL APPLICATION: LEVELS OF ASSESSMENT .....................................................................................................3
                  Level 1 (Screening) Assessment ...................................................................................................3
                  Level 2 (Detailed) Assessment ......................................................................................................4
                  Level 3 (Comprehensive) Assessment ..........................................................................................5
         THE 11 STEPS TO GOOD MODELLING PRACTICE ....................................................................................................5

    APPROVED MODELS AND THEIR APPLICATION ............................................................................................... 7
         LEVEL 1 (SCREENING) MODELS ..........................................................................................................................7
                   AERSCREEN ...................................................................................................................................7
         LEVEL 2 AND 3 (REFINED) MODELS ....................................................................................................................8
                   AERMOD .......................................................................................................................................8
                   CALPUFF .......................................................................................................................................9
         CHOOSING A REFINED MODEL .........................................................................................................................10
                   Alternate Models........................................................................................................................11
                   Modifications to Models ............................................................................................................11

    DISPERSION MODEL INPUT – SOURCE PARAMETERS ................................................................................... 12
          SOURCE TYPES: POINT, AREA, VOLUME, AND LINE ..............................................................................................12
          MULTIPLE SOURCES, GROUPING (AERSCREEN) ................................................................................................12
          SOURCE EMISSION RATES ...............................................................................................................................13
                   Approved/Proposed Emissions Limits ........................................................................................13
                   Continuous Stack Monitoring .....................................................................................................14
                   Equipment Manufacturer Emission Specifications ....................................................................14
                   Published Emission Factors ........................................................................................................14
                   Modelled Emissions....................................................................................................................14
                   Stack Sampling Survey Data .......................................................................................................15
          EMISSION RATE VARIABILITY ...........................................................................................................................15
                   Emissions Under Normal Operations .........................................................................................15
                   Emissions Under Reduced Operations .......................................................................................15
                   Emissions Under Abnormal Operations .....................................................................................16
                   Emissions from Existing Sources ................................................................................................16
          APPROPRIATE TIME AVERAGES ........................................................................................................................16
          PM2.5 AND PM10 STACK EMISSIONS ................................................................................................................17
          PM EMISSIONS AND SIZE FRACTIONS FOR PARTICLE DEPOSITION ESTIMATES ............................................................17
          FUGITIVE SOURCES ........................................................................................................................................18
          EMISSIONS FROM OTHER SOURCES ...................................................................................................................19

    DISPERSION MODEL INPUT - GEOPHYSICAL DATA ........................................................................................ 20
          TERRAIN DATA..............................................................................................................................................20
          LAND USE DATA ...........................................................................................................................................21
          PROCESSING LAND USE DATA FOR AERSCREEN AND AERMOD (AERMET) SURFACE PARAMETERS ..........................22
                   Surface Roughness Height ..........................................................................................................24
                   Albedo ........................................................................................................................................24
                   Bowen Ratio ...............................................................................................................................25
          PROCESSING LAND USE DATA FOR CALPUFF (CALMET) SURFACE PARAMETERS ...................................................26



Version: July 2026                                                                                                                                         Page i
                              British Columbia Air Quality Dispersion Modelling Guideline


                           Processing GeoBC BTM Land Use Data ......................................................................................27
                           Processing GeoGratis Canadian Land Cover, Circa 2000 (Vector) Data .....................................33
                           Processing Manually Assigned Land Use Types..........................................................................35

    DISPERSION MODEL INPUT - METEOROLOGICAL DATA ................................................................................ 36
          ON-SITE AND EXISTING ALTERNATE SOURCES OF METEOROLOGICAL DATA...............................................................36
          DATA SOURCES: SURFACE DATA ......................................................................................................................37
                   Environment and Climate Change Canada (Meteorological Service of Canada: MSC) Data ......37
                   BC Ministry of Environment and Parks and Metro Vancouver Data ..........................................38
                   Other Sources of Surface (Land and Marine) Meteorological Data ...........................................39
          DATA SOURCES: UPPER AIR DATA ....................................................................................................................40
          LENGTH OF RECORD.......................................................................................................................................41
          MISSING DATA .............................................................................................................................................41
          LIGHT AND NO WIND CONDITIONS ...................................................................................................................42
          NWP MODEL OUTPUT ..................................................................................................................................43
          METEOROLOGICAL DATA FOR AERMET (AERMOD’S PRE-PROCESSOR) ................................................................43
                   Minimum Meteorological Data Requirements ..........................................................................43
                   Calms Treatment ........................................................................................................................44
                   Meteorological Data Acquisition and Preparation .....................................................................45
                   5.8.3.1 Reformatting Hourly Surface Data into AERMET Ready Files ......................................45
                   5.8.3.2 Wind Direction Treatment for MSC Hourly Airport Data .............................................45
                   5.8.3.3 Upper Air Data .............................................................................................................45
          METEOROLOGICAL DATA FOR CALMET (CALPUFF’S PRE-PROCESSOR) .................................................................46
                   Surface Meteorological Data Preparation for CALMET ..............................................................46
                   Upper Air Data Preparation ........................................................................................................48
                   Precipitation Code and Hourly Precipitation for Long-Term, Wet Deposition Calculations ......48

    METEOROLOGICAL MODELLING ................................................................................................................... 50
         NWP MODEL OUTPUT FOR DISPERSION MODELLING ..........................................................................................50
         BC WRF DATA AVAILABILITY ..........................................................................................................................53
         MESOSCALE MODEL INTERFACE PROGRAM (MMIF): PROCESSING NWP OUTPUT FOR DISPERSION MODELS ................53
         CALMET METEOROLOGICAL MODELLING .........................................................................................................54
                CALMET Modelling Modes .........................................................................................................54
                6.4.1.1 No Observation (No-Obs) .............................................................................................54
                6.4.1.2 Observation Only (Obs-Only) .......................................................................................55
                6.4.1.3 Hybrid...........................................................................................................................55
                Domain and Grid Resolution ......................................................................................................56
                Guidance on Key CALMET Model Options .................................................................................56

    DISPERSION MODELLING ............................................................................................................................. 64
          SIZE OF DOMAIN ...........................................................................................................................................64
          RECEPTOR SPACING .......................................................................................................................................64
          PLANT BOUNDARY.........................................................................................................................................65
          SENSITIVE RECEPTORS ....................................................................................................................................65
          FLAG POLE RECEPTORS...................................................................................................................................65
          BUILDINGS ...................................................................................................................................................65
          AERMOD: REGULATORY DEFAULT SETTINGS (CORE MODEL OPTIONS) .................................................................66
          CALPUFF: INPUT GROUP 2 AND 12 SWITCH SETTINGS .......................................................................................67

    POST PROCESSING ....................................................................................................................................... 70
          ADDING BASELINE AIR QUALITY CONCENTRATIONS..............................................................................................70
                  Site Specific Monitoring Data .....................................................................................................70
                  Monitoring Data from a Different Location ...............................................................................70



Version: July 2026                                                                                                                                           Page ii
                               British Columbia Air Quality Dispersion Modelling Guideline


                        Establish a Monitoring Program .................................................................................................71
                        Selecting a Baseline Level Using Monitoring Data .....................................................................71
                        Modelling Baseline (Contributing Sources) ................................................................................72
                        Future Baseline ...........................................................................................................................72
               NO TO NO2 CONVERSION ..............................................................................................................................73
                        100% Conversion ........................................................................................................................73
                        Ambient Ratio Method ...............................................................................................................73
                        Ozone Limiting Method ..............................................................................................................74
                        Plume Volume Molar Ratio Method (AERSCREEN and AERMOD Only) .....................................75
                        RIVAD/ARM3 and RIVAD/ISORROPIA Chemical Transformation ...............................................75
               DISPERSION MODEL OUTPUT ..........................................................................................................................75
                        Level 1 Assessments ...................................................................................................................76
                        Level 2 or 3 Assessments ............................................................................................................76
                        Computer Files ...........................................................................................................................77

      QUALITY ASSURANCE AND QUALITY CONTROL ............................................................................................ 78
            CALMET/CALPUFF QA/QC PROCESS............................................................................................................78
                    CALMET/CALPUFF QA Files ........................................................................................................78
                    CALMET Input Data ....................................................................................................................78
                    CALMET Output Data .................................................................................................................79
            MODEL PERFORMANCE AND UNCERTAINTY ........................................................................................................79

               SPECIAL TOPICS ................................................................................................................................. 82
               FLARING ......................................................................................................................................................82
                         Flare Source Parameters ............................................................................................................82
                         Intermittent Emission: Abnormal (Unscheduled) Flaring ...........................................................83
                         Intermittent, Transient Releases: Scheduled Pipeline Blow-Down Flaring ................................84
                         Well Test Flaring .........................................................................................................................85
                         10.1.4.1 Level 1 Assessment ......................................................................................................85
                         10.1.4.2 Level 2 Assessment ......................................................................................................85
                         10.1.4.3 Human Health: Assessment, Mitigation and Monitoring ............................................85
                         10.1.4.4 Vegetation: Assessment, Mitigation and Monitoring ..................................................86
               STAGNATION CONDITIONS ..............................................................................................................................87
               SHORE / COASTAL EFFECTS .............................................................................................................................88
               HORIZONTALLY ORIENTED STACKS AND STACKS WITH RAIN CAPS ...........................................................................88
               LIQUID STORAGE TANKS .................................................................................................................................89
               PLUME CONDENSATION (FOGGING) AND ICING ...................................................................................................89
               WET AND DRY SULPHUR AND NITROGEN DEPOSITION ..........................................................................................89
               POINT SOURCE PLUME: SECONDARY PM2.5 AND O3 ............................................................................................90

               MODEL RESOURCES .......................................................................................................................... 92
               SUMMARY OF BC MINISTRY-RECOMMENDED ACCEPTANCE ITEMS .........................................................................92

               REFERENCES ...................................................................................................................................... 93

APPENDIX A: DISPERSION MODELLING PLAN ....................................................................................................... 95




Version: July 2026                                                                                                                                               Page iii
                     British Columbia Air Quality Dispersion Modelling Guideline



ACKNOWLEDGEMENTS
The Guideline is the result of contributions from modelling and atmospheric-science experts who are experienced
in the application of dispersion models for regulatory processes in British Columbia (BC). The expert community
includes modelling practitioners from consultants (Arcadis, Levelton, RWDI Air, and Stantec) and regulatory
agencies (BC Ministry of Environment and Parks, Metro Vancouver).




Version: July 2026                                                                                      Page iv
                       British Columbia Air Quality Dispersion Modelling Guideline



PREFACE – A GUIDE TO THE GUIDE
The document provides key guidance on:

          model selection,
          application of models for regulatory purposes in BC, and
          best modelling practices.

The document is written for modelling practitioners who have a working knowledge of air quality and dispersion
models.

The complete package is available from the Ministry website1




SPECIFIC GUIDANCE
In addition to this document, the Ministry has published the following specific guidance:

          GUIDANCE FOR NO2 DISPERSION MODELLING IN BRITISH COLUMBIA.
          LOWER FRASER VALLEY PHOTOCHEMICAL MODELLING GUIDELINE.
          GUIDANCE FOR DISPERSION MODELLING OF MINING-RELATED FUGITIVE PARTICULATE MATTER
           EMISSIONS IN BRITISH COLUMBIA.




1
    https://www2.gov.bc.ca/gov/content?id=5D4A6114DAA14400860111A33E189B10


Version: July 2026                                                                                       Page v
                       British Columbia Air Quality Dispersion Modelling Guideline



     INTRODUCTION

        PURPOSE OF GUIDELINES

The BC Ministry of Environment and Parks (the Ministry) has produced the Guideline to help ensure that dispersion
modelling studies developed for the Ministry are appropriate for the needs of the application, correctly and
consistently applied and used to reliably inform air quality management decisions.

This Guideline applies to modelling submitted to the Ministry as well as Metro Vancouver (MV). Please be sure to
complete a modelling plan before commencing modelling work. The modelling plan for BC is located in Appendix A:
Dispersion Modelling Plan. Within MV’s boundaries, please use the Metro Vancouver Modelling Plan2.

Any deviation from the core modelling methodology should be presented to the Ministry prior to submission of
the modelling report.

This Guideline:

        Is not for regional scale, urban photochemical modelling although many topics in the Guideline are
         relevant for these purposes.

        Recognizes the importance of professional judgment and allows departures from or additions to Guideline
         practice as long as variations are detailed in a modelling plan.

        Will be updated as modelling practice changes (as driven by advances in science and user experience with
         the Guideline).

Please also note:

        Using this Guideline for air quality assessments is ministry policy, however it is not legally binding.

        Following the Guideline and/or obtaining written approval of a dispersion model plan may be mandatory
         if referenced in permit or other authorization under the Environment Management Act (EMA) or in an
         information requirements table when applying for an authorization to discharge waste under EMA.

        THE APPLICATION OF DISPERSION MODELS: BC’S UNIQUE SCIENTIFIC AND
        REGULATORY CONTEXT

The selection of a specific model and its application are driven primarily by two needs: scientific and regulatory.

From a scientific perspective, if the source and geophysical situation is simple (e.g., single source and flat terrain), a
simple model can be applied. However, BC is characterized by “complex” terrain (rolling hills, valleys, bowls,
coastal areas, and rugged mountains) and varied surfaces (water, open plains, forests, snow/ice, and urban zones).
These geophysical characteristics result in “complex flow” (curved trajectories, non-steady wind speed and
direction, stagnation, enhanced turbulence, and internal boundary layers) and present many challenges for
dispersion models to simulate.



2
 https://metrovancouver.org/services/environmental-regulation-enforcement/air-quality-regulatory-
program/dispersion-modelling-reviews


Version: July 2026                                                                                                 Page 1
                       British Columbia Air Quality Dispersion Modelling Guideline


From a regulatory perspective, a simple model may be applied if the source is small and the decision-maker is only
interested in knowing whether the air quality objectives/standards will be exceeded (independent of where or
when it occurs). On the other hand, sometimes the spatial and temporal distributions of the predicted
concentrations are of interest in particular if the source is large and near an environmentally sensitive or
populated area. This requires a more comprehensive model that produces spatial distributions of concentrations
or the location, time and frequency of exceedances of ambient air quality criteria.

        DISPERSION MODELS: SCREENING, REFINED AND ADVANCED

There are a variety of dispersion models that range from very simple to very complex that can be categorized into
three levels: Screening, Refined and Advanced.

Screening models are a simple way to estimate a “worst case” concentration that may possibly occur for a given
emission. They calculate hourly concentrations for a wide spectrum of different combinations of meteorological
conditions, and from this output the highest (worst-case) concentration is selected as the “worst case”. Due in part
to the use of a matrix of meteorological conditions (and not actual site-specific meteorology), they may either
underestimate or overestimate the “worst case”. Despite their limitations, they are a useful tool that provides a
quick means to flag the potential for unacceptable air quality.

Refined models include a more detailed treatment of the atmospheric processes, require more detailed input data,
and provide time and space concentration (or deposition) distributions. Refined models include more realism (physics
and meteorological input) in order to make predictions that are specific to the site and closer to reality than screening
model predictions.

Model input consists of geophysical data such as terrain and surface roughness, user-defined receptors, and a
sequential, hourly time series of meteorological data that are representative of the conditions at the location of the
source. In addition, hourly changes in source emissions can be input. This level of detail produces a sequential hour-by-
hour contaminant concentration (or deposition) simulation for an array of receptor locations. Output can include both
short-term (one hour) and long-term (multiple-hour, daily, seasonal, and annual) average concentrations at every
receptor location. Thus, it is possible to obtain the maximum concentration as well as details on the space and time
variation of the concentrations — something that may be of importance to decision makers.

Advanced models include comprehensive treatments of meteorology, emissions and chemistry and require
considerable expertise and computer resources to initiate, execute, and interpret the results. An example would be a
regional scale photochemistry models such as the Community Multi-Scale Model (CMAQ) (Byun & Schere, 2006),
where simulations of secondary contaminants over large domains are conducted. Their primary use is to examine the
effectiveness of large-scale source management decisions on air quality over a regional area (e.g., changes to vehicle
fuel standards, planning for new subdivisions).

Specific guidance on the use of Advanced models for the Lower Fraser Valley can be found in the Lower Fraser Valley
Photochemical Modelling Guideline. The following are considerations as to when advanced models could be applied:

        A regional scale (on the order of 100 km) assessment where the formation of ozone and/or secondary
         particulate matter is of critical importance.
        Comprehensive emissions inventories are available that include the sources, spatial distribution, and chemical
         species required as inputs for the chemistry module of the model.
        Ambient levels of chemical species are available as required by the chemistry module.



Version: July 2026                                                                                                Page 2
                       British Columbia Air Quality Dispersion Modelling Guideline



        Supporting meteorology (i.e., mesoscale meteorological model output) for specific periods that are deemed
         to be representative episode periods.
        An assessment of model performance.
        Expertise and resources required to set up, run and interpret the output.

        DISPERSION MODEL TYPES

The Screening and Refined categories of model complexity include two types of dispersion models – both are
based on the assumption that the horizontal and vertical distribution of the contaminant has a Gaussian
distribution.

Straight Line Gaussian Plume models calculate contaminant concentrations for each hour assuming
meteorological conditions are uniform over the horizontal and vertical space that is included in modelling domain.
The vertical and horizontal distribution of the pollutants is assumed to be Gaussian (“bell shaped”). Because of the
steady-state, straight-line nature of these models, they do not explicitly account for curved plume trajectories and
variable wind conditions that occur in complex flow situations. In addition, these models cannot handle low wind
speeds - a frequent occurrence in the bowls and deep valleys of BC especially during the winter. These types of
models can be in the Screening or Refined category of model complexity.

Curved Trajectory Gaussian Puff models treat the emissions as a series of puffs in time and space varying
meteorological conditions. Although these models require more computing resources as they track puffs that
represent discrete amounts of contaminants over time, they have the advantage of allowing meteorological
conditions to vary in space (horizontally and vertically). In addition, they can address the accumulation of
contaminants during calm conditions, the curved paths of plumes, and the effects of causality (where the previous
position and conditions of the plume is accounted for in determining the current plume position). This type of
model is in the Refined category of model complexity.

        MODEL APPLICATION: LEVELS OF ASSESSMENT

There are a number of dispersion models available and it is important to choose and apply the model best suited
to the situation. There are three general levels of assessment:

        Level 1 (Screening) Assessment

        Level 2 (Detailed) Assessment

        Level 3 (Comprehensive) Assessment


         LEVEL 1 (SCREENING) ASSESSMENT
A Level 1 Assessment is appropriate when decisions can be made based only on an estimate of the possible worst-
case air quality, independent of where or when they occur. Screening Assessments are appropriate for situations
such as:

        “Go, no-go” evaluations (a critical acceptance/rejection criteria is exceeded or not exceeded: typically for
         low-risk sources).

        Permit/approval decisions for low-risk sources.




Version: July 2026                                                                                            Page 3
                       British Columbia Air Quality Dispersion Modelling Guideline



        Preliminary identification of air quality issues associated with proposed new sources or modifications to
         existing sources.

        Planning purposes (internal resources required to conduct assessment, need to consider other studies to
         support the decision-making process, need to contact other agencies).

        Identification of the need for more detailed modelling using Level 2 or 3 Assessment approaches (if
         exceedances of short-term objectives are predicted).

        Confirmation of refined model results that appear unusually high or low.


         LEVEL 2 (DETAILED) ASSESSMENT

A Level 2 Assessment provides a more realistic and detailed determination of air quality dispersion than what is
provided in a Level 1 Assessment. This level requires a Refined model that uses a time series of hourly meteorological
data (over a period of at least one year) and the geophysical conditions representative of the site. A Level 2 Assessment
is appropriate when:

        A Screening (Level 1) Assessment indicates the potential for an exceedance of ambient objectives.

        There is a need to produce a maximum concentration for different time averages and distributions of the
         concentrations and/or depositions in time and space that reflect the actual meteorological conditions.

        The contaminant can be reasonably modelled by a straight-line, steady-state, Gaussian plume model with
         no chemical transformation. Although more complicated processes may be occurring (i.e., curved plume
         trajectory), a more complicated model that explicitly treats these processes is not necessary depending
         on the purposes of the modelling and the zone of interest. For example, if the area of interest is within
         100 m, then curvilinear trajectories and chemical transformations are likely not critical.

        The emissions are from small sources (e.g., a small compressor station stacks) where the greatest
         concentrations are in the order of 100 m downwind.

        The source is considered to be low risk.

        The purpose is for a standard/generic permit or amendment process (such as an emissions reduction).

        To define conditions under which well test flaring can occur.

        Supporting other air management related investigations such as:

              o   identify potential contributing sources;

              o   identify worst-case meteorological conditions;

              o   identify areas of air quality concern;

              o   analyze historical air quality trends; and

              o   design monitoring networks (locations, contaminants, sampling period, frequency).




Version: July 2026                                                                                              Page 4
                       British Columbia Air Quality Dispersion Modelling Guideline



         LEVEL 3 (COMPREHENSIVE) ASSESSMENT

Level 3 Assessments require Refined models and corresponding input data, resources and model operator expertise to
properly account for these factors. These models require detailed meteorological, geophysical and source input that
may include:

        One or more years of representative meteorological data (e.g., wind speed, wind direction, temperature,
         turbulence, and mixing height) at a number of sites in the domain of interest.

        Detailed emission inventories for point, line, area and volume sources in an airshed that could vary in time
         and where there is a mix of urban, industrial and natural sources.

        The speciation, emissions and time variation of different contaminants.

In general, a Level 3 Assessment is recommended in situations where:

        A Level 1 or 2 Assessment indicates predicted exceedances of ambient objectives.

        The purpose of the assessment requires detailed time and space variation of the concentrations.

        It is important to account for multiple source types, chemical transformations and effects associated with
         complex topography such as causality, calms, curvilinear plume trajectories, spatial variations in turbulent
         mixing.

        A source is considered to be high risk.

More specifically, Level 3 Assessments are recommended in situations where there is a need to:

        Evaluate air quality consequences under a permitting or Environmental Assessment process for large
         industrial developments that have considerable social, economic and environmental consequences.

        Assess contaminants resulting from non-linear processes (e.g., deposition, ground-level ozone, particulate
         formation, visibility).

        Evaluate consequences of air quality management approaches that involve multi-source, multi-sector
         contributions from permitted and non-permitted sources in an airshed.

        Provide information to support environmental, human and economic effects studies.

        Examine specific receptors that may be sensitive or of special interest such as individual residences,
         sensitive ecosystem areas.

        Assess contaminants in meteorologically complex situations (“complex flow” such as mountain valley
         flows, reversals, sea breeze, and fumigation).

        Assist in understanding of the underlying source and meteorological causes of episodes.

         THE 11 STEPS TO GOOD MODELLING PRACTICE

The following are recommended steps to follow for every modelling application. The steps help determine the
modelling approach, the assessment level and help ensure the model output addresses the needs of the regulatory
agencies.




Version: July 2026                                                                                           Page 5
                     British Columbia Air Quality Dispersion Modelling Guideline



                 11 Steps to Air Quality Dispersion Modelling

    Step 1: Set the Context              • Define objectives and scope of the study.



  Step 2: Characterize Sources           • Gather information on sources, emissions and
        / Contaminants                    contaminants.


     Step 3: Characterize
                                         • Define the modelling domain and review geophysical
  Physical and Meteorological             characteristics and atmospheric behaviour of the area.
             Setting

        Step 4: Determine                • Determine Assessment Level required to meet the
        Assessment Level                  objectives.



                                         • Select a dispersion model by reviewing the technical
     Step 5: Select a Model               capabilities of each model and its recommended use.



    Step 6: Determine Model              • Define the meteorological and geophysical data needs
             Inputs                       based on the Assessment Level and model selected.



   Step 7: Determine Baseline            • Consider the baseline concentrations of air contaminants
           Air Quality                    if the intent is to assess the cumulative effects.



   Step 8: Submit a Dispersion           • A Dispersion Modelling Plan is required. The plan is a
                                          tool to facilitate communication between the different
         Modelling Plan                   parties in order to avoild misunderstandings and delays.


   Step 9: Prepare Input Files,          • Format meteorological and geophysical data input files
        Execute Models                    and excecute the model.


   Step 10: Quality Assurance            • Exercise a QA/QC procedure to confirm the accuracy of
      and Quality Control                 the inputs, meteorological data and behaviour of the
            (QA/QC)                       model.

                                         • In order for the Ministry to conduct a thorough review of
   Step 11: Model Output and              the air quality assessment, input and output files and
         Documentation                    corresponding documentation may be required.




Version: July 2026                                                                                   Page 6
                       British Columbia Air Quality Dispersion Modelling Guideline



     APPROVED MODELS AND THEIR APPLICATION

Table 2.1 provides dispersion models recommended for use in BC and the appropriate assessment level. A brief
description of each model and the specific situations under which they can be applied are provided in the following
sections.

Table 2.1         Assessment Level and Corresponding Guideline Dispersion Model


            Air Quality Assessment                       Recommended Guideline Model


                                                                Screening Models:
              Level 1: Screening
                                                                   AERSCREEN

                                                                 Refined Models:
               Level 2: Detailed                                     AERMOD
                                                             CALPUFF (No-Obs mode)
                                                            CALPUFF (Obs-Only mode)
                                                                 Refined Models:
                                                                     AERMOD
            Level 3: Comprehensive
                                                              CALPUFF (Hybrid mode)
                                                            CALPUFF (No-Obs mode)*
*depends on several considerations (see Section 6.4.1).

        LEVEL 1 (SCREENING) MODELS


         AERSCREEN

AERSCREEN is a screening version of the Refined model AERMOD. AERSCREEN is a shell that runs AERMOD with an
input matrix of meteorological conditions that represent a wide range of possible conditions. This matrix is
prepared by the MAKEMET program through the use of interactive command-prompts that specify surface
characteristics, ambient temperatures, minimum wind speed, and the anemometer height. The surface
characteristics can be the default values, user specified, or produced through the land use pre-processor program,
AERSURFACE (see Section 5.8: Meteorological Data for AERMET (AERMOD’s Pre-Processor)).

AERSCREEN produces the worst-case 3-h, 8-h, 24-h and annual average concentrations from a single source. Note
that if there are buildings and/or there is terrain to be included, then model initiation and execution can require
the use of additional pre-processors BPPIPPRM (prepares building parameters) and AERMAP (prepares terrain
data).

The use of AERSURFACE and AERMAP for BC applications requires additional processing steps as the data formats
for BC land use and terrain may need further manipulation before applying the preprocessors.

Recommended use:

        maximum 1-h, 3-h, 8-h, 24-h and annual average concentrations

        single point, capped stack, horizontal stack, flare, area (circular or rectangular), volume sources



Version: July 2026                                                                                             Page 7
                         British Columbia Air Quality Dispersion Modelling Guideline



          building wake effects on point, capped stack, horizontal stack, and flare sources

          flat and elevated terrain

          urban and rural areas

          transport distances of less than 50 km (depends on terrain)


Approved Version: The latest version available via the U.S. EPA Screening Models3 should be applied.

Although the model can only handle a single source, multiple point sources can be accounted for by following the
recommendations in Section 3.2: Multiple Sources, Grouping (AERSCREEN).

AERSCREEN is the preferred screening level model for use in BC and MV.

          LEVEL 2 AND 3 (REFINED) MODELS


           AERMOD

AERMOD is a regulatory straight-line, steady-state plume modelling system that consists of three components:
AERMOD (calculates the concentrations), AERMET (prepares the meteorological input) and AERMAP (prepares the
terrain input).

The meteorological preprocessor AERMET is required to process available meteorological data into a format
suitable for use by AERMOD. AERMET is designed to operate on certain types of data: U.S. National Weather
Service, Federal Aviation Administration or other hourly surface observations, upper air soundings, and data
collected from on-site measurement program such as from an instrumented tower.

When applying AERMET, the appropriate surface characteristics (surface roughness length, albedo and Bowen
ratio) must be specified. This can be done manually or through use of the preprocessor AERSURFACE which reads
U.S. Geological Survey (USGS) Land Cover datasets and uses look up tables for surface characteristics that vary by
land cover type and season to produce AERMET-ready surface data. Since AERMOD results are sensitive to these
surface characteristics, AERSURFACE provides a level of consistency and objectivity to this step in the data
preparation.

The preprocessor AERMAP processes terrain data to produce terrain base elevations for receptors and sources and
other AERMOD required information (i.e., hill height scale for each receptor). If there are buildings that could
cause downwash, then AERMOD requires building related information produced by the processor BPIPPRM.

The application of AERMOD for locations in BC requires additional processing steps to those noted above. The data
formats for BC terrain, land cover, and meteorological data are different than the U.S. formats, so they may need
further manipulation before applying the preprocessors. These input data issues will be discussed in Section 4.3:
Processing Land Use Data for AERSCREEN and AERMOD (AERMET) Surface Parameters and Section 5.8:
Meteorological Data for AERMET (AERMOD’s Pre-Processor).




3
    http://www.epa.gov/scram/air-quality-dispersion-modeling-screening-models


Version: July 2026                                                                                          Page 8
                        British Columbia Air Quality Dispersion Modelling Guideline


Recommended use:

         sources in an industrial complex (single or multiple point, area, line, volume sources)

         situations where a straight-line, steady state model applies (non-complex flow)

         single or multiple buildings with or without building downwash

         gas and particle depositions

         constant or time-varying emissions

         rural or urban areas

         transport distances less than 50 km (depends on terrain)

         concentration estimates for all terrain locations, except in areas located on the lee-side of the
          topographic features

Approved Version: The latest version of AERMOD publicly available via the U.S. EPA Preferred/Recommended
Models4 website should be applied.


          CALPUFF

CALPUFF is a curved trajectory, Gaussian puff model that can account for time- and space-varying meteorological
conditions, different source configurations and contaminants, and chemical transformations. The CALPUFF
modelling system is comprised of three components: CALMET (meteorological model), CALPUFF (calculates
concentration and deposition output), and CALPOST (analysis and display of output). The meteorological fields
used by CALPUFF are produced by CALMET — a meteorological model that includes a diagnostic wind field model.
CALMET inputs include surface and upper air meteorological data as well as the option to use the output produced
by gridded numerical weather prediction models. CALMET can blend those input and apply treatments of slope
flows, valley flows, terrain blocking effects, kinematic terrain effects (i.e., speed up over hills), lake and sea breeze
circulations, and a procedure to ensure mass is conserved in the domain.

CALPUFF generates a variety of outputs that include concentrations, wet and dry depositions, and visibility
parameters (extinction coefficients). It describes a continuous plume as a series of puffs, which in turn allows a
better description of how a plume can curve, deform (e.g., passing through a narrowing in the valley), and even
split apart in complex wind areas. Furthermore, unlike other straight-line Gaussian plume models, it can be applied
under calm conditions.

Recommended use:

         complex flow: non-steady-state meteorological conditions (calms, time and space variability in wind and
          turbulence fields) such as found in complex terrain and coastal situations

         local scale (<10 km), regional scale (10-50 km) to long-range transport and dispersion (>50 to 200 km)

         multiple sources, source types (point, area, volume) and building(s)

         gaseous and particulate deposition



4
    http://www.epa.gov/scram/air-quality-dispersion-modeling-preferred-and-recommended-models


Version: July 2026                                                                                               Page 9
                        British Columbia Air Quality Dispersion Modelling Guideline



          wet and dry Sulphur and Nitrogen deposition

          PM2.5 secondary formation

          visibility assessments for regional (>10 km) and long-range transport distances (> 50 km to 200 km)

          constant or time varying source conditions of gaseous and particulate contaminants

          fogging, icing and odour effects

Approved Version: The Ministry strongly recommends the use of the latest version of the CALPUFF system (i.e.,
CALMET, CALPUFF, CALPOST and associated utilities) that is publicly available via Lakes Environmental5.

Earlier versions of CALPUFF or any of its associated components can be used provided that the modifications in the
later versions are not relevant to the application.

CALPUFF can be run in different “modes”, which reflect the source of input meteorology used for CALMET
processing. The mode selected depends on several factors which include the Assessment Level, the availability of
meteorological input, whether the flow is complex, and whether output from a Numerical Weather Prediction
(NWP) model output is available. See Section 6.4.1: CALMET Modelling Modes for more information on selecting a
CALPUFF mode.

Depending on the mode, CALPUFF can require professional judgment on selecting options and adjustable
parameters, as well as establishing the right combination of these parameters requires expert understanding of
large-scale and small-scale meteorological effects, especially when applied in complex flow situations.

          CHOOSING A REFINED MODEL

Determining whether to use AERMOD or CALPUFF depends on the following considerations based on the CALPUFF
FAQ’s6:

          The degree in which the flow is “complex” (wind and turbulence patterns change over short distance,
           prolonged stagnations).

          The area of interest (i.e., transport distance) — shorter transport distances can be less influenced by
           complex flow and can be approximated by a straight line trajectory.

          The objectives of the application — even in a very complex flow situation, if only maximum
           concentrations are desired, then a straight line model may be appropriate, however if the concentrations
           at receptors around the bend of a valley is required for example, then CALPUFF is the better choice if
           representative meteorological data are available.

In BC, complex flow is inherent for most applications, and information needs for Level 2 and 3 Assessments require
a model that provides spatial variations of the concentration/depositions fields (not just maximum concentrations).
As such, CALPUFF is technically superior for these situations and it is the only scientifically-based regulatory tool
for complex flow.




5
    https://www.calpuff.org/
6
    https://calpuff.org/calpuff/FAQ-answers.htm


Version: July 2026                                                                                            Page 10
                         British Columbia Air Quality Dispersion Modelling Guideline



           ALTERNATE MODELS

Although the models recommended here can handle the vast majority of applications, there may be circumstances
where an alternate model would be better suited. The conditions under which an alternate model would be considered
are:

          The recommended models are not appropriate due to technical limitations for the application.

          The alternate model performs better than the Guideline model based on peer reviewed evidence where
           the proposed alternate model has been applied in a similar situation using performance indicators
           relevant for this application (maximum hourly concentrations comparisons, exceedance threshold
           comparisons, spatial pattern comparisons, etc.).

          The alternate model is publicly available.


           MODIFICATIONS TO MODELS

Modifications to the model program can result in unexpected results due to the complexity of the inter-related
model components. If a model is modified, include the following information in the Dispersion Modelling Plan:

          The justification for such a modification.

          The modification is documented and testing results provided.

          For CALPUFF, proof that modified versions (with documentation) have been submitted to the program
           developer as per the CALPUFF End-User License Agreement7.




7
    https://www.calpuff.org/calpuff/calpuff_eula.htm


Version: July 2026                                                                                        Page 11
                       British Columbia Air Quality Dispersion Modelling Guideline



     DISPERSION MODEL INPUT – SOURCE PARAMETERS

The following provides guidance on determining source types and emission parameters as well as approaches to treat
time-varying emissions for dispersion modelling applications. Any uncertainties and errors in these inputs will be
reflected in the model results, so this step is critical to the quality of the dispersion modelling effort.

        SOURCE TYPES: POINT, AREA, VOLUME , AND LINE

Emission sources can be categorized into four types based on geometry: point, area, volume, and line sources. All
of the dispersion models listed in Section 2 can be used for point, area, and volume source types; however, with
AERSCREEN, there are limitations for multiple source situations.

A point source is a stationary, specific point of origin where contaminants are emitted into the atmosphere (such
as a stack). A flare is considered to be a point source, but requires special treatment (see Section 10.1). Stacks with
rain caps and those horizontally oriented are special circumstances that AERSCREEN, AERMOD, and CALPUFF can
handle explicitly. Section 10: Special Topics for guidance on these types of sources.

An area source is an emission into the atmosphere that is distributed over a stationary spatial area such as settling
ponds and even urban regions that include multiple point sources (which combined together act as an area source).
See Section 3.8: Fugitive Sources for guidance on fugitive emissions such as open pit mines.

A volume source is an emission to the atmosphere that has an initial width and depth at a stationary release point
such as the dust emissions from an aggregate storage pile.

A line source is an emission to the atmosphere that is distributed over a line such as conveyor belts, roadways, and
rail lines. Some models may not treat line sources explicitly – in this case sources of this type can be handled as
area sources (long, thin rectangles) or as a string of volume sources.

        MULTIPLE SOURCES, GROUPING (AERSCREEN)

The emissions from a facility can come from a number of sources of different types, locations, and characteristics.
Deciding which sources should be included is largely dictated by the magnitude of the contaminant emission rate
(e.g., a large emission rate has a large area of influence). A dispersion model may need to include sources outside
the facility if they contribute to the contaminant load in the area (Section 8.1: Adding Baseline Air Quality
Concentrations). Decisions regarding the inclusion of nearby sources in the modelling effort may involve an
iterative process, where subsequent model runs indicate which sources are important.

If the plumes from adjacent stacks are close enough to merge, the plume rise can be enhanced. None of the
recommended models account explicitly for the merged plume rise situation although the non-treatment in
models is generally a conservative assumption, given that the modeled plume rise in these situations will be lower
than in reality.

For Level 1 Assessments for multiple source situations, since AERSCREEN can only account for a single source, the
following is recommended:

        Estimate the concentration by direct addition of the predicted worst-case concentrations for each source.
         This very conservative estimate is consistent with a screening approach.




Version: July 2026                                                                                            Page 12
                        British Columbia Air Quality Dispersion Modelling Guideline



          A group of closely spaced stacks can be modelled as a single source where stacks located less than one-
           stack diameter apart and with the same release height, similar exit velocities, and flow rates can be
           treated as a single source. Follow Section 2.2 of EPA Screening Procedures for Estimating the Air Quality
           Impact of Stationary Sources8 to calculate a pseudo diameter for the single source based on the total
           volume flow rate and exit velocity of the stacks.

          If sources cannot be grouped, then consider the use of a Refined model.

For Level 2 and 3 Assessments, the AERMOD and CALPUFF are capable of handling multiple sources. However, as
mentioned earlier, there is no explicit treatment for plume rise in situations where the plume merges with others.

          SOURCE EMISSION RATES

The basis for the emission rates used as input depends on whether dispersion modelling is used to assess the air
quality consequences of a new or modified existing source, or whether it is used in a retrospective analysis –
where historical air quality from existing sources are of interest.

If the emitted contaminant is assumed not to chemically transform in the atmosphere, the downwind
concentrations are directly proportional to the emission rate of the contaminant (if all other source emission
characteristics such as exit temperature do not change). Thus it is critical to use an emission rate that is both
accurate and appropriate for the purpose of the modelling application.

Methods that can be used to establish emission rates are the following:

          approved/proposed emission limits

          continuous stack monitoring

          equipment manufacturer emission specifications

          published emission factors

          modelled emissions

          stack sampling survey (limited use)


           APPROVED/PROPOSED EMISSIONS LIMITS

Maximum emission limits for authorized discharges are often set in permits under the BC Environmental
Management Act9 and reflect emissions associated with the process and control technology. Emission criteria (in
the form of an in-stack concentration) in combination with volume-flow limits establish a contaminant mass
emission rate for particulate matter, sulphur oxides, nitrogen oxides, and other air contaminants of concern. The
averaging time associated with these approved emission rates may not be specified in the permit, so it is
important to establish whether the permit criteria concentration (and thus the emission rate) corresponds to a 1-h
averaging time. This is further discussed in Section 3.4: Emission Rate Variability.




8
    https://www.epa.gov/sites/default/files/2020-09/documents/epa-454r-92-019_ocr.pdf
9
    https://www.bclaws.gov.bc.ca/civix/document/id/complete/statreg/03053_00


Version: July 2026                                                                                            Page 13
                       British Columbia Air Quality Dispersion Modelling Guideline



          CONTINUOUS STACK MONITORING

Large stacks at industrial facilities that emit a significant amount of air contaminants are often equipped with
Continuous Emission Monitoring (CEM) systems that measure exit temperature, exit flow rate, and contaminant
mass emission rates. The data can be analyzed to develop appropriate maximum, average, and upper percentile
(e.g., 95th or 98th) operating conditions of the stack. Furthermore, the sequential time series of hourly average exit
velocity, exit temperature, and emission rate data can be used by many of the recommended models to assess the
effect of variations in emission parameters on predicted ambient air quality.


         EQUIPMENT MANUFACTURER EMISSION SPECIFICATIONS

For new facilities or in cases where measurements are not available, manufacturer specifications of contaminant
emission rates can be used. Emission rates may be provided for different load capacities and operating conditions.
Confidence in manufacturer’s ratings is reduced if the equipment:

        is old;

        has been retrofitted or modified;

        is not operating under optimum conditions; or

        is operated in substantially different conditions than the emission test (i.e., climatic conditions, altitude,
         process changes).

In such cases, the original manufacturer’s specifications can be modified if there is evidence
(i.e., stack samples, data from other studies) that supports a different value.


         PUBLISHED EMISSION FACTORS

When no appropriate measured emission rates are available, published emission factors can be used (see the U.S.
EPA’s AP-42 emission factors10). These emission factors provide the mass of contaminants discharged per mass of
fuel consumed, product processed or activity rate and are rated (i.e., from A to E) to reflect their uncertainty. This
rating and the applicability of the AP-42 emission factors to conditions in BC must be considered before they are
used. For example, emission factors associated with dust from mineral crushing have a rating of E (poor).
Modelling for this type of source using these factors will be limited in its value as it is subject to large uncertainty.

In some situations there may be more recent information available for a specific source type that would be more
appropriate than the AP-42 emission factors. In this case, provide the reference and the rationale for this
alternative factor(s) in the Dispersion Modelling Plan.


         MODELLED EMISSIONS

There are various models available that estimate emissions using a calculation methodology that is specific to a
source type. For example, emission models for wastewater treatment plants, landfills, and other sources are




10
  https://www.epa.gov/air-emissions-factors-and-quantification/ap-42-compilation-air-emissions-factors-
stationary-sources


Version: July 2026                                                                                              Page 14
                        British Columbia Air Quality Dispersion Modelling Guideline


available from U.S. EPA’s Emission Factors Estimation Tools11. In addition, follow the link for information on U.S.
EPA emissions estimation methods for on-road and non-road mobile sources12.


           STACK SAMPLING SURVEY DATA

Stack sampling survey data provide a snapshot in time of the emissions and are normally conducted for permit
compliance reasons, rather than to characterize the emissions for modelling purposes. Unless “sufficient” survey
data (based on expert judgment) are available to determine the range of possible emissions from a source, they
should only be used in situations where the air quality at the time of the survey is of interest or when there are no
or little reliable data from other sources of information on emissions. Stack survey information may also be useful
to estimate emission rates under reduced operation/production rates (discussed in the following section) if there
are corresponding records of production rate.

         EMISSION RATE VARIABILITY

Most sources of air emissions will exhibit variability in the contaminant emission rates over short or long-term
periods. This variability can be due to changes in operating/production capacity and abnormal/upset conditions.
Although the recommended Refined models have the ability to handle time-varying emission rates, many model
applications are for proposed (i.e., yet to be built) sources. In these cases, a single emission value is assumed to
apply for the entire simulation period. The following provides guidance on selecting the appropriate emission rate
and examining the air quality consequences of this variability.


           EMISSIONS UNDER NORMAL OPERATIONS

Emission rates may vary when a process is not operated continuously 24 hours per day or when the process rate
(e.g., loading capacity, fuel consumption) varies with time. This variability means that the maximum short-term
emission rate experienced over a period of a year can be much greater than the annual average emission rate.

If modelling is conducted to assess air quality that results from a future contaminant emission scenario, it is
desirable to use the maximum emissions under normal operating conditions to determine compliance with
ambient air quality objectives. Given that many ambient objectives are 1-h averages, this means using the
maximum expected 1-h emission rate for dispersion modelling. See Section 3.5: Appropriate Time Averages for
recommendations on longer time averages.


           EMISSIONS UNDER REDUCED OPERATIONS

Maximum predicted ambient concentrations may not necessarily occur under maximum normal operating conditions.
Although emission rates can be lower at a reduced capacity, there can be a corresponding reduction in plume rise due
to the lower gas effluent volumes and temperatures. Higher concentrations could result under the reduced operation
scenario than under maximum capacity emissions especially when there is nearby terrain, where a lower plume means
impingement on the surface closer to the stack. Furthermore, some contaminants, such as NOx, may have higher
emission rates when a combustion source is not operating at full load due to lower operating efficiencies. In such cases:



11
     https://www.epa.gov/air-emissions-factors-and-quantification/emissions-estimation-tools
12
     https://www.epa.gov/moves


Version: July 2026                                                                                             Page 15
                       British Columbia Air Quality Dispersion Modelling Guideline



        For elevated sources, examine the air quality consequences of the 25, 50, 75, and 100 % capacity scenarios
         (if such scenarios could occur) by using a Screening model (i.e. AERSCREEN) to ensure that the source
         conditions causing the highest concentration are identified.


         EMISSIONS UNDER ABNORMAL OPERATION S

Ambient concentrations should also be predicted for abnormal operating conditions since it is these situations that can
result in the poorest air quality. “Abnormal operations” are known situations based on operational history, anticipated
scheduled maintenance of control equipment, by-pass operations for plant safety, and start-up/shut-down situations.

Abnormal emissions are anticipated, but it is not known whether they will occur and when the dispersion conditions
are good or poor. Although the probability of abnormal emissions occurring at the same time as poor dispersion
conditions may be very small, decision makers are often interested in the worst-case air quality scenario that could
potentially occur. In this case, the following is recommended:

        Establish the realistic emission scenarios associated with abnormal emissions and their anticipated
         frequency of occurrence (start-up, shut-down, maintenance, etc.).

        Apply a Screening model (i.e. AERSCREEN) to determine the potential worst-case air quality and the
         associated meteorology under these scenarios.

        Estimate the likelihood of the abnormal emission occurring at the same time as the worst-case
         meteorology.

Whether the abnormal emission scenario is an important factor in management decisions regarding the source
depends on an assessment of both the probability of the maximum concentration occurring and the air quality
consequences if it does occur.

Finally, “upset conditions” on the other hand are unanticipated emissions (i.e., accidents) that result from
unexpected equipment/process failure. Since the emissions under these scenarios are unknown, modelling the air
quality that results from these scenarios is not possible.


         EMISSIONS FROM EXISTING SOURCES

If the objective is to reconstruct a past air quality event, actual emissions that correspond to that event should be
used. An hourly emissions file can be input into the Refined models recommended here, and variations in the
emissions can be accounted for explicitly if the data are available (such as from a continuous emission monitor).
Alternatively, equipment manufacturer emission specifications can be used. If specific events are of interest, stack
sampling data can be used if it corresponds to the time of the event.

        APPROPRIATE TIME AVERAGES

Given that emission rates can vary over time, the maximum short-term emission rate experienced over a period of
a year can be much greater than the annual average emission rate. In order to determine compliance with short-
term (1-h) and longer-term average (8-h, 24-h, annual) ambient air quality objectives, the following is
recommended:

        For 1-h average concentrations use the maximum 1-h emission rate over the complete period of
         simulation.



Version: July 2026                                                                                           Page 16
                       British Columbia Air Quality Dispersion Modelling Guideline



        For longer term averages (8-h, 24-h, annual), the maximum emission rate that corresponds to the
         concentration averaging period can be used (e.g., annual average emission rate for annual average
         predictions) over the complete period of simulation.

        PM 2.5 AND PM 1 0 STACK EMISSIONS

Emissions of PM2.5 and PM10 consist of a filterable portion (directly emitted “primary” solid and liquid particles)
and can include a condensable fraction (vapours that condense into PM 2.5 when cooled to ambient temperatures).
This latter component can be significant as some sources, such as veneer dryers, may have a condensable fraction 5
– 10 times greater than the filterable portion. Other sources may have only a small condensable fraction, in which
case it can be ignored.

The information available on PM emissions may not be size specific and may not include the condensable fraction.
For example, if emission factors or manufacturer’s specifications are used, the distinction between sizes and/or
condensables may not be available. Given this situation, the following is recommended for estimating PM emission
rates:

        If information that distinguishes PM10 and PM2.5 is available, the corresponding emission rates should be
         used as model emissions input to determine PM 10 and PM2.5 concentrations.

        If the filterable and condensable fractions of the PM emissions are specified, then use the total (i.e.,
         filterable + condensable) as the emission rate for modelling.

        If there is no information available on whether the emission rates include the condensable portion, then
         use the rate as is.

        PM EMISSIONS AND SIZE FRACTIONS FOR PARTICLE DEPOSITION ESTIMATES

AERMOD and CALPUFF can model particulate deposition for a range of user specified particulate sizes, mass
fractions and the particle densities. If this is enabled, then a range of particle size categories is required along with
information such as corresponding mass fractions and the particle densities.

 For CALPUFF, using the pre-configured (default) PM2.5 and PM10 distribution curves can lead to concentrations of
PM2.5 greater than PM10 due to inconsistencies in the distribution within the PM2.5 range. In order to avoid this
situation, for INPUT GROUP 8 in the CALPUFF.INP file, use Table 3.1 for PM2.5 and PM10 deposition calculations.




Version: July 2026                                                                                              Page 17
                       British Columbia Air Quality Dispersion Modelling Guideline


Table 3.1          PM Granulometry Distributions for 6 PM Species

                                            Geometric           Geometric                Affiliation
              PM        Size Range
                                            Mean Dia.           Standard
            Species        (µm)
                                              (µm)              Deviation         PM2.5           PM10
              P1         0.5 - 0.75           0.625                 0.0              X                 X
              P2         0.75 - 1.0           0.875                 0.0              X                 X
              P3         1.0 - 1.25           1.125                 0.0              X                 X
              P4         1.25 - 2.5           1.875                 0.0              X                 X
              P5           2.5 - 6            4.25                  0.0                                X
              P6           6 - 10               8                   0.0                                X

When the Geometric Standard Deviation is set to 0.0, CALPUFF assigns the diameter to the entire bin (i.e., there is
no size distribution about the geometric mean diameter). In addition, CALPUFF uses the number of species size
categories as input by the user and the number of categories as specified by NINT is ignored. PM2.5 and PM10
concentrations can be calculated at the post processing stage (using CALSUM or POSTUITL) by adding the relevant
PM size categories. In other words:

         PM2.5 = P1 + P2 + P3 + P4

         PM10 = P1 + P2 + P3 + P4 + P5 + P6

For larger size ranges (such as for TSP calculations where particle sizes may be 40 µm), Table 2.6.1 from the
Newfoundland Guideline for Plume Dispersion Modelling 13 (Lawrence, 2012) can be used. When using this Table,
the particles size intervals must be specified.

CALPUFF assumes the density of each particle is 1 g/cm3. There is no option in the CALPUFF input control file to
change this. In some situations (for example with heavy metal deposition), this density may not apply, so refer to
Table 2.6.2 in the Newfoundland Guidelines14 for the “effective” Geometric mass mean diameters and Geometric
standard deviations that would be correspond to these particle densities.

       FUGITIVE SOURCES

Fugitive sources are difficult to characterize since their emissions may vary with wind speed and time of day (for
particle emissions) or process changes, and the control efficiency of mitigation measures applied to reduce
emissions can be only crudely estimated with attendant large uncertainties. Fugitive sources are typically near the
ground and have their greatest effect near the source. Whether they need to be considered in the assessment
starts with an identification of these sources and the potential for any sensitive receptors to be in the affected area.
The emissions estimates for fugitives are frequently based on AP-42 emission factors, Fifth Edition, Volume I
Chapter 13: Miscellaneous Sources15, which in turn have a rating of E (poor, i.e., large uncertainty).




13
   https://www.gov.nl.ca/eccc/files/env-protection-science-gd-ppd-019-2.pdf
14
   https://www.gov.nl.ca/eccc/files/env-protection-science-gd-ppd-019-2.pdf
15
   https://www.epa.gov/air-emissions-factors-and-quantification/ap-42-fifth-edition-volume-i-chapter-13-
miscellaneous-0


Version: July 2026                                                                                           Page 18
                       British Columbia Air Quality Dispersion Modelling Guideline


AERMOD and CALPUFF can handle various source configurations often associated with such emissions. These
models can account for the deposition of fugitive dust (with particle-size distributions and settling velocities) from
areas such as gravel parking lots.

Due to the large uncertainties associated with establishing fugitive emissions and the difficulties in modelling them,
consider:

        An emphasis on ambient monitoring to establish the fugitive emission contribution to air quality.

        Dispersion modelling to provide the concentration magnitude in order to demonstrate whether proactive
         fugitive emissions control/management needs to be adopted. Furthermore, the modelling can be used to
         assess the effectiveness of mitigation plans (relative changes) with less emphasis on the absolute results.

        If possible, report the modelling results of fugitive sources separately from the modelling results
         associated with sources where there is greater certainty in the emissions (point sources for example).

        EMISSIONS FROM OTHER SOURCES

It is possible that the emissions from other nearby sources need to be included in the dispersion model simulation.
Guidance on when to include such sources is provided in Section 8.1: Adding Baseline Air Quality Concentrations.

The contaminant emission rates for these other sources depend on their type and the information available (some
may be operating under permit, and others such as marine sources need to be estimated). The methods used to
estimate the contaminant emission rates would follow the same guidance as provided the previous sections.




Version: July 2026                                                                                           Page 19
                         British Columbia Air Quality Dispersion Modelling Guideline



       DISPERSION MODEL INPUT - GEOPHYSICAL DATA

AERSCREEN, AERMOD, and CALPUFF require a description of the geophysical characteristics of the location to be
modelled. This can include detailed topographical and land use information with corresponding surface
parameters such as surface roughness, albedo, and Bowen ratios.

           TERRAIN DATA

There are numerous sources of terrain data available; however, some do not cover Canada, and some are not
suited for the Guideline recommended models.

AERMAP, the AERMOD terrain processor, requires terrain data in USGS Digital Elevation Model (DEM) format or
USGS National Elevation Dataset (NED) GeoTIFF format. There is also an option to input XYZ gridded terrain
elevations manually, but they must be first converted to a form that mimics the USGS DEM data format (US
Environmental Protection Agency, 2024).

Unfortunately, USGS terrain data do not cover Canada. However, other types of terrain data that cover Canada
(e.g., Canadian Digital Elevation Data and XYZ terrain data) can be converted to AERMAP compatible format by
commercially available software or by special processing by the user. Instructions on processing XYZ data into the
USGS DEM format are available on the SCRAM website16 in a document titled “On inputting XYZ data into AERMAP”
which is embedded in the zip file XYZ Elevation Data Preparation and Entry Procedure.

TERREL, the CALPUFF terrain processor, supports processing of terrain data in the USGS DEM format, as well as
several other types of data that provide terrain at different spatial resolutions – although there are only a few that
are relevant to BC.

For BC applications of AERSCREEN, AERMOD, and CALPUFF, the recommended terrain data are:

           Canadian Digital Elevation Data (CDED) provided by Natural Resources Canada. The CDED terrain data are
            in USGS DEM compatible data formats at scales of 1:50,000 and 1:250,000. The CDED data are based on
            the North American Datum 1983 horizontal reference datum and consist of elevations relative to Mean
            Sea Level. For these data, note the following:

                o   The 250,000 map scale data have a grid resolution range of 3 to 12 arc-seconds, depending on
                    the latitude of the CDED section. For data south of 68˚ N latitude, the grid spacing is about 93 m in
                    the north-south direction and 35-65m in the east-west direction. This data type can be read
                    directly by AERMAP and TERREL without the need for any conversions.

                o   The 1:50,000 map scale data have a minimum grid resolution of 0.75 arc-seconds and a
                    maximum of 3 arc-seconds. For data south of 68° N latitude, the grid resolution is roughly 23 m in
                    the north-south direction and 11-16 m in the east-west direction. These data are not readable by
                    AERMAP, so they must be converted to a compatible format (e.g., into the equivalent USGS 7.5-
                    minute DEM data format) before they can be used by AERMAP.




16
     https://www.epa.gov/scram/air-quality-dispersion-modeling-related-model-support-programs


Version: July 2026                                                                                             Page 20
                        British Columbia Air Quality Dispersion Modelling Guideline



         GeoBC17 provides BC specific terrain data in the form of gridded DEM files with a resolution of 25 m,
          created from the 1:20,000 map scale Terrain Resource Information Management (TRIM) DEM data. The
          BC DEM files are produced in two planar projections – UTM and BC Albers. All coordinates in the file are
          based on the 1983 North American Datum, and all elevations are calculated in metres above mean sea
          level. For the BC TRIM DEM data, the following is recommended:

              o    The UTM projection of the BC TRIM DEM data can be converted into an XYZ terrain elevations
                   data file using commercially available software. The generic XYZ data file (where the first column
                   must contain the X coordinate of a specific location, the second column contains the Y
                   coordinate of the location, and the third column contains the terrain elevation) can be directly
                   read by TERREL.

              o    Alternatively, the UTM projection of the BC TRIM DEM data can also be converted to the USGS
                   7.5-minute DEM format (readable by AERMAP and TERREL) by using commercially available
                   software or by special processing by the user.

         For TERREL only, USGS SRTM3 terrain data18 a resolution of 1 arc-seconds (~30 m resolution) up to 60° N
          are available. Note that these data provide tree-top elevations (not ground-level). For high tree canopy
          areas and/or where the modelled sources are near ground-level, these data may not be appropriate.

         Terrain data in XYZ format (where X and Y are the coordinates, and Z is the terrain elevation) can be
          manually extracted from a contour map or from site specific survey data.

Before any terrain data are used as input, the following is recommended:

         Plot the terrain surface and visually scan for any anomalous points that are not consistent with the
          observed topography.

         Ensure that the terrain data are of a spatial resolution that is sufficient to resolve the important terrain
          features so that the dispersion models can properly characterize the effects of the terrain on the
          transport and dispersion.

         LAND USE DATA

For AERSCREEN and AERMOD, the following sources of data are recommended:

         GeoBC Baseline Thematic Mapping (BTM) land use data (available at the GeoBC website19).

         Canadian Land Cover, Circa 2000 (Vector) data (available at the Open Government of Canada website20).

         Extract land use manually from contour maps, aerial photography, or Google Earth images.

For CALPUFF (CALMET), a file of land use data in the form of gridded land use types is required by CTGPROG, a
preprocessor which produces the fractional land use for each grid cell in the modelling domain or it can be input
directly into the GEO.DAT file. Three sources of land use data in BC are recommended:




17
   https://www2.gov.bc.ca/gov/content?id=42995B86D9B94DE0B3DF424CE3E5E2A7
18
   https://doi.org/10.5067/MEASURES/SRTM/SRTMGL1.003
19
   https://www2.gov.bc.ca/gov/content?id=5C9D77FBC4C64A61979F78383F3DEC93
20
   https://open.canada.ca/en


Version: July 2026                                                                                             Page 21
                       British Columbia Air Quality Dispersion Modelling Guideline



         GeoBC BTM land use data. The BTM data are in polygonised format, and they cannot be read directly be
          CTGPROG. The BTM data files must be translated into a raster image in order to provide a land use code
          for the location of each grid cell. Such a raster file must be further prepared for CTGPROG (several steps
          are needed – these are described in Section 4.4.1: Processing GeoBC BTM Land Use Data). Furthermore, if
          the BTM data are used, the BC land use codes must be translated into the equivalent USGS land use codes
          used in GEO.DAT.

         GeoGratis Canadian Land Cover, Circa 2000 (Vector) data21. Since these data are also in polygonised
          format, they should be first rasterized and further prepared for CTGPROG to use.

         USGS Global Land Use Characterization data22 file for North America (Lambert Azimuthal Equal Area
          Projection File). This provides land use in USGS land use codes at a 1 km grid resolution that can be read
          and processed directly by CTGPROG. This resolution may be adequate for regional scale applications of
          CALPUFF.

         Extract land use manually from 1:50,000 contour maps, aerial photography, or Google Earth images.

         PROCESSING LAND USE DATA FOR AERSCREEN AND AERMOD (AERMET) SURFACE
         PARAMETERS

AERSCREEN and AERMOD require the specification of surface roughness, albedo, and Bowen ratio which are
defined by different land use types in the area surrounding the location where the input meteorological data were
collected. The user can enter direction specific (i.e., wind sector), monthly, annual, and season-dependent surface
values as well. AERSCREEN and AERMOD results can be sensitive to these variables, so their selection (especially
surface roughness) is important. The AERMOD Implementation Guide23 (US Environmental Protection Agency,
2024) provides guidance on how to determine these parameters:

         The surface roughness length “should be based on an inverse-distance weighted geometric mean for a
          default upwind distance of 1 kilometer relative to the measurement site.”

         The Bowen ratio “should be based on a simple unweighted geometric mean (i.e., no direction or distance
          dependency) for a representative domain, with a default domain defined by a 10km by 10km region
          centered on the measurement site.”

         The albedo “should be based on a simple unweighted arithmetic mean (i.e., no direction or distance
          dependency) for the same representative domain as defined for Bowen ratio, with a default domain
          defined by a 10km by 10km region centered on the measurement site.”

Surface parameters can be entered manually according to the guidance provided in the AERMOD Implementation
Guide24 (US Environmental Protection Agency, 2024), or through the use of AERSURFACE – a utility that processes
land cover data and applies objective methods to obtain realistic and reproducible surface characteristics.
AERSURFACE cannot be used in BC as it requires the USGS National Land Cover Data. The original method (i.e., pre-
AERSURFACE) to determine surface parameters is recommended for AERMOD applications in BC.



21
   https://open.canada.ca/data/en/dataset/97126362-5a85-4fe0-9dc2-915464cfdbb7
22
   https://www.usgs.gov/centers/eros/science/usgs-eros-archive-land-cover-products-global-land-cover-
characterization-glcc?qt-science_center_objects=0#qt-science_center_objects
23
   https://gaftp.epa.gov/Air/aqmg/SCRAM/models/preferred/aermod/aermod_implementation_guide.pdf
24
   https://gaftp.epa.gov/Air/aqmg/SCRAM/models/preferred/aermod/aermod_implementation_guide.pdf


Version: July 2026                                                                                           Page 22
                       British Columbia Air Quality Dispersion Modelling Guideline


The steps in this process are as follows:

        Obtain land use from contour maps, aerial photography, Google Earth images, or maps created from
         either the GeoBC BTM land use data or the GeoGratis Canadian Land Cover, Circa 2000 (Vector) data. On a
         map, draw a 1-km circle around the location of the meteorological data collection site.

        Within the 1 km radius, identify areas of different land use according to the 8 types listed in the first
         column of Table 4.1. For example, if the area within the radius includes forest and urban areas, then
         outline those areas that are forest and those that are urban. If GeoBC BTM land use maps are used, use
         Table 4.1 as guidance to determine the equivalent land use types.

Table 4.1          Land Use Types and Equivalent GeoBC BTM Land Use

                            Land Use Type                           GeoBC BTM Land Use Category
            Water (fresh water and sea water)                Salt Water, Fresh Water, Rivers, Lakes
            Deciduous Forest                                 Old Forest, Young Forest
            Coniferous Forest                                Old Forest, Young Forest
            Swamp                                            Wetlands, Estuaries
            Cultivated Land                                  Agriculture
            Grassland                                        Transmission Line Corridor, Rangeland
            Urban                                            Urban, Residential Agriculture Mixes
            Desert Shrubland                                 Shrubs, Rangeland

        AERMOD can account for different land use types within the radius by allowing the user to specify land
         use types for up to 12 different contiguous, non-overlapping wind direction sectors that define unique
         land use types. Note the following:

               o   The sectors are defined as compass degrees from the north (north is 360 degrees), clockwise.

               o   The minimum sector width is 30 degrees.

               o   The sectors must cover the full circle so that the end value of one sector matches the beginning
                   of the next sector.

               o   The beginning direction is considered to be part of the sector, while the ending direction is not.

         For example, if there are grasslands that dominate the land use type to the east of the meteorological
         data collection site, a sector can be then defined for that area with the unique land use type as
         “Grassland”.

        Once the dominant land use type for the different sectors has been defined, then the corresponding
         surface parameters (roughness height, albedo, Bowen ratio) need to be assigned for the different seasons
         and wind direction sectors (and in the case of Bowen ratio, for the different moisture conditions).

The following sections describe each surface parameter and provide Tables of values for the user to select. The
definitions of seasons used in the following Tables are based on the surface conditions as defined below:

        “Spring”: when vegetation is emerging or partially green. This applies for 1–2 months after the last killing
         frost.

        “Summer”: when vegetation is lush and healthy.



Version: July 2026                                                                                           Page 23
                        British Columbia Air Quality Dispersion Modelling Guideline



          “Autumn”: periods when freezing conditions are common, deciduous trees are leafless, crops are not yet
           planted or are already harvested (bare soil exposed), grass surfaces are brown, and no snow is present.

          “Winter”: for snow-covered surfaces and subfreezing temperatures.

Surface parameters can be specified for each month, season or for a whole year. Given the range of surface
conditions that can apply through the year in BC the seasonal option is recommended. Note that AERMET defines
seasons by monthly groupings of Winter (Dec, Jan, Feb), Spring (March, April, May), Summer (June, July, Aug), and
Autumn (Sept, Oct, Nov). These seasonal definitions are designed for mid-latitude regions and therefore may not
apply to the northern or southwestern portion of BC. For example, Winter: “snow-covered surfaces and
subfreezing temperatures” may extend from October to April in the north. In this case, it would be more
appropriate to use the monthly option to specify surface parameters, where Winter surface parameter values are
assigned for October through April. Similarly for south western BC “Winter” as defined above does not occur
(unless it is in elevated terrain). In this case it would be more appropriate to use the Autumn surface parameters
for Winter.


           SURFACE ROUGHNESS HEIGHT

The AERMOD Implementation Guide25 (US Environmental Protection Agency, 2024) indicates that “the relationship
between the surface roughness upwind of the measurement site and the measured wind speeds is generally the
most important consideration.”

AERMOD can account for monthly, seasonal, annual, and wind sector variations in surface roughness. As per
recommendations above, classify land use type according to Table 4.1 then assign the corresponding roughness
height for each season and if required, each direction sector according to Table 4.2.

Table 4.2       Surface Roughness Heights (m), For Typical Land Use Types and Seasons (Environmental
Research and Technology, 1987)

                       Land Use Type                     Spring          Summer         Autumn         Winter
       Water (fresh water and sea water)              0.0001           0.0001          0.0001        0.0001
       Deciduous Forest                               1.00             1.30            0.80          0.50
       Coniferous Forest                              1.30             1.30            1.30          1.30
       Swamp                                          0.20             0.20            0.20          0.05
       Cultivated Land                                0.03             0.20            0.05          0.01
       Grassland                                      0.05             0.10            0.01          0.001
       Urban                                          1.00             1.00            1.00          1.00
       Desert Shrubland                               0.30             0.30            0.30          0.15


           ALBEDO

The albedo is the fraction of total incoming solar radiation that is reflected from the ground back to the space
without absorption. Table 4.3 provides seasonal values of noon-time albedo for different land use types.




25
     https://gaftp.epa.gov/Air/aqmg/SCRAM/models/preferred/aermod/aermod_implementation_guide.pdf


Version: July 2026                                                                                          Page 24
                       British Columbia Air Quality Dispersion Modelling Guideline


Table 4.3         Noon-Time Albedo For Typical Land Use Types and Seasons

                     Land Use Type                        Spring          Summer          Autumn          Winter
     Water (fresh water and sea water)                 0.12             0.10             0.14           0.20
     Deciduous Forest                                  0.12             0.12             0.12           0.50
     Coniferous Forest                                 0.12             0.12             0.12           0.35
     Swamp                                             0.12             0.14             0.16           0.30
     Cultivated Land                                   0.14             0.20             0.18           0.60
     Grassland                                         0.18             0.18             0.20           0.60
     Urban                                             0.14             0.16             0.18           0.35
     Desert Shrubland                                  0.30             0.28             0.28           0.45



         BOWEN RATIO

The Bowen ratio is an indicator of surface moisture, and it is the ratio of the sensible heat flux to the latent heat
flux. In AERMOD, the value depends on a number of factors including land use type, season, and moisture
conditions.

Table 4.4 a) – c) provide seasonal values of Bowen ratios for different land use types and moisture conditions (dry,
average, wet). Wet, average, and dry moisture conditions are defined by the climatological normals based on a 30-
year precipitation data record. If the amount of precipitation during the simulation period (the period covered by
the input meteorological data used for modelling) is in the upper 30th percentile, then assign “wet” conditions; if in
the lower 30th percentile then assign “dry” conditions, and if in between then assign “average” condition.

Table 4.4 a       Daytime Bowen Ratios, For Typical Land Use Types and Seasons

                                              (Dry Moisture Conditions)

                     Land Use Type                           Spring       Summer          Autumn         Winter*
     Water (fresh water and sea water)                 0.1              0.1              0.1            2.0**
     Deciduous Forest                                  1.5              0.6              2.0            2.0
     Coniferous Forest                                 1.5              0.6              1.5            2.0
     Swamp                                             0.2              0.2              0.2            2.0
     Cultivated Land                                   1.0              1.5              2.0            2.0
     Grassland                                         1.0              2.0              2.0            2.0
     Urban                                             2.0              4.0              4.0            2.0
     Desert Shrubland                                  5.0              6.0              10.0           10.0
*        Winter Bowen ratios depend on whether a snow cover is present continuously, intermittently or seldom.
         The values for winter are applicable for continuous snow covers; the values between autumn and winter
         may be more applicable for seldom snow covers.
**       If water body frozen over, use this value.




Version: July 2026                                                                                            Page 25
                      British Columbia Air Quality Dispersion Modelling Guideline


Table 4.4 b      Daytime Bowen Ratios, For Typical Land Use Types and Seasons

                                         (Average Moisture Conditions)

                     Land Use Type                         Spring      Summer         Autumn        Winter*
     Water (fresh water and sea water)               0.1             0.1             0.1           1.5**
     Deciduous Forest                                0.7             0.3             1.0           1.5
     Coniferous Forest                               0.7             0.3             0.8           1.5
     Swamp                                           0.1             0.1             0.1           1.5
     Cultivated Land                                 0.3             0.5             0.7           1.5
     Grassland                                       0.4             0.8             1.0           1.5
     Urban                                           1.0             2.0             2.0           1.5
     Desert Shrubland                                3.0             4.0             6.0           6.0
*        Winter Bowen ratios depend on whether a snow cover is present continuously, intermittently or seldom.
         The values for winter are applicable for continuous snow covers; the values between autumn and winter
         may be more applicable for seldom snow covers.
**       If water body frozen over, use this value.

Table 4.4 c      Daytime Bowen Ratios, For Typical Land Use Types and Seasons

                                           (Wet Moisture Conditions)

                     Land Use Type                         Spring      Summer         Autumn        Winter*
     Water (fresh water and sea water)               0.1             0.1             0.1           0.3**
     Deciduous Forest                                0.3             0.2             0.4           0.5
     Coniferous Forest                               0.3             0.2             0.3           0.3
     Swamp                                           0.1             0.1             0.1           0.5
     Cultivated Land                                 0.2             0.3             0.4           0.5
     Grassland                                       0.3             0.4             0.5           0.5
     Urban                                           0.5             1.0             1.0           0.5
     Desert Shrubland                                1.0             1.5             2.0           2.0
*        Winter Bowen ratios depend on whether a snow cover is present continuously, intermittently or seldom.
         The values for winter are applicable for continuous snow covers; the values between autumn and winter
         may be more applicable for seldom snow covers.
**       If water body frozen over, use this value.

       PROCESSING LAND USE DATA FOR CALPUFF (CALMET) S URFACE PARAMETERS

CALMET requires a GEO.DAT file which contains the dominant land use type for each grid cell, gridded terrain
elevation, and land use weighted fields of surface parameters (surface roughness length, albedo, Bowen ratio, soil
heat flux, vegetation leaf area index, and anthropogenic heat flux). In order to create the land use portion of
GEO.DAT, raw land use data can be first processed by CTGPROG, a preprocessor. The file produced by CTGPROG is
in turn used by MAKEGEO to create the gridded land use type and surface parameters that make up part of the
GEO.DAT file.




Version: July 2026                                                                                       Page 26
                       British Columbia Air Quality Dispersion Modelling Guideline


Due to the wide variation in climate in BC, the definition of seasons can also vary dramatically. It is good practice to
select seasonally specific surface parameters to be entered into CALMET as gridded fields using five seasonal
categories as follows:

        Season 1: Midsummer with lush vegetation.

        Season 2: Autumn with cropland that has not been harvested.

        Season 3: Winter 1, late autumn after frost, no snow on the ground.

        Season 4: Winter 2, snow on the ground and subfreezing.

        Season 5: Transitional spring with partially green short annuals.

Note that winter is defined as two seasons: one associated with frozen, snow-covered water bodies, and the other
associated with open water.

The seasonal categories for regions in BC vary due to the different climates experienced in those regions. Regions
can have winter seasons that last from 4 to 6 months with some regions never having persistent snow on the
ground and hence never experience Winter 2 conditions. Table 4.5 provides a list of seasonal categories for
different months and latitude. As cited in Brook et al. (1999), “it was assumed that for areas with the same latitude,
the season category is the same. In reality, areas with the same latitude experience different seasonal
characteristics depending on the distance from the sea, the underlying surfaces, etc.” Therefore, it is
recommended that the user refer to Table 4.5 as a guide and corroborate the selection with observations of snow
on the ground for the modelling period selected. Please consult with the Ministry to determine season selection
for a modelling project.

Table 4.5         Variation of Seasonal Category by Month and Latitude

 Latitude        Jan     Feb      Mar      Apr     May       Jun      Jul     Aug      Sep      Oct     Nov      Dec
 45ºN-50ºN        4       4        4        4       5         1        1       2        2        3       3        4
 50ºN-55ºN        4       4        4        4       5         1        1       2        2        3       4        4
 55ºN-60ºN        4       4        4        4       5         5        1       2        2        3       4        4
 60ºN-65ºN        4       4        4        4       4         5        1       2        2        3       4        4


         PROCESSING GEOBC BTM LAND USE DATA

The GeoBC BTM land use data cannot be processed directly by CTGPROG, and the raw data must be first
manipulated according to the following steps:

    1.   convert the polygon vector BTM file into a fine grid-resolution (e.g., 100 m) raster image using a GIS utility.
         From the raster image, create an XYZ file where X is the BTM land use type, and Y, Z are the coordinates
         (in longitude and latitude).




Version: July 2026                                                                                            Page 27
                         British Columbia Air Quality Dispersion Modelling Guideline


       2.   create a generic land use data set (XYZ file) for CTGPROG to use. Table 4.6 provides a sample of generic
            land use data file. For more information on the form of the generic file, see Table 2-1226 of The CALPUFF
            User’s Manual (Exponent, 2011).

Table 4.6           Sample File of Generic Land Use Data Set

GENERIC.LANDUSE 1.0             LU, Longitude, Latitude (free-format)
2
Prepared by User
Longitude is positive to east, Latitude is positive to north
LL
WGS-84 02-21-2003
DEG
24
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
    9 -119.440857          49.3442345
    9 -119.439484          49.3442612
    9 -119.438103          49.3442917
    9 -119.436729          49.3443184
    9 -119.435356          49.3443489
    14 -119.433983         49.3443794
    14 -119.432602         49.3444061
    14 -119.431229         49.3444366
    14 -119.429855         49.3444633
    14 -119.428482         49.3444939
    14 -119.427101         49.3445244
    14 -119.425728         49.3445511
    8 -119.424355          49.3445816
    8 -119.422981          49.3446083
    8 -119.421600          49.3446388
    22 -119.998535         49.9995651
    22 -119.997147         49.9996033
    22 -119.991570         49.9997444
    13 -119.988785         49.9998207

Before the output file from CTGPROG is used by MAKEGEO to compute dominant land use categories and
weighted surface parameter data for GEO.DAT, the default table in Subgroup (4b) in the MAKEGEO control input
file (MAKEGEO.INP) must be replaced.

The default table assigns surface parameters for the USGS 38-Category System and thus cannot be used for the
GeoBC BTM land use categories as they are not directly equivalent. Table 4.7 provides a cross-reference list that
maps the GeoBC BTM land use categories to the USGS categories to use with CALMET.




26
     https://www.calpuff.org/calpuff/download/CALPUFF_Version6_UserInstructions.pdf


Version: July 2026                                                                                         Page 28
                                           British Columbia Air Quality Dispersion Modelling Guideline



Table 4.7      Translation Table of GeoBC BTM Land Use Categories to CALMET Categories

                      GeoBC BTM Land Use Category                                    CALMET Land Use Category
            Land Use Type                                       Land Use Type
                                       Description                                                     Description
            Numeric Value                                       Numeric Value
                 1                     Agriculture                   20                             Agricultural Land
                 2           Residential Agriculture Mixtures        20                             Agricultural Land
                 3                       Alpine                      70                                Barren Land
                 4             Sub alpine Avalanche Chutes           70                                Barren Land
                 5                   Barren Surfaces                 70                                Barren Land
                 6                  Recently Burned                  70                                Barren Land
                 7                      Estuaries                    60                                 Wetlands
                 8                     Old Forest                    42                          Forest Land (coniferous)
                 9                    Young Forest                   42                          Forest Land (coniferous)
                 10                     Highways                     70                                Barren Land
                 11                Glaciers and Snow                 90                           Perennial Snow or Ice
                                                                     51                          Water (small water body)
                 12                       Lakes
                                                                     55            Water (large water body with buoy temperature data)
                 13                 Recently Logged                  30                                Rangeland
                 14                Selectively Logged                42                            Forest (coniferous)
                 15                      Mining                      70                                Barren Land
                 16                   Range Lands                    30                                Rangeland
                 17               Recreation Activities              42                          Forest Land (coniferous)
                 18                Rivers, double line               51                                  Water
                 19                      Shrubs                      32                                   Shrub
                 20            Transmission Line Corridor            30                                Rangeland
                 21                      Urban                       10                          Urban or Built-up Land
                 22                     Wetlands                     60                                  Wetland
                 23                   Fresh Water                    51                                  Water
                 24                    Salt Water                    55            Water (large water body with buoy temperature data)



Version: July 2026                                                                             Page 29
                             British Columbia Air Quality Dispersion Modelling Guideline



Table 4.8 through 4.12 depict specific parameters suggested for each land use type for the five seasons. The use of
these Tables is recommend as there can be dramatic changes that can occur to the surface parameters over a year
which, in turn, can affect the concentration predictions.

The seasonally specific surface parameters of surface roughness length, albedo and Bowen ratio are derived from the
AERSURFACE User’s Guide27. The Leaf Area Index values shown in those Tables are derived from (Zhang, Brook, & Vet,
2003) and (Zhang, Moran, Makar, Brook, & Gong, 2002) where the soil heat flux values are the CALMET default values
for each category. As shown in Table 4.8 – Table 4.12, the Soil Heat Flux is the only surface parameter with no
seasonal variation, where only the “Water” category experiences a change in Season 4.

Anthropogenic Heat Flux (AHF) can be calculated based on population density and per capita energy usage (Oke,
1987). The values given in Table 4.8 through Table 4.12 are for a mid-latitude city with mild winters and summers such
as Vancouver. For cities with colder winters and warmer summers, different values for the per capita energy usage
need to be selected and hence different AHF values will be assigned. Gridded population data needed to calculate
AHF can be downloaded from Canadian census data available from Statistics Canada28 and can be converted to match
the chosen CALMET domain.

Large bodies of water for which air temperature and Sea Surface Temperature (SST) data exist should be flagged by a
different land use category in GEO.DAT in order for CALMET to use specific overwater temperature algorithms.
Typically such water bodies are flagged with a land use category of 55. The overwater algorithms are only used for
those large water bodies if JWAT1 and JWAT2 are also set to 55 in CALMET.INP, and if buoy/ship temperature data
(air temperature and SST) are available in a SEA.DAT file (or several SEA.DAT files). Note that JWAT1 and JWAT2 are
set in CALMET.INP and only matter for the offshore water temperature interpolation algorithms, whereas IWAT1 and
IWAT2, set in the GEO.DAT header, should encompass the whole range of water categories (i.e., IWAT1=50 and
IWAT2=55) and should affect winds , turbulence, and mixing heights over all bodies of water.




27
     https://gaftp.epa.gov/Air/aqmg/SCRAM/models/related/aersurface/aersurface_ug_v24142.pdf
28
     https://www12.statcan.gc.ca/census-recensement/2021/geo/sip-pis/boundary-limites/index-eng.cfm


Version: July 2026                                                                                        Page 30
                          British Columbia Air Quality Dispersion Modelling Guideline


Table 4.8       Season 1 (Mid-Summer) Geophysical Parameters

CALMET                               Surface
                                                           Bowen    Soil Heat   Anthropogenic    Leaf Area
Land Use         Description        Roughness   Albedo
                                                            Ratio     Flux        Heat Flux        Index
  Type                                Length
   10               Urban              0.54      0.16       0.8       0.25           8.0            0.3
   20            Agricultural          0.20      0.20       0.50      0.15           0.0            2.0
   30             Rangeland            0.15      0.20       0.5       0.15           0.0            1.0
   32               Shrub               0.3      0.18         1       0.15           0.0            4.5
   40         Transition Forest         0.3      0.18       1.0       0.15           0.0            4.5
   41             Deciduous            1.30      0.16       0.3       0.15           0.0            3.4
   42             Coniferous           1.30      0.12       0.3       0.15           0.0            5.0
   43               Mixed               1.3      0.14       0.3       0.15           0.0            4.5
   51         Small Water Body         0.001     0.1        0.1       1.00           0.0            0.0
   55         Large Water Body         0.001     0.1        0.1       1.00           0.0            0.0
   60              Wetland              0.2      0.14       0.1       0.3            0.0            0.2
   61         Forested Wetland          0.7      0.14       0.2       0.3            0.0            0.2
   62       Non-forested Wetland        0.2      0.14       0.1       0.3            0.0            0.2
   70            Barren Land           0.05      0.2        1.5       0.15           0.0            0.0
   90       Perennial Snow or Ice       0.2      0.7        0.5       0.15           0.0            0.0



Table 4.9       Season 2 (Autumn) Geophysical Parameters

CALMET                               Surface
                                                           Bowen    Soil Heat   Anthropogenic    Leaf Area
Land Use         Description        Roughness   Albedo
                                                            Ratio     Flux        Heat Flux        Index
  Type                               Length
   10               Urban             0.54       0.16       1.0       0.25          12.0            0.2
   20            Agricultural         0.20       0.20       0.70      0.15           0.0            1.5
   30             Rangeland           0.15       0.20       0.7       0.15           0.0            1.0
   32               Shrub              0.3       0.18       1.5       0.15           0.0            3.5
   40         Transition Forest        0.3       0.18       1.5       0.15           0.0            3.5
   41             Deciduous           1.30       0.16       1.0       0.15           0.0            1.9
   42            Coniferous           1.30       0.12       0.8       0.15           0.0            5.0
   43               Mixed              1.3       0.14       0.9       0.15           0.0            3.5
   51         Small Water Body        0.001      0.1        0.1       1.00           0.0            0.0
   55         Large Water Body        0.001      0.1        0.1       1.00           0.0            0.0
   60              Wetland             0.2       0.14       0.1       0.3            0.0            0.2
   61         Forested Wetland         0.7       0.14       0.2       0.3            0.0            0.2
   62       Non-forested Wetland       0.2       0.14       0.1       0.3            0.0            0.2
   70            Barren Land          0.05       0.2        1.5       0.15           0.0            0.0
   90       Perennial Snow or Ice      0.2       0.7        0.5       0.15           0.0            0.0




Version: July 2026                                                                              Page 31
                           British Columbia Air Quality Dispersion Modelling Guideline


Table 4.10       Season 3 (Winter 1) Geophysical Parameters

CALMET                                Surface
                                                              Bowen    Soil Heat   Anthropogenic    Leaf Area
Land Use          Description        Roughness   Albedo
                                                               Ratio     Flux        Heat Flux        Index
  Type                                Length
   10                Urban             0.50        0.18        1.0       0.25          21.0           0.1
   20             Agricultural         0.02        0.18        0.7       0.15           0.0           1.0
   30              Rangeland           0.02        0.18        0.70      0.15           0.0           1.0
   32                Shrub              0.3        0.18        1.5       0.15           0.0           2.3
   40          Transition Forest        0.3        0.18        1.5       0.15           0.0           2.3
   41              Deciduous           0.60        0.17        1.0       0.15           0.0           0.1
   42             Coniferous           1.30        0.12        0.8       0.15           0.0           5.0
   43                Mixed             0.95        0.14        0.9       0.15           0.0           2.3
   51          Small Water Body        0.001       0.10        0.10      1.00           0.0           0.0
   55          Large Water Body        0.001       0.10        0.10      1.00           0.0           0.0
   60               Wetland             0.2        0.14        0.1       0.3            0.0           0.1
   61          Forested Wetland         0.6        0.14        0.3       0.3            0.0           0.1
   62        Non-forested Wetland       0.2        0.14        0.1       0.3            0.0           0.1
   70             Barren Land          0.05        0.2         1.5       0.15           0.0           0.05
   90        Perennial Snow or Ice      0.2        0.7         0.5       0.15           0.0           0.0



Table 4.11       Season 4 (Winter 2) Geophysical Parameters

CALMET                                Surface
                                                              Bowen    Soil Heat   Anthropogenic    Leaf Area
Land Use          Description        Roughness   Albedo
                                                               Ratio     Flux        Heat Flux        Index
  Type                                Length
   10                Urban             0.50        0.45        0.5       0.15          17.0           0.0
   20             Agricultural         0.01        0.60        0.5       0.15           0.0           0.0
   30              Rangeland           0.01        0.60        0.5       0.15           0.0           1.0
   32                Shrub             0.15        0.5         0.5       0.15           0.0           2.3
   40          Transition Forest        0.2        0.50        0.5       0.15           0.0           2.3
   41              Deciduous           0.50        0.50        0.5       0.15           0.0           0.0
   42             Coniferous           1.30        0.35        0.5       0.15           0.0           5.0
   43                Mixed              0.9        0.42        0.5       0.15           0.0           2.3
   51          Small Water Body        0.002       0.7         0.5       0.15           0.0           0.0
   55          Large Water Body        0.002       0.7         0.5       0.15           0.0           0.0
   60               Wetland             0.1        0.3         0.5       0.3            0.0           0.0
   61          Forested Wetland         0.5        0.3         0.5       0.3            0.0           0.0
   62        Non-forested Wetland       0.1        0.3         0.5       0.3            0.0           0.0
   70             Barren Land          0.05        0.6         0.5       0.15           0.0           0.05
   90        Perennial Snow or Ice      0.2        0.7         0.5       0.15           0.0           0.0




Version: July 2026                                                                                 Page 32
                           British Columbia Air Quality Dispersion Modelling Guideline


Table 4.12       Season 5 (Transitional Spring) Geophysical Parameters

CALMET                                   Surface
                                                                 Bowen      Soil Heat    Anthropogenic    Leaf Area
Land Use          Description           Roughness   Albedo
                                                                  Ratio       Flux         Heat Flux        Index
  Type                                   Length
   10                Urban                0.52       0.16         0.8         0.25            15.0           0.2
   20             Agricultural            0.03       0.14         0.30        0.15            0.0            1.0
   30              Rangeland              0.03       0.14         0.3         0.15            0.0            1.0
   32                Shrub                 0.3       0.18           1         0.15            0.0            3.3
   40          Transition Forest           0.3       0.18         1.0         0.15            0.0            3.3
   41              Deciduous              1.00       0.16         0.7         0.15            0.0            0.8
   42             Coniferous              1.30       0.12         0.7         0.15            0.0            5.0
   43                Mixed                1.15       0.14         0.7         0.15            0.0            3.3
   51          Small Water Body           0.001      0.1          0.1         1.00            0.0            0.0
   55          Large Water Body           0.001      0.1          0.1         1.00            0.0            0.0
   60               Wetland                0.2       0.14         0.1         0.3             0.0            0.1
   61          Forested Wetland            0.7       0.14         0.2         0.3             0.0            0.1
   62        Non-forested Wetland          0.2       0.14         0.1         0.3             0.0            0.1
   70             Barren Land             0.05       0.2          1.5         0.15            0.0            0.0
   90        Perennial Snow or Ice         0.2       0.7          0.5         0.15            0.0            0.0

Refer to Table 4.8 through Table 4.12 to create a GeoBC BTM-specific Subgroup (4b) table with the new appropriate
values of seasonally specific surface parameters for each land use category. Table 4.13 provides an example of a
GeoBC BTM-specific Subgroup (4b) table in MAKEGEO.INP. When the original default table is replaced by a user
updated table, run MAKEGEO using the output from both CTGPROG and TERREL to create a GEO.DAT file.

Table 4.13       Example of Subgroup (4b) Table of the MAKEGEO Control Input File for Season 1 (Mid-Summer)

                                LAND USE PROPERTIES AND OUTPUT MAP (NINICAT entries)

       Input                                         Soil Heat   Anthropogenic        Leaf     Output
                    z0       Albedo       Bowen
     Category                                          Flux        Heat Flux          Area    Category
                   (m)       (0 to 1)      Ratio
         ID                                         Parameter       (W/m2)           Index       ID
     ! X = 10,    0.54,         0.16,       0.8,       0.25,          8.0,            0.3,      10 !       !END!
     ! X = 20,    0.20,         0.20,       0.5,       0.15,          0.0,            2.0,      20 !       !END!
     ! X = 30,    0.15,         0.20,       0.5,       0.15,          0.0,            1.0,      30 !       !END!
     ! X = 32,    0.30,         0.18,       1.0,       0.15,          0.0,            0.0,      32 !       !END!
     ! X = 40,    0.30,         0.18,       1.0,       0.15,          0.0,            4.5,      40 !       !END!
     ! X = 42,    1.30,         0.12,       0.3,       0.15,          0.0,            5.0,      42 !       !END!
     ! X = 51,    0.001,        0.10,       0.1,        1.0,          0.0,            0.0,      51 !       !END!
     ! X = 55,    0.001,        0.10,       0.1,        1.0,          0.0,            0.0,      55 !       !END!
     ! X = 60,    0.20,         0.14,       0.1,       0.30,          0.0,            0.2,      60 !       !END!
     ! X = 70,    0.05,         0.20,       1.5,       0.15,          0.0,            0.0,      70 !       !END!
     ! X = 90,    0.20,         0.70,       0.5,       0.15,          0.0,            0.0,      90 !       !END!



        PROCESSING GEOGRATIS CANADIAN LAND COVER, CIRCA 2000 (VECTOR) DATA

Follow the method mentioned earlier in 4.2: Land Use Data to process the GeoGratis Canadian Land Cover, Circa 2000
(Vector) data, except that Table 4.14 should be used as a cross-reference list for converting the GeoGratis land use
categories to the USGS land use categories.



Version: July 2026                                                                                       Page 33
                            British Columbia Air Quality Dispersion Modelling Guideline


Table 4.14        Translation Table of GeoGratis Land Use Categories to CALMET Categories

 GeoGratis Canadian Land Cover, Circa 2000 (Vector) Category                   CALMET Land Use Category

 Land Use Type                                                       Land Use Type
                                    Description                                               Description
Numerical Value                                                     Numerical Value
       20                              Water                              50                     Water
       30                      Barren/Non-vegetated                       70                  Barren Land
       31                             Snow/Ice                            90              Perennial Snow or Ice
       32                           Rock/Rubble                           70                  Barren Land
       33                          Exposed Land                           70                  Barren Land
       34                            Developed                            10             Urban or Built-up Land
       35                   Sparsely vegetated bedrock                    70                  Barren Land
       36                 Sparsely vegetated till-colluvium               70                  Barren Land
       37            Bare soil with cryptogam crust - frost boils         70                  Barren Land
       40                              Bryoids                            70                  Barren Land
       50                            Shrubland                            32                     Shrub
       51                            Shrub - Tall                         32                     Shrub
       52                           Shrub - Low                           32                     Shrub
       53                      Prostrate dwarf shrub                      32                     Shrub
       80                             Wetland                             60                    Wetland
       81                         Wetland – Treed                         61                Forested Wetland
       82                         Wetland – Shrub                         62             Non-forested Wetland
       83                         Wetland – Herb                          62             Non-forested Wetland
      100                               Herb                              32                     Shrub
      101                    Tussock graminoid tundra                     32                     Shrub
      102                           Wet sedge                             32                     Shrub
                     Moist to dry non-tussock graminoid/dwarf
      103                                                                 32                     Shrub
                                   shrub tundra
      104           Dry graminoid prostrate dwarf shrub tundra            32                     Shrub
      110                            Grassland                            30                   Rangeland
      120                   Cultivated Agricultural Land                  20                  Agricultural
      121                         Annual Cropland                         20                  Agricultural
      122                 Perennial Cropland and Pasture                  20                  Agricultural
      200                       Forest/Tree classes                       42                   Coniferous
      210                        Coniferous Forest                        42                   Coniferous
      211                        Coniferous Dense                         42                   Coniferous
      212                         Coniferous Open                         42                   Coniferous
      213                        Coniferous Sparse                        42                   Coniferous
      220                        Deciduous Forest                         41                   Deciduous
      221                         Broadleaf Dense                         41                   Deciduous
      222                         Broadleaf Open                          41                   Deciduous
      223                         Broadleaf Sparse                        41                   Deciduous



Version: July 2026                                                                                       Page 34
                             British Columbia Air Quality Dispersion Modelling Guideline



         230                         MixedWood                              40                  Transition Forest
         231                      Mixedwood Dense                           43                       Mixed
         232                       Mixedwood Open                           40                  Transition Forest
         233                      Mixedwood Sparse                          40                  Transition Forest


          PROCESSING MANUALLY ASSIGNED LAND USE TYPES

Finally, land use types can be manually assigned for each grid point by using contour maps, aerial photography,
Google Earth images, or land use maps created from the GeoBC BTM land use data or the GeoGratis Canadian Land
Cover Circa, 2000 (Vector) data . The steps to do this are listed as follows:

    1.    Examine the land cover in the domain and assign an equivalent CALMET land use type (i.e., USGS land use
          type) that is included in Table 4.8 (first column) for each grid point.

    2.    Create a generic land use data file in the form of XYZ, where X is the CALMET land use type, and Y, Z are the
          coordinates (in longitude and latitude). See Table 4.6 for a sample generic land use data file.

    3.    Run CTGPROG with the generic land use data file.

    4.    Replace the original Subgroup (4b) table in MAKEGEO.INP with a new table with appropriate values of
          surface parameters (following the instructions regarding this table described earlier).

    5.    Run MAKEGEO with the output from CTGPROG and TERREL to create a GEO.DAT file.




Version: July 2026                                                                                           Page 35
                                British Columbia Air Quality Dispersion Modelling Guideline



       DISPERSION MODEL INPUT - METEOROLOGICAL DATA

Air quality dispersion models require meteorological data as input, either in the form of a matrix of pre-determined
meteorological conditions (Screening Models) or a time series of hourly sequential meteorological data (Refined
Models). Regarding the latter, in order to reduce uncertainty in the model predictions, the meteorological data must
be representative of the expected conditions.

In order of preference, the Guideline recommends the following for meteorological input to dispersion models:

           A site specific collection program.

           Measurements from a different location that represent the key features of the conditions at the site.

           Output produced by an NWP model.

           ON-SITE AND EXISTING ALTERNATE SOURCES OF METEOROLOGICAL DATA

Ideally, the meteorological input data would come from a specially designed and sited meteorological collection
program and would represent the dispersion conditions at the site. Such data can be from a site specific (on-site)
meteorological collection program or from a nearby collection program that provides data that are “representative”
of the site conditions.

Based on Meteorological Monitoring Guidance for Regulatory Modeling Applications 29 (US Environmental Protection
Agency, 2000) consider the following factors in determining whether an alternate source of meteorological data from
a nearby existing collection program is appropriate for the site under consideration:

           Determine whether the geophysical situation at an alternative site characterizes the geophysical conditions
            at the site under consideration.

           Determine whether the climatic regimes between the existing meteorological program is similar to the site
            under consideration.

           Determine the purpose of the existing collection program and the measured parameters, the type of
            instrumentation and data collection, and the height of the measurements. Certain parameters may be
            missing or not suitable for modelling.

           Determine the instrument specifications, siting criteria and data treatment (response thresholds, data
            collection and handling protocols, special data entry flags). This will indicate the limitations and quality of the
            data. If this information is not available, then review the following and apply professional judgment to assess
            data quality:

                o    low wind speed thresholds and frequencies

                o    consistency with expected behaviour for the given location

                o    range checks

                o    hour-to-hour changes




29
     https://www.epa.gov/sites/default/files/2020-10/documents/mmgrma_0.pdf


Version: July 2026                                                                                                Page 36
                                British Columbia Air Quality Dispersion Modelling Guideline


Finally, if the application is for a Level 3 assessment for a major project with extensive emissions in a sensitive receptor and
meteorologically complex area where alternate sources of information do not exist or are inadequate, a new site specific
meteorological collection program may be required. Before such a program is started, the following is recommended:

           Consult with the Ministry regarding the need for a new site specific data collection program. This should be
            identified in the Dispersion Modelling Plan. For meteorological monitoring it is critical to start as early as
            possible in order to collect as much data as possible.

           DATA SOURCES: SURFACE DATA

There are several potential sources of meteorological data in BC, each having their own unique set of characteristics
(such as parameters, time averages, accuracies). Sources of such data are listed below. Specific information on the
available parameters for dispersion modelling and other unique features for each of the data sources can be found in
the following sections.


            ENVIRONMENT AND CLIMATE CHANGE CANADA (METEOROLOGICAL SERVICE OF
            CANADA: MSC) DATA

MSC has archives of meteorological data collected at reporting airports, climate stations, buoys, lighthouses, and
special studies. Typically, airport data provide most if not all of the required surface meteorological parameters for
AERMOD and CALPUFF. Data from non-airport stations can also be useful as they may be more representative of the
site conditions. If there are multiple stations, they can be used to better define the spatial variability of the
meteorological fields over the domain (important for CALMET applications).

MSC data30 are available online. Available hourly variables include element 076 (wind speed – km/s), element 156
(wind direction – 10s of degrees), element 078 (dry bulb temperature – 0.1 deg C), element 080 (relative humidity - %),
and element 077 (station pressure – 0.01 Kilopascals). Some stations are full 24-h reporting stations and others
operate only 12-16 hours/day.

The web available data do not provide all the necessary data, so for elements 071 (ceiling height – 30’s of m), 081
(total cloud opacity – tenths), 082 (total cloud amount – tenths), and elements 086 to 091 (present weather codes,
which can be used to determine precipitation codes), contact the regional MSC Climate Services Office at climatouest-
climatewest@ec.gc.ca. Request the data to be provided in a column oriented text file or Excel file.

MSC hourly airport observation data are primarily designed to generate long-term climate statistics and to satisfy
aviation weather requirements. In particular, it is important to note that:

           Wind speed is usually measured at 10 metres above the ground.

           The wind measurement may not be a true hourly average, but can be an observer estimated average over
            the last two minutes of every hour. It is important to determine what method is applied. If the latter, the
            quasi-instantaneous measure of wind speed/direction to predict the hourly average concentrations at a
            specific location increases the uncertainty of the individual hourly predictions.




30
     http://climate.weather.gc.ca/index_e.html


Version: July 2026                                                                                                Page 37
                              British Columbia Air Quality Dispersion Modelling Guideline



          Wind directions are reported to the nearest 10˚.

          The wind direction is assigned a zero when the wind speed is zero.

          There can be differences in wind instrumentation from these stations – for example, some stations use sonic
           anemometers (with a starting threshold of 0) and others use older instruments with an Anemometer Starting
           Threshold (AST) of 1 or 2 knots (approximately 1 m/s). Any wind less than the AST is reported as zero for
           speed and direction. If there are frequent periods when the wind speeds are less than the AST (which
           commonly occurs in valleys/basins), there will be an unrealistically high frequency of calms (zero wind speed).
           For MSC data, check the record to confirm the lowest non-zero wind speed in the data record as this will
           define the AST.

          The location of the airport may not adequately represent the meteorological conditions at the site under
           consideration. Note however that some parameters, such as cloud amounts, ceiling heights and precipitation,
           can have a broad spatial area of applicability, so observation of these parameters made at a distant airport
           could be applicable to the location of interest.


           BC MINISTRY OF ENVIRONMENT AND PARKS AND METRO VANCOUVER DATA

The Ministry downloads and archives hourly meteorological data at various stations across BC, including MV. The
Ministry operates some of the sites, and MV operates and maintains meteorological instrumentation throughout the
Lower Fraser Valley at air quality monitoring network stations. The rest of the sites are operated by industry under
permit to these agencies.

Wind and temperature are measured along with variables more suited for air quality analysis and dispersion
modelling, such as the vector wind speed and the standard deviation of the wind direction fluctuations (sigma theta,
σθ). In addition to wind and temperature, other observed parameters are relative humidity, barometric pressure,
precipitation, and incoming solar radiation.

For more information on how these parameters are calculated and their importance, see “Meteorological Monitoring
Guidance for Regulatory Modeling Applications” (US Environmental Protection Agency, 2000).

Although the data quality is maintained through a maintenance program and data quality checks, it is recommended
that information on the siting, instrumentation and data treatment are reviewed for appropriateness before using the
data. Data are available online for current and historical sites in BC from the BC Air Data Archive website31. To obtain
validated data from MV meteorological stations, an email request should be sent to: AQInfo@metrovancouver.org. While
MV data may also be available on the Ministry website (BC Air Data Archive website32), it should be emphasized that MV
station data available from this site may not be validated and/or up-to-date.

For Ministry and MV meteorological data there are a variety of different wind parameters available, so the Guideline
recommends the following:

          For all models listed in the Guideline, use the hourly scalar mean wind speed and wind direction (referred to
           as the scalar wind direction).




31
     http://envistaweb.env.gov.bc.ca/
32
     http://envistaweb.env.gov.bc.ca/


Version: July 2026                                                                                          Page 38
                             British Columbia Air Quality Dispersion Modelling Guideline



        Alternatively, a unit vector wind direction can be used instead of the hourly mean wind direction.

Note that anemometers at most Ministry and MV sites have a starting threshold of 1 km/hr (approximately 0.5 m/s).
However, due to changes in the database structure along with hardware changes, there have been some historical
changes to the way winds less than the AST have been reported. The following is recommended for light wind/calm
treatment of the data:

        For MV data, consult with Metro Vancouver.

        For Ministry data, wind speeds less than the AST should be set to zero, and wind directions should also be
         assigned as zero.

For BC wind data consider the following:

        Winds are typically measured at 10 m with the exception of a few sites where measurements are made on
         top of a building.

For MV wind data consider the following:

        Wind speed and direction measurements are made at various heights throughout the network. Anemometer
         heights range from 10 metres to 30 metres above the ground. Anemometer heights can be obtained for MV
         stations in a report titled: “Station Information: Lower Fraser Valley Air Quality Monitoring Network 33”
         (Metro Vancouver, 2012).


         OTHER SOURCES OF SURFACE (LAND AND MARINE) METEOROLOGICAL DATA

Other sources of data include the fire weather network program (Ministry of Forests, Lands and Natural Resource
Operations), the road maintenance and avalanche forecast program (Ministry of Transportation and Infrastructure),
university research programs, programs operated within national parks, marine observations (ships and buoy data)
and industry operated stations. In data-sparse areas, or areas where a network of meteorological data is needed to
characterize complex meteorological fields (i.e., for areas where CALPUFF is applied), the other sources of data can be
useful.

Instrument siting, type, maintenance and data formats (meta-data) are very important given that these stations are
set up for purposes other than air quality analysis. For example, many Ministry of Transportation and Infrastructure
meteorological stations are located along roadways, so wind directions may be influenced by trees that border the
roads. Other considerations include the parameter units of measurement (wind speed recorded in km/h whereas a
model requires m/s). Sometimes information on the siting, instrumentation, maintenance and data treatments may
be lacking, so care is required before using the data from these sources.

Meteorological data collected by several agencies (along with meta-data information) can be obtained from the
Pacific Climate Impact Consortium (PCIC) data portal34. If more recent data are not available from PCIC, contact the




33
   https://metrovancouver.org/services/air-quality-climate-action/Documents/lower-fraser-valley-air-quality-
monitoring-network-station-information-2012.pdf
34
   https://services.pacificclimate.org/met-data-portal-pcds/app/


Version: July 2026                                                                                        Page 39
                               British Columbia Air Quality Dispersion Modelling Guideline


listed agencies directly. Marine (ocean) observation data are available on-line from National Oceanic and Atmospheric
Administration’s (NOAA) National Data Buoy Center35.

If data from these sources are considered for modelling purposes, the following is recommended:

           Review siting to determine whether the data are influenced by obstacles or terrain features.

           Review instrument types for thresholds.

           Review maintenance and data quality assurance protocols.

           Review sampling frequencies.

           Determine the measurement units (e.g., m/s or km/h) and whether they are compatible with the
            measurement units required for the dispersion model.

           Assess the validity of the data for the purposes of modelling through the generation of wind roses and
            frequency of calms.

           If documentation on the data is not available, then apply tests to confirm quality and representativeness
            such as wind roses, time series plots and percentiles.

           DATA SOURCES: UPPER AIR DATA

Upper air data are required by both CALPUFF (depending on the CALMET mode) and AERMOD (morning sounding only). In
BC, the density of upper air stations is sparse, and the spatial applicability of the data for modelling dispersion at
lower levels (i.e., within a mountain valley) may not be applicable to the site unless the location of the sounding is in
the same valley and at approximately the same elevation. For more regional scale applications and/or where
dispersion is modelled at levels above the mountains (or above the boundary layer), a distant upper air sounding
could be applicable depending on the professional judgment of a meteorologist.

A list of upper air stations that could have applicability for BC modelling applications is provided in Table 5.1. The
upper air data are available online from the Integrated Global Radiosonde Archive (IGRA).36




35
     http://www.ndbc.noaa.gov/index.shtml
36
     https://www.ncei.noaa.gov/products/weather-balloon/integrated-global-radiosonde-archive


Version: July 2026                                                                                            Page 40
                             British Columbia Air Quality Dispersion Modelling Guideline


Table 5.1         Upper Air (Radiosonde) Stations in the Western/Pacific Region

                                       Station, Location              Latitude, Longitude (°)
                             Yakukat, Alaska                               59.51, -139.66
                             Whitehorse, Yukon                             60.70, -135.06
                             Fort Nelson, British Columbia                 58.83, -122.60
                             Edmonton-Stony Plain, Alberta                 53.53, -114.10
                             Prince George, British Columbia               53.90, -122.80
                             Kelowna, British Columbia                     49.97, -119.38
                             Spokane, Washington                           47.68, -117.63
                             Quillayute, Washington                        47.95, -127.36
                             Port Hardy, British Columbia                  50.68, -127.36
                             Annette Island, Alaska                        55.03, - 131.56



        LENGTH OF RECORD

For Level 2 or Level 3 Assessments, the following is recommended:

        Minimum period of data is one recent year with up to five recent and consecutive years preferred.

        At least three years of data should be used for the applications where there are significant public concerns
         about impacts of air quality.

        One year of data is acceptable if these data are obtained from a site specific meteorological collection
         program. If more data are available, they should be used.

        MISSING DATA

AERMOD (AERMET) and CALPUFF (CALMET) require an hour-by-hour sequential time series of meteorological input
data. AERMET skips missing data periods while CALMET will not run with gaps in the data series.

The recommended limit for data completeness is:

        90% (i.e., if more than 10% of data are missing over the simulation period, more data should be collected
         before running the model). If the data completeness recommendation of 90% cannot be met, the
         consequences of a lesser amount of data need to be outlined in the Dispersion Modelling Plan.

        The following is a guide to define “data completeness”:

             o    Lost data due to calibrations or other quality assurance procedures are considered missing data.

             o    The 90% recommendation applies on a monthly basis such that 12 consecutive months with 90%
                  recovery are required for an acceptable one-year database.

             o    The 90% recommendation applies to each of the variables required in a model, such as wind direction, wind
                  speed, and temperature.




Version: July 2026                                                                                              Page 41
                                British Columbia Air Quality Dispersion Modelling Guideline


                o    The 90% recommendation for temperature may be relaxed as model results are typically not
                     sensitive to variations in temperature.

                o    A variable is not missing if data from a backup, co-located sensor are available.

           If the data is greater than 90% complete, use the data filling procedures as described below for both
            AERMOD and CALPUFF.

Based on the Meteorological Monitoring Guidance for Regulatory Modeling Applications37 (US Environmental
Protection Agency, 2000), the following are recommended protocols to “fill” missing data gaps:

           If the missing period is four hours or less, use linear interpolation (do not use persistence – use of data from
            the previous time period (hour)).

                o    For transition periods (e.g., sunrise/sunset), use an appropriate interpolation technique depending
                     on the parameter.

           For periods greater than four hours, first determine if data from a nearby representative site are available to
            substitute for the missing hours. If such data are not available, use interpolation techniques based on how
            the parameter typically changes during this period.

                o    Use the criteria listed in Section 5.1: On-Site and Existing Alternate Sources of Meteorological Data
                     to determine if the station data are adequately representative.

           For data with large data gaps during a day (i.e., anything greater than 6 hours), replace the entire day with
            the previous valid full day.

                o    Data from a different station can be used to fill in gaps provided they are representative of the site.

           If there are multiple measurements at different levels on a tower, or upper air sounding data, then missing
            data for one level can be replaced by data from an alternative level. If the data are available, this method is
            the preferred approach to fill longer missing data periods.

Any data which are filled should be flagged to assist in the interpretation of model output and the uncertainty
associated with the concentrations predicted during periods of substituted data. The greater the missing data period,
the greater the potential for error in the filled data.

Other data filling approaches can be used but introduce greater uncertainty in the predictions. The following methods
should be detailed in the Dispersion Modelling Plan submitted prior to modelling. These methods include:

           substitution from measurements at nearby locations with dissimilar meteorological characteristics.

           substitution of NWP output, subject to the validity checks in Section 6.1: NWP Model Output for Dispersion
            Modelling. Ensure that the time period of the replacement data is the same as the missing data, and also the
            user should adjust elevation levels between the replacement station and the missing data station.

           LIGHT AND NO WIND CONDITIONS

The term “calm” is defined as a period of time when the wind speed is recorded as zero. Zero values can be due to:




37
     https://www.epa.gov/sites/default/files/2020-10/documents/mmgrma_0.pdf


Version: July 2026                                                                                              Page 42
                                British Columbia Air Quality Dispersion Modelling Guideline



           a “meteorological” calm which refers to the condition of no movement of air (no wind).

           an “instrument” calm, where there may be wind, but it is below the AST.

It is important to review the hourly data to understand what the lowest wind speed recorded (the low wind speed cut off),
and the frequency of calms. For MSC, Ministry and MV wind data, the calm treatment is described in Section 5.2: Data
Sources: Surface Data.

While CALPUFF can explicitly handle calms, AERMOD does not. AERMOD will skip the period when wind speeds are less
than approximately 0.28 m/s and will not calculate concentrations. While this is an acceptable treatment for an occasional
calm, it is inappropriate if calms occur frequently and/or last for long durations (a common occurrence in BC valleys during
the winter). AERMOD will not calculate a 24 h average (e.g., for PM2.5) if there are more than 8 hrs in a 24 h period that are
skipped due to calms. Since it is common for the highest 24 h average concentrations to occur under light wind conditions,
the highest 24 h averages may be missed completely as a result of the calms treatment.

Given this, determine the frequency and duration of winds less than the threshold of 0.28 m/s. Depending on the results,
CALPUFF may be the only appropriate model to use.

           NWP MODEL OUTPUT

Recommendations on the use of NWP model output for dispersion modelling and the availability of such output for BC are
discussed in Section 6.1: NWP Model Output for Dispersion Modelling and Section 6.2: BC WRF Data Availability. Critical
to the decision on the use of the output for these purposes are tests to confirm the validity of the data for the location of
interest.

           METEOROLOGICAL DATA FOR AERMET (AERMOD’S PRE -PROCESSOR)


            MINIMUM METEOROLOGICAL DATA REQUIREMENTS

AERMET requires hourly surface meteorological data observed at an airport or a site specific collection program and
the morning upper air sounding from a nearby (or representative) upper air station.

The following are recommendations on the minimum data requirements for AERMET from the User`s Guide for the
AERMOD Meteorological Preprocessor (AERMET)38 (US Environmental Protection Agency, 2024):

           Data must be horizontally and vertically representative - judged independently for each variable.

           Surface characteristics around the meteorological site must be similar to the surface characteristics in the
            modelling domain.

           Surface characteristics around the meteorological site should be input.

Adequately representative data for each of the following variables constitute the minimum set of meteorological
variables that AERMOD requires:

           wind speed (scalar mean, generally at a height of 10 m).




38
     https://gaftp.epa.gov/Air/aqmg/SCRAM/models/met/aermet/aermet_userguide.pdf


Version: July 2026                                                                                               Page 43
                             British Columbia Air Quality Dispersion Modelling Guideline



        wind direction (scalar mean or unit vector, generally at a height of 10 m).

        ambient temperature (generally at a height of 2 m).

        cloud cover – use opaque cloud cover first; use total cloud amount if cloud opacity is unavailable.

        morning upper air sounding (12 UTC in TD-6201 or original FSL format).

If gaseous and particulate deposition (wet, dry) is to be modelled, in addition to the above, AERMET requires station
pressure and hourly precipitation (amount) data.

Data from a site specific meteorological program can be used by AERMET to provide meteorology in situations where
there are no representative airport data. However since cloud cover is a required input, airport data must be used as
typically observers estimate cloud cover. If there are no representative cloud cover data, an equivalent cloud cover
can be calculated by AERMET, using the Bulk Richardson Number approach (select the BULKRN option) if temperature
measurements at two levels and wind speed at one level are available from a site specific measurement program.

In addition, given the paucity of upper air stations and the complex terrain in BC, the nearest upper air sounding will
likely not be representative of the location of interest especially near the surface. In such situations, first test the
sensitivity of the modelled maximum concentrations using soundings from different locations. If the results are
sensitive to this input, one of the following methods is recommended:

        Use the sounding that results in the highest maximum ground level concentrations.

        NWP model output can be mined (subject to the conditions of use in Section 6.1: NWP Model Output for
         Dispersion Modelling) to create a morning upper air sounding by extracting the vertical profiles of wind and
         temperatures at the location of interest.

        Use commercial software available to estimate the upper air sounding for AERMET; however these
         estimation methods assume flat terrain, so the application to BC is limited.


         CALMS TREATMENT

As discussed in Section 5.6, a calm (where the wind speed = 0.0) can be a true calm (i.e., no wind movement) or instrument
induced where the wind speed is less than AST. With respect to calms, AERMET is able to handle wind speeds ≥ 0.28 m/s
(21/2 * σvmin, where σvmin = 0.2 m/s). For wind speeds less than the lesser of the AST and 0.28, AERMET skips the hours and
flags those multiple-hour average concentrations that include a model calm period in the output file.

For data from a site specific meteorological program, the user must specify the AST. AERMET assigns the wind speed
as calm (0.0 m/s) in the meteorological file if the wind speeds are less than a user-specified AST. Any wind speeds less
than approximately 0.28 m/s will be treated as calm (even if the AST is lower than this value). Note that AERMET does
not allow an AST > 1.0 m/s when data from a site specific meteorology program are used.

For airport data obtained from MSC (see Section 5.2.1: Environment and Climate Change Canada (Meteorological Service
of Canada: MSC) Data), AERMET does not require the AST to process the SURFACE pathway as any wind speeds less than
the AST would be already be recorded as zero. As previously discussed, AERMET will skip these calm hours.

Finally, AERMOD includes several options related to calms treatment to address concerns about model performance under
low wind speed conditions. These provide the user flexibility to change the minimum values of σvmin, as well as the low wind
speed\stable friction velocity (ADJ_U*) and meander factor (FRAN). For these parameters, the Guideline recommends the
default values (i.e., do not enable these options).



Version: July 2026                                                                                             Page 44
                             British Columbia Air Quality Dispersion Modelling Guideline



         METEOROLOGICAL DATA ACQUISITION AND PREPARATION


5.8.3.1 REFORMATTING HOURLY SURFACE DATA INTO AERMET READY FILES

Data from a site specific meteorological program can be used by selecting the ONSITE pathway in AERMET. Any ASCII
text files (e.g., .txt, and .csv files) of hourly sequential data can be read – as long as the order of the parameters and
their corresponding formats are specified.

If hourly site specific data are in an Excel spreadsheet, the file should be saved as a text file which contains numeric
values only. The variables in the site specific hourly data must have the following units:

        wind direction (degrees from north)

        wind speed (m/s)

        dry bulb temperature (o Celsius)

        sky cover/cloud cover (tens of percent)

        station pressure (millibars*10)

        precipitation amount (centimeters)

        relative humidity (percent)

Data from hourly airport data obtained from MSC can be processed by AERMET by selecting the SURFACE pathway.
However, these data must first be converted to a format readable by AERMET, such as CD-144, SCRAM, or SAMSON.
Care is required to ensure that the proper units associated with each of these formats are assigned (e.g., the wind
speeds in CD-144 data are in knots).


5.8.3.2 WIND DIRECTION TREATMENT FOR MSC HOURLY AIRPORT DATA

Wind directions in MSC data are in 10’s of degrees. For use in AERMOD, the secondary keyword RANDOM should be
selected so that AERMET will randomize the wind direction within a 10 ° sector.


5.8.3.3 UPPER AIR DATA
The morning upper air sounding data for processing by AERMET can be in a format supported by the version of
AERMET being used, including TD-6201, original FSL, or IGRA format. For AERMOD applications within BC, the
required morning sounding corresponds to the 12 UTC sounding.
Original FSL-format upper air soundings are no longer available for new downloads from the former NOAA/ESRL
Radiosonde Database. For new downloads, users should obtain upper air sounding data from the NOAA/NCEI
Integrated Global Radiosonde Archive (IGRA) and specify the IGRA format in the UPPER AIR pathway of AERMOD.
Existing archived TD-6201 or original FSL files may still be used where they are available and can be processed
successfully by the selected AERMET version.
If there are no appropriate upper air soundings available for the modelling domain, extracting upper air soundings
from NWP model output (subject to the conditions on the use of NWP output listed in Section 6.1: NWP Model
Output for Dispersion Modelling) is an alternative.




Version: July 2026                                                                                            Page 45
                              British Columbia Air Quality Dispersion Modelling Guideline



         METEOROLOGICAL DATA FOR CALMET (CALPUFF’S PRE -PROCESSOR)


          SURFACE METEOROLOGICAL DATA PREPARATION FOR CALMET

SMERGE is the primary pre-processing utility for creating a CALMET-ready, surface data file (called SURF.DAT). It
assumes that the data have already been validated through quality assurance checks by using a utility called
METSCAN (for CD-144 formatted data) or some other equivalent method. Typically data obtained from MSC, the
Ministry or MV have already been validated, thus avoiding the need to run METSCAN (and the need to reformat the
data into CD-144 format in order to run it).

For validated data, SMERGE can read data in a variety of U.S. data formats (including CD-144), as well as a generic
comma delimited (.csv) format. This latter format is most convenient for BC applications as hourly data obtained from
government agencies are normally stored in a text file (.txt) or spreadsheet. A spreadsheet can easily be manipulated
to SMERGE requirements (described as follows) then saved in the .csv format – ready for processing by SMERGE. The
instructions are as follows:

    1.    Place each parameter listed below into a specific column, with the proper units as listed as follows:

              Column 1 – month (e.g., 1, 2, ... , 12)
              Column 2 – day (e.g., 1, 2, 3, ... , 31)
              Column 3 – year (4 digits, e.g., 2010)
              Column 4 – hour (multiplied by 100, e.g., 0, 100, 200, ... , 2300)
              Column 5 – temperature (Celsius)
              Column 6 – precipitation amount (mm)
              Column 7 – pressure (mb)
              Column 8 – relative humidity (%)
              Column 9 – wind direction (degrees)
              Column 10 – wind speed (m/s)
              Column 11 – cloud cover (tenths)
              Column 12 – ceiling height (hundreds of feet)

    2.    A missing value of a real variable must be replaced with 9999.0 and for an integer variable: 9999 with the
          exception of the station ID, month, day, year, and hour.

    3.    No blank cells are allowed.

    4.    Ceiling height and cloud cover are only available from MSC. These parameters can be combined with other
          parameters from an on-site measurement program if they apply to the site under consideration.

    5.    For ceiling height, MSC uses 888 for unlimited ceiling height whenever cloud opacity is less than 6/10ths (see
          Env Canada, 2013). It can be used as is.

Table 5.2 is a sample spreadsheet which is used to prepare meteorological data input for SMERGE and reflects the
data and formats as described in the previous steps 1-5. In this example, the Station ID=1432. The following is based
on TRC (2011b).




Version: July 2026                                                                                           Page 46
                             British Columbia Air Quality Dispersion Modelling Guideline


Table 5.2         A Sample of a Spreadsheet Used to Prepared Data for SMERGE

                                                                                                    Cloud     Ceiling
  Station ID       =    1432 Temp        Precip    Pressure      RH     Wind Dir Wind Speed
                                                                                                    Cover     Height
                                                                                                              100’s of
  Month Day Year Hour            ̊C       mm          Mb         %         deg         ms-1         tenths
                                                                                                                feet
     8      1    2010     0     10.8     9999.0     961.4        89        177          0.2           3         888
     8      1    2010   100     9.5      9999.0     961.8        95        334          0.5           0         888
     8      1    2010   200     7.4      9999.0     961.9        98        211          0.4           0         888
     8      1    2010   300     7.3      9999.0     961.9       100        324          0.2           3         888
     8      1    2010   400     6.1      9999.0     961.9       100        323          0.1           3         888
     8      1    2010   500     5.9      9999.0     961.8       100        153          0.6           4         888
     8      1    2010   600     6.4      9999.0     961.8       100        157          1.6           5         888
     8      1    2010   700     8.5      9999.0     961.6        94        144          1.2           6          20
     8      1    2010   800     11.4     9999.0     961.5        82        132          0.8           8          50
     8      1    2010   900     13.2     9999.0     960.7        74        172          0.7           9          40
     8      1    2010   1000    15.1     9999.0     9999.0     9999       9999         9999.0       9999       9999

Once the spreadsheet is arranged, save it in a comma delimited file (.csv file) as in the example shown in Table 5.3.
No blank spaces are allowed between the commas.

The file must contain the first three lines as shown in Table 5.3 exactly as shown with the exception of the Station ID
number. In this example, it is 1432, but could be any integer number.

Table 5.3         Sample of a Comma Delimited Input File Ready for SMERGE

GENERIC,Version,'2.0',Manually generated,Time as ending hour
Station,ID,=',1432,Temp,Precip,Pressure,RH,Wdir10m,Wspeed10m,Ccover,Cheight
Month,Day,Year,Hour,DegC,mm,mb,%,deg,ms-1,tenths,hundreds_of_feet
8,1,2010,0,10.8,9999.00,961.4,89,177,0.2,3,888
8,1,2010,100,9.5,9999.00,961.8,95,334,0.5,0,888
8,1,2010,200,7.4,9999.00,961.9,98,211,0.4,0,888
8,1,2010,300,7.3,9999.00,961.9,100,324,0.2,3,888
8,1,2010,400,6.1,9999.00,961.9,100,323,0.1,3,888
8,1,2010,500,5.9,9999.00,961.8,100,153,0.6,4,888
8,1,2010,600,6.4,9999.00,961.8,100,157,1.6,5,888
8,1,2010,700,8.5,9999.00,961.6,94,144,1.2,5,888
8,1,2010,800,11.4,9999.00,961.5,82,132,0.8,6,20
8,1,2010,900,13.2,9999.00,960.7,74,172,0.7,8,50
8,1,2010,1000,15.1,9999.00,960.1,72,53,0.9,9,40
8,1,2010,1100,16.4,9999.00,9999.00,9999,9999,9999.00,9999,9999
8,1,2010,1200,19.1,9999.00,959.1,53,80,1.4,8,80
8,1,2010,1300,19.8,9999.00,958.8,49,337,1.6,6,100
8,1,2010,1400,21.8,9999.00,958.1,48,336,2.1,8,80




Version: July 2026                                                                                           Page 47
                             British Columbia Air Quality Dispersion Modelling Guideline


For the merging of multiple surface data files specify each file name, station identifier (i.e., Station ID) and
corresponding time zone in the SMERGE.INP file see the CALPUFF Modeling System Version 6 User Instructions 39
(Exponent, 2011) . SMERGE will combine these individual files into a singular SURF.DAT file - ready for input to
CALMET.

For missing data, when multiple surface stations are selected for CALMET, the only requirement is that a valid value of
each parameter is available for each hour from at least one of those stations. The exception to this is the precipitation
type code which can be missing from all of stations. For the other parameters CALMET will automatically substitute a
missing parameter from another surface station (if it exists for that period). However, if such a substitution is
inappropriate, or there are no other surface data available for that missing period then data filling methods can be
used depending on the parameter and the extent of the missing record. SMERGE will generate a list of missing data,
which can be further examined to determine how these periods are to be treated (see Section 5.5: Missing Data).


         UPPER AIR DATA PREPARATION

Data for the twice-daily soundings released at 00 and 12 UTC for one or more stations are required in a format that
can be processed into CALMET-ready UP.DAT format. READ62 is the upper air data preprocessor that produces a
formatted UP.DAT file for use by CALMET and conducts a number of quality assurance checks (missing data, out of
range values, etc.). These missing values can be eliminated or if retained there are options on how to treat missing
data. Users are referred to the CALMET User’s Guide40 and the CALPUFF FAQ’s41 for different upper air data filling
methods.

As mentioned in Section 5.3, original FSL-format upper air data are no longer available for new downloads from the
former NOAA/ESRL Radiosonde Database. Existing archived original FSL files may still be used where they are available
and can be processed successfully by READ62.

Where compatible, archived original FSL-format upper air data are unavailable, users should apply a documented and
quality-assured method to generate CALMET-ready UP.DAT files from NOAA/NCEI IGRA-format data. If suitable upper
air soundings are not available for the modelling domain, users should consider using NWP model output, subject to
the conditions described in Section 6.1: NWP Model Output for Dispersion Modelling. In all cases, users should
document the data source, processing method, and QA/QC procedures and retain this information for potential
review by the Ministry.


         PRECIPITATION CODE AND HOURLY PRECIPITATION FOR LONG -TERM, WET
         DEPOSITION CALCULATIONS

In order to model wet deposition, CALMET requires hourly precipitation and precipitation code (i.e., frozen vs. liquid
precipitation). However, the precipitation code is not included in the surface data available from MSC, the Ministry or
MV. Fortunately, if the surface data are processed in a *.csv format by SMERGE (as per Section 5.9.1), the
precipitation code is calculated automatically based on hourly precipitation and dry bulb temperature and then is
written into the list of parameters that comprise the SURF.DAT file.




39
   https://www.calpuff.org/calpuff/download/CALPUFF_Version6_UserInstructions.pdf
40
   https://www.calpuff.org/calpuff/download/CALMET_UsersGuide.pdf
41
   https://calpuff.org/calpuff/FAQ-answers.htm


Version: July 2026                                                                                          Page 48
                              British Columbia Air Quality Dispersion Modelling Guideline


MSC data can include hourly precipitation, although some stations only report daily precipitation amounts. In this
case, the hourly weather code can be used (elements 086 to 098 that indicate rain, snow, drizzle, etc.) to determine
the hours when precipitation occurs. Partitioning of the total daily precipitation equally over the precipitation hours is
acceptable given that calculated depositions are long term accumulations (i.e., annual wet-deposition).

For wet deposition calculations, hourly precipitation data are required in a separate PRECIP.DAT file. Pre-processors,
PXEXTRACT (for quality assurance of precipitation data) and PMERGE (to merge data from multiple stations, account
for stations in different time zones, etc.), are available to create a PRECIP.DAT file if the hourly precipitation data are
in U.S. NWS TD-3240 format. Unfortunately, this format is not used by Canadian agencies.

For preparing Canadian precipitation data, users can select the free-formatted option by creating the PRECIP.DAT file
directly in free-format as described in Section 8 of the CALPUFF Modeling System Version 6 User Instructions42.

Note that the use of precipitation data collected at airports may underestimate the orographic enhancement of
precipitation at higher elevations, or the precipitation data may not be representative of the domain. In such cases, if
there is no PRECIP.DAT file, the precipitation fields provided by the NWP model output (when used either in Hybrid or
No-Obs mode) can be used by CALPUFF for wet deposition calculations.




42
     https://www.calpuff.org/calpuff/download/CALPUFF_Version6_UserInstructions.pdf


Version: July 2026                                                                                             Page 49
                             British Columbia Air Quality Dispersion Modelling Guideline



     METEOROLOGICAL MODELLING

        NWP MODEL OUTPUT FOR DISPERSION MODELLING

NWP models include sophisticated physics and produce hourly forecasts of meteorological parameters at horizontal
grid resolutions in the order of a kilometre and in multiple vertical levels over a large domain. This ability means that
NWP models can produce meteorological information in areas where no observations exist – an attractive benefit for
air quality assessments in BC, given that the meteorological monitoring network in BC is sparse with limited spatial
applicability due to the complex terrain.

NWP model output can be used for dispersion modelling purposes, either to supplement observational data or as a
substitute for observations depending on the Assessment Level and the dispersion model (see Section 2). However,
whether the output data from an NWP model can be used for dispersion modelling depends on the applicability and
quality of the output and as such it must undergo a QA/QC process described in this Section.

Subject to the considerations and QA/QC tests, the following is recommended:

        For AERMOD, NWP output can be used to extract surface meteorological variables and vertical profiles in
         situations where there is no meteorological data. However, the priority should always be to obtain
         representative meteorological data rather than relying on NWP model output. Section 6.3 provides guidance
         on extracting AERMET-ready files.

        For CALPUFF, NWP output can be used in several ways, depending on the application (and the corresponding
         CALMET mode) as per Section 6.4.1.

Since air quality assessments involve dispersion models that use historical meteorological data, NWP models can be
run for periods in the past to produce historical prediction (hindcast) output that is pieced together to create a
continuous historical, model-produced meteorological record. This hindcast approach can be taken a step further,
where actual observations can be used to “nudge” the model toward what was actually observed at these locations,
and in doing so improve the rest of the meteorological fields.

NWP models offer an alternative to meteorological measurements as input for dispersion models such as AERMOD
and CALPUFF. This approach offers many advantages:

        provides data for any location (important in regions where there are no surface and/or upper air data).

        takes less time in hindcast mode to generate a data set than a monitoring program (in forecast mode it takes
         the same time).

        avoids subjective decisions regarding the applicability of meteorological data collected in a different location
         (for example, adjusting/rotating winds to account for different terrain orientations).

        provides details of the space and time variability of the meteorology in three dimensions within a modelling
         domain, of critical importance for BC given the complex geophysical setting.

While NWP models offer many advantages, they require computational resources (processing, storage) that escalate
rapidly with domain size, grid resolution and simulation duration. Therefore, it is impractical to apply these models for
dispersion modelling in situations where it is required to simulate one or multiple years of hourly meteorology at a
fine enough grid scale (on the order of 100 m) and vertical levels in order to resolve the effects of small-scale terrain
features - a common need in BC.



Version: July 2026                                                                                           Page 50
                             British Columbia Air Quality Dispersion Modelling Guideline


In order to take advantage of their sophisticated physics in situations which their application would be limited (i.e.,
fine grid resolutions over long period simulations), the relatively coarse spatial resolution NWP model output can be
used as an initial guess field for the fine spatial resolution diagnostic model, CALMET. The diagnostic model further
applies dynamically consistent diagnostic algorithms based on terrain effects and observations to develop
meteorological adjustments on a fine scale. This hybrid approach uses the strengths of both models and makes it
practical to produce fine-scale meteorological fields for long period simulations.

Note that there is a spatial limit to the terrain effects and weather phenomena that NWP model can resolve. NWP
models can completely resolve phenomena with spatial scales greater than 7 to 10 times the grid resolution. Anything
less will be only partly resolved, and at 2 times the grid resolution such effects will be filtered out completely. Typical
grid resolutions for fine scale NWP models are 4 km, which means that the flow features that affect the pollution
transport in BC for many local-scale applications will only be partly accounted for at best and may be completely
ignored. For this reason, combining the coarse scale NWP output with a fine scale model such as CALMET is the only
way to resolve these small scale terrain flow features.

“Research on the Applicability of Modelled Site Specific Meteorological Data to Well Test Flaring Assessments in BC”
(Levelton Consultants Ltd., 2002) concluded that the hybrid approach “significantly improves resolution of the local
scale wind system in complex terrain.” Furthermore, the study found that this combination of models is an objective
means to produce meteorological data at a site for straight-line, Gaussian models that are at least as representative
for that site as applying and adapting data from other locations.

NWP output can be used in ways other than the hybrid approach as CALMET has the ability to use NWP output as the
only source of input meteorology. Guidance on the recommended application of these various modes is provided in
Section 6.4.1.

Although NWP models offer many advantages, there are other factors to consider:

        They require considerable expertise and computational resources to run.

        The boundary layer has traditionally been a challenge to simulate for these models especially in geophysically
         complex settings.

        The limitations of grid resolution mean that terrain effects and weather phenomena with small spatial scales
         may only be partly resolved or completely ignored.

        There are several NWP models that can be configured and initialized in different ways all providing
         potentially different results (McEwan and Murphy, 2004).

        The installation of an on-site meteorological station(s) may still be required given the ability of CALMET to be
         used in Hybrid mode to resolve small scale terrain flow features.

Given that NWP models are evolving and experience on their use for dispersion modelling is relatively recent, NWP
model output for dispersion modelling purposes can be applied only if the use of a particular NWP model and its
modelling configuration has been carefully determined based on an initial analysis of the simulations, especially
within the boundary layer. This would involve model testing in the domain of interest to determine whether it
produces realistic meteorological fields in a qualitative sense, as well as comparing model output to observations
using quantifiable performance measures such as those described in “Use of High Resolution Numerical
Meteorological Fields with the CALPUFF Modelling System: An Analysis of RAMS and MC2 Fields over Kamloops B.C.”
(McEwan & Murphy, 2004). These include an assessment of:




Version: July 2026                                                                                            Page 51
                            British Columbia Air Quality Dispersion Modelling Guideline



       wind roses (seasonal and annual) produced at key locations
            o    Do they make qualitative sense?
            o    How well do they compare to windroses based on actual data?
       thermally generated flows and diurnal variations
            o    Do the meteorological fields under clear sky conditions reflect the expected day and night flow fields
                 for the terrain situations?
            o    Do the meteorological fields in the afternoon of the spring and summer months reflect a wind that
                 blows from the ocean inland?
            o    Do the meteorological fields at night of the summer months reflect a wind that blows from the land
                 to the ocean?
       vertical profiles of wind and temperature (and diurnal variations)
            o    Do these vertical profiles make qualitative sense, especially as the temperature profile varies
                 diurnally?
            o    How well do they compare to nearby upper air soundings?
       the different levels of wind fields over the domain (and diurnal variations)
            o    Is there evidence of unrealistic wind jets at low elevations (this has occurred in some NWP output)?
       the boundaries of the NWP domain
            o    Is the domain of interest close to the edge of the NWP output domain (i.e., within 2 or 3 NWP grids)
                 where unrealistic boundary effects occur?
NWP model output may not be necessary or may offer marginal benefit in the following situations:

       There are relatively flat areas with representative upper air soundings and a network of surface observing
        stations.

       The modelled sources are near the surface (<50 m) and the domain is limited (on the order of a few km), the
        surface observing station(s) in the domain may provide data representative of the conditions at the location
        of interest.




Version: July 2026                                                                                         Page 52
                            British Columbia Air Quality Dispersion Modelling Guideline



         BC WRF DATA AVAILABILITY

WRF model output for every hour from 2019 to 2023, covering all of BC and surrounding areas, was produced by
Weatherlogics and the University of British Columbia under contract. This output, produced using the Weather
Research and Forecasting (WRF) model at 4 km grid resolution with Four Dimensional Data Assimilation performed,
provides a quality controlled, consistent source of NWP model output suitable for use in dispersion modelling
applications.

In addition to the 4 km province-wide dataset, high-resolution WRF model output at 1.33 km grid spacing has been
developed for selected areas of BC, including Kitimat, Prince George, Trail, and Metro Vancouver. This high-resolution
WRF output provides more detailed meteorological fields in areas where finer-resolution NWP output may better
represent local and regional terrain-influenced flows.

WRF output in CALMET-ready format (3D.DAT) can be extracted for specific domains. The Ministry is also developing a
web application to allow WRF output in netCDF-format to be extracted for user-specified locations.

To provide assurance that the WRF output can be used for dispersion modelling purposes, the output has been
quantitatively evaluated through comparisons with observations at MV and the Ministry stations, as well as data from
NAV CANADA and Environment and Climate Change Canada. The WRF output has also been qualitatively evaluated
for its ability to reproduce specific phenomena, including thermally generated flows and diurnal variations; hourly
time series of wind speed, direction, temperature, and humidity; vertical profiles of wind and temperature; and wind
fields at different levels over the WRF domains.

The WRF model output is available here.43

         MESOSCALE MODEL INTERFACE PROGRAM ( MMIF): PROCESSING NWP OUTPUT FOR
         DISPERSION MODELS

User’s Manual for the Mesoscale Model Interface Program (MMIF) 44 facilitates the use of NWP output for air quality
modelling applications. The current version available from the U.S. EPA is MMIF Version 4.1.1.

MMIF can convert output fields from an NWP model (the Fifth Generation Mesoscale Model [MM5], version 3 and
WRF-ARW, version 2 through 4) to parameters and formats suitable for input to AERMET. MMIF can also replace
AERMET by producing output data which can be input directly into AERMOD.

MMIF is able to act as an alternative to CALMET to process NWP model output directly to a suitable format for
CALPUFF. However, if MMIF is used as a replacement for CALMET, the addition of terrain effects that CALMET can
apply to produce a final wind field is by-passed.

MMIF does not perform any coordinate transformation. Therefore, CALPUFF must be run in the same coordinate
system and datum as the NWP model output. In addition, MMIF does not provide output with a grid resolution finer
than the one of the NWP output, and MMIF cannot process additional observation data as well. All of these factors
limit the use of MMIF for CALPUFF.




43
     https://www2.gov.bc.ca/gov/content?id=6C49603C7EAA4050979CF908162EA91D
44
     https://gaftp.epa.gov/Air/aqmg/SCRAM/models/related/mmif/MMIFv4.1.1_Users_Guide.pdf


Version: July 2026                                                                                        Page 53
                             British Columbia Air Quality Dispersion Modelling Guideline


Although MMIF is still under testing and development, the following is recommended:

         The use of MMIF should be identified in the Dispersion Modelling Plan.

         MMIF can be used to extract meteorological output suitable for AERMOD applications in situations where
          there are no meteorological data and terrain within a modelling domain is flat. However, the priority should
          always be to obtain representative meteorological data rather than fully relying on NWP model output (see
          Section 6.1).

         Given CALMET’s current ability to ingest and process NWP output in No-Obs mode and its ability to apply
          further terrain effects, the use of MMIF as a replacement for CALMET is not recommended until further
          experience with MMIF demonstrates its superiority.

         CALMET METEOROLOGICAL MODELLING

CALMET consists of a diagnostic wind field module and micrometeorological modules for overwater and overland
boundary layers (Scire, Strimaitis, & Yamartino, A User’s Guide for the CALPUFF Dispersion Model (Version 5), 2000). A
two-step approach can be used to produce wind fields where the Step 1 fields are created from an Initial Guess field
that is adjusted for the kinematic effects of terrain, slope flows, and terrain blocking effects. In Step 2 an objective
analysis procedure blends observational data (if available) into the Step 1 field in order to produce a final Step 2 wind
field.


          CALMET MODELLING MODES

CALMET can be run in 3 different ways as follows:

    1.    No Observation (No-Obs): CALMET relies entirely on NWP model output

    2.    Observation Only (Obs-Only): CALMET relies entirely on observation data

    3.    Hybrid: CALMET uses observations and NWP model output

The selection of a particular mode depends on the recommendations provided in the following sections.


6.4.1.1 NO OBSERVATION (NO-OBS)

In this mode, output from an NWP model is CALMET’s only source of meteorological data. For CALPUFF applications,
the critical considerations should this mode be selected are that the key features of the complex flow that define the
dispersion and transport can be resolved by the grid resolution of the NWP model output. The use of No-Obs mode is
recommended for:

         Level 2 Assessments.

         Level 3 Assessments only if there are no appropriate surface observations to run in the Hybrid mode.

         The use of NWP model output is subject to the tests/conditions described in Section 6.1.

         The NWP model output is introduced as an Initial Guess field.

Regarding the last recommendation, CALMET’s grid resolution is typically finer than the grid resolution of the NWP
model. This finer resolution is required in order to capture the relevant complex flow features of the situation. In this



Version: July 2026                                                                                           Page 54
                             British Columbia Air Quality Dispersion Modelling Guideline


case, introducing the NWP model output as an Initial Guess field takes advantage of CALMET’s ability to apply the
finer scale terrain effects (kinematic, slope, and blocking) to create a more realistic flow field.


6.4.1.2 OBSERVATION ONLY (OBS-ONLY)

Obs-Only mode works best when available representative surface and upper air data are close to a facility and the
expected area of impact is within a few to several kilometres. For complex flow situations, relying fully on
observational data as input to CALMET would require a network of surface and upper air stations to adequately
simulate the meteorological fields. Given that such networks are rare in BC, the use of this mode using actual
meteorological observations would be limited to:

       Simple (i.e., non-complex) flow where available observational data have a wide spatial applicability.

       Level 2 Assessments.


6.4.1.3 HYBRID

Hybrid mode combines three-dimensional NWP output with surface (or both surface and upper air station) data and
optional data such as overwater buoy data and precipitation data. Coarse grid resolution NWP model output provides
a spatially and temporally varying meteorological field. The NWP output is interpolated to the CALMET grid with a grid
resolution finer than the NWP output’s (e.g., 4-km grid spacing), and CALMET uses fine-resolution terrain data and
observations to make adjustments to produce a final Step 2 wind filed. The Hybrid approach uses the strength of both
NWP and CALMET model to produce at least as good as or often better results than the use of observations alone.

Often appropriate surface station data are available, but due to the paucity of upper air stations in BC, appropriate
upper air observations may be limited. As such this mode allows the use of surface station observations in
combination of NWP output to “fill in” the upper air information.

For CALPUFF applications, this mode is recommended for the following situations:

       Level 3 Assessments.

       Appropriate observational data (surface data at a minimum with upper air data if available).

       “Complex” flow situations.

       The NWP model output is subject to the quality checks described in Section 6.1.

       The NWP model output is introduced as the Initial Guess field.

When upper and surface observations are used in CALMET, there are seven user defined parameters that require
expert judgment: BIAS, IEXTRP, RMAX1, RMAX2, TERRAD, R1, and R2. Due to their importance in creating a realistic
CALMET wind field, based on TRC (2011), explanation and guidance on these parameters are provided in Table 6.1.

One common problem associated with Hybrid mode is “doughnut” patterns that occur around surface observation
stations under low wind speed conditions. This may point to a problem with the NWP model output at that specific
location or an issue with relatively high anemometer starting thresholds where low wind speeds are reported as zero.
In these situations a difference in the wind speed at the meteorological station vs the surrounding area as defined by
the NWP model output appears as a “doughnut” pattern.

Eliminating or minimizing the spatial extent of this artifact can be done through the following:


Version: July 2026                                                                                          Page 55
                               British Columbia Air Quality Dispersion Modelling Guideline



          Review the user defined values: R1 and RMAX1 (increase or reduce; see Table 6.1).

          Review the NWP model output at the specific location for reasonableness (see Section 6.1).


           DOMAIN AND GRID RESOLUTION

The CALMET domain depends in part on the area where the predicted concentrations/depositions are needed. Given
this, select the CALMET domain that will include all terrain affecting the flows that determine the dispersion within
the area of interest (i.e., CALPUFF domain). In addition, if meteorological data outside of the area of interest can help
better define the CALMET generated meteorological fields within the area of interest, then extend the meteorological
domain to include these observations.

The choice of grid resolution depends on the whether the resolution will allow CALMET to capture the important
features of the flows that govern the behaviour of the dispersion. For example, modelling the dispersion of a plume in
a narrow, winding valley that is 2 km wide (a common situation in BC) would require a CALMET grid resolution on the
order of 250 m. This is a common grid resolution for local scale applications in BC. For regional scale applications
where dispersion on the order of 100 km is of interest, then a coarse grid resolution (i.e., 1 km) may be appropriate.

In all cases, selecting the appropriate grid resolution must be balanced by the need to resolve the important features
of the flow fields and the computer resources required to generate the winds for every grid point. Given the
professional judgement involved, it is critical that the grid resolution and domain size are identified in the Dispersion
Modelling plan and discussed with the Ministry.


           GUIDANCE ON KEY CALMET MODEL OPTIONS

Due to the number of user-defined variables and options, many of which require expert judgment, guidance on the
switches/options and user defined factors for CALMET (Version: 6.4.0, Level: 121203) Input Group 4 and 5 are
provided in Tables 6.2. The switches/user defined parameters are shaded according to the following:

       Black = do not touch (if changed, provide justification)
       Dark Grey = recommended default
       Light Grey = expert judgment required
       White = not used

This guidance is based on the collective experience of practitioners in BC, CALPUFF Modeling System Version 6 User
Instructions45, the CALPUFF FAQ’s46, and the New South Wales CALPUFF guidelines (New South Wales Department of
Environment and Conservation, 2005). Although these Tables are intended to be as specific as possible, expert
judgment is required as every CALMET application is unique. For these reasons, it is recommended that CALPUFF
users:

          Are familiar with the above documents,

          Have attend a CALPUFF basic and advanced training course, and

          Have a background in boundary layer meteorology.




45
     https://www.calpuff.org/calpuff/download/CALPUFF_Version6_UserInstructions.pdf
46
     https://calpuff.org/calpuff/FAQ-answers.htm


Version: July 2026                                                                                           Page 56
                                                   British Columbia Air Quality Dispersion Modelling Guideline


Table 6.1          An Explanation of the 7 Critical User-Defined, Site Specific Parameters When Using Observational Data in CALMET (Barclay & Scire, 2011)


 Option                                   Parameter     Recommended Value               Explanation and Justification


 Terrain radius of influence (km)         TERRAD        No Default                      TERRAD, a terrain scale, used in computing slope flow effects (ISLOPE) and
                                                                                        terrain blocking effects (IFRADJ) on the wind field. Consider TERRAD as the
                                                                                        distance (km) that CALMET 'looks' at in computing each of these effects. For
                                                                                        instance the distance of the slope of the nearby terrain is needed to compute the
                                                        Requires user input. Value in
                                                                                        slope flow. TERRAD should not be too small otherwise nearby valley walls which
                                                        km specific to each model
                                                                                        contribute to the slope flow will not be seen. On the other hand, TERRAD must
                                                        domain.
                                                                                        not be so large that hills more than one valley away are seen. TERRAD can be
                                                                                        estimated as the typical ridge-to-ridge distance divided by two, and usually
                                                                                        rounded up. Typical values of TERRAD are between 5 and 15 km with an upper
                                                        Used in both No-Obs and Obs     limit of about 20 km (except for very large grid spacing simulations). See TRC
                                                        modes                           (2011) Section 3.2.3 for further guidance.


 Vertical extrapolation of surface wind   IEXTRP        Default (-4): ‘to extrapolate   This switch affects whether the model allows vertical extrapolation of surface
 observations                                           using similarity theory” and    data or not. This switch was developed since upper air observations are typically
                                                        to exclude upper air            only taken every 12 hours. The vertical extrapolation of surface wind
                                                        observations from Layer 1       observations allows for the hourly surface data to impact layers above the
                                                                                        surface layer.

                                                                                        The default is -4, which means similarity theory is used to extrapolate the surface
                                                        Not used in No-Obs mode         winds into the layers aloft, which provides more information on the observed
                                                                                        local effects to the upper layers.

                                                                                        A value of IEXTRP < 0 means that the lowest layer of the upper air observations
                                                                                        will not be considered in any interpolations.

                                                                                        IEXTRP = 0 means that no vertical extrapolation from the surface wind data is
                                                                                        used.




Version: July 2026                                                                                             Page 57
                                                        British Columbia Air Quality Dispersion Modelling Guideline


Table 6.1 (Cont’d)             An Explanation of the 7 Critical User-Defined, Site Specific Parameters When Using Observational Data in CALMET (Barclay & Scire,
2011)


Option                                          Parameter     Recommended Value                Explanation and Justification


Layer dependent weighting factor of             BIAS (NZ)     Default (NZ * 0) is to not       BIAS is most often used in complex terrain. It ranges from -1 to +1, and a value is
surface vs. upper air wind observations in                    change the 1/R2 weighting        input by the user for each vertical layer. A value of -1 means the surface station
defining the Initial Guess Field (IGF) winds.                 given equally to surface and     has 100% weight for that layer, while a value of +1 means the upper air station
Observations are always weighted by                           upper air data                   has 100% weight for that layer. In simple terrain situation, BIAS is often set to
inverse distance squared (1/R2) from the                                                       zero (0) for each vertical layer (i.e., the upper air and surface wind and
station to the grid point. The BIAS                           Not used in No-Obs mode          temperature observations are given equal weight in the 1/R2 interpolations used
parameter changes that weight.                                                                 to initialize the computational domain).
                                                              Requires user input,
                                                              depending on validity of         The BIAS affects how the initial Step 1 winds will be interpolated to each grid cell
                                                              surface and upper air stations   in each vertical layer based on upper air and surface observations. By setting BIAS
                                                                                               to -1, upper air observations are eliminated in the interpolations for this layer.
                                                                                               Conversely by setting BIAS to +1, surface observations are eliminated in the
                                                                                               interpolations for this layer.

                                                                                               An example where non-default settings for BIAS may be used is for a narrow,
                                                                                               twisting valley, where the only upper air observations were 100 km away, and the
                                                                                               only local surface wind observations were in one location in the valley. For this
                                                                                               example, set BIAS to -1 within the valley forcing surface data only to be used for
                                                                                               the lowest layers, and BIAS to +1 above the valley (which could be double the
                                                                                               valley depths depending on how much the valley effects the flow above it) forcing
                                                                                               upper air data only to be used aloft, and BIAS might go from -1 to +1 in the
                                                                                               transitional layers at the top of the valley.




Version: July 2026                                                                                                   Page 58
                                                        British Columbia Air Quality Dispersion Modelling Guideline


Table 6.1 (Cont’d)            An Explanation of the 7 Critical User-Defined, Site Specific Parameters When Using Observational Data in CALMET (Barclay & Scire,
2011)


Option                                         Parameter      Recommended Value               Explanation and Justification


Weighting parameter for Step 1 wind field      R1 and R2      No Default                      The value of R1 and R2 are used in the construction of the Step 2 wind field,
vs. observations in Layer 1 (R1) and Layer 2                                                  where the observed winds are 'blended' in with the Step 1 winds. R1 represents
and above (R2)                                                Requires user input. Value in   the distance from a surface observation station at which the surface observation
                                                              km specific to each model       and the Step 1 wind field are weighted equally.
                                                              domain, 1 value represents
                                                              all stations                    R2 represents the comparable distance for winds in layers 2 and above. It is
                                                                                              important to note that all the results of the diagnostic wind model (kinematics,
                                                              Not used in No-Obs mode         slope and blocking effects) are contained in the Step 1 wind field, thus if too much
                                                                                              weight is given to the observations, then you will essentially erase all the
                                                                                              information generated in creating the Step 1 winds. Rule of thumb is to start with
                                                                                              small R1 and R2 values and slowly increase these values if you do not believe the
                                                                                              surface stations are showing enough weight. Typically for observation sites in flat
                                                                                              terrain values of R1 and R2 are larger than in mountainous terrain where a
                                                                                              station’s flow is limited by the valley segment.


Maximum radius of influence for                RMAX1          No Default                      The values of RMAX are also used in the construction of the Step 2 wind field,
meteorological stations in layer 1 (Step 2)                                                   where the observed winds are 'blended' in with the Step 1 winds. Any observation
and layers aloft (Step 2)                      and            Requires user input. Value in   for which Rk (the distance from the grid cell to the k-th observation location) is
                                                              km specific to each model       greater than RMAX1 in the surface layer or RMAX2 aloft is excluded from the
                                               RMAX 2         domain                          'blending' formula. RMAX1 and RMAX2 can be used to exclude observations from
                                                                                              being inappropriately included (as they are in the next valley, on the other side of
                                                              One value represents all
                                                                                              a mountain, etc.). Note, if RMAX1 and RMAX2 are used to exclude observations,
                                                              stations
                                                                                              then do not set LVARY to T, as CALMET will increase the values of RMAX1 and
                                                                                              RMAX2 to at least capture the nearest observation, regardless of whether this
                                                              Not used in No-Obs mode
                                                                                              makes sense.

                                                                                              Typically values of R1 and R2 are smaller than RMAX1 and RMAX2, this way
                                                                                              ‘sharp’ boundaries between the Step 1 wind field and the weighted observation
                                                                                              station are prevented.




Version: July 2026                                                                                                  Page 59
                                  British Columbia Air Quality Dispersion Modelling Guideline


Table 6.2           Recommendations for Key CALMET (Version: 6.4.0, Level: 121203) Model Options in Input Group 4
and 5

                                  Black (do not touch) Dark Grey (recommended default)
                           Light Grey (expert judgement required to determine) White (not used)


 Option                              Parameter    Value        Explanation & Justification
 Determines whether
                                                                (0) CALMET is run with observation data only, (1) a
 observation data are used, or in
                                        NOOBS      0, 1 or 2   combination of observations and NWP output, (2) or just NWP
 combination with NWP model
                                                               output. See Section 6.4.1 for further guidance.
 output, or NWP data only
                                                               If gridded NWP model used: (4) compute the gridded cloud
 Cloud Data Option: 1,2,3,4            MCLOUD         4
                                                               cover from prognostic relative humidity at all levels.
 Wind field model selection                                    (1) Use diagnostic wind module for wind field creation via 2-
                                        IWFCOD        1
 variable.                                                     Step approach, (0) objective analysis only: not recommended.
                                                               Used to evaluate thermodynamic blocking effects of the terrain
 Compute Froude number
                                        IFRADJ        1        on the wind flow and are described using the critical Froude
 adjustment effects?
                                                               number (see CRITFN to define). Used only if IWFCOD=1.
                                                               (0) Do not compute a terrain-forced vertical velocity in the
 Compute kinematic effects?              IKINE        0        initial guess wind field. When IKINE=0, DIVLIM, NITER and
                                                               ALPHA do not need to be specified.
 Use O’Brien procedure for
                                                               No adjustment to the vertical velocity profile at top of model
 adjustment of the vertical              IOBR         0
                                                               domain.
 velocity?
                                                               Slope flow is parameterized as a function of sensible heat flux,
 Compute slope flows?                   ISLOPE        1
                                                               distance to the crest, and terrain slope.
                                                               (-4) Extrapolate surface observations using similarity theory.
 Extrapolate surface wind                                      (1) no extrap, (2,3) other extrap methods used. See Table 6.1.
                                        IEXTRP        -4
 observations to upper levels                                  If IEXTRP<0: ignore layer 1 data of upper air stations if surface
                                                               station nearby as it is likely more representative.
                                                               If surface meteorological data and upper air soundings are
                                                               used alone to run CALMET, ICALM should be:
                                                               • (0) when unrealistically high % of calm (due to high
                                                                  anemometer starting threshold) and the location is not
 Extrapolate calm winds aloft?          ICALM       0 or 1        prone to calms/inversions.
                                                               • (1) when % calm is realistic (low threshold anemometer) and
                                                                  the location is prone to frequent stagnant conditions
                                                                  (valleys).
                                                               If NWP model output is used in CALMET, ICALM=0.
                                                               -1<=BIAS<=1. See Table 6.1
 Layer-dependent biases.                 BIAS       varies     Note: BIAS is only used when using observations to develop
                                                               initial guess field. Not active when No-Obs =1,2
                                                               (-1) when IEXTRP = ± 2,3,4 to ensure extrapolation of all
                                                               surface stations (i.e., unlimited distance). Option designed to
 Min. distance (km) between                                    avoid extrapolated surface data “competing” with upper air
 upper air stn and srfc stn for                                measurements when both surface and upper air
                                        RMIN2         -1
 which extrapolation of srfc                                   measurements are co-located. However, the better time
 winds will be allowed.                                        resolution of the hourly surface data suggests extrapolating
                                                               may be appropriate. RMIN2 defines the distance between
                                                               measurements defining “co-located”.
 Use gridded prognostic wind
                                        IPROG       0 or 14    (0) No (14) where NWP output used as Initial Guess field.
 field (NWP) model output?



Version: July 2026                                                                                                 Page 60
                                  British Columbia Air Quality Dispersion Modelling Guideline



 Time step (hrs) of the NWP
                                       ISTEPPGS       3600      (1) Usually, it is an hourly time step (3600 seconds)
 output used as input data
                                                                (0) Off. If (1) the coarse CALMET fields from an earlier run will
 Use coarse CALMET fields as
                                        IGFMET         0        be used to define the initial guess fields (could be used if no
 initial guess fields?
                                                                NWP output available).
                                                                (F) Off (T) use objective analysis only rather than the 2 Step
                                                                diagnostic wind module (when IWFCOD=0). Radius of
                                                                influence is expanded when no stations are within the fixed
 Use varying radius of influence?       LVARY          F
                                                                radius of influence value. Caution: RMAX is effectively
                                                                enlarged to incorporate the “nearest” station regardless of it
                                                                suitability.
 Max radius of influence over                                   Used to exclude obs at surface from being inappropriately
                                        RMAX1        varies
 land of the surface layer.                                     included (i.e., if they are in the next valley). See Table 6.1
 Max radius of influence over                                   Used to exclude obs aloft from being inappropriately included.
                                        RMAX2        varies
 land aloft.                                                    See Table 6.1
 Max radius of influence of over                                Must be large enough so that all grid points over water are
                                        RMAX3        varies
 water – all layers                                             within the radius of influence of at least one observation.
 Min. radius of influence used in                               Use small value (0.1 km). Prevents a divide-by-zero error when
                                         RMIN         0.1
 the wind field interpolation.                                  a grid point and station are co-located.
 Radius of influence of terrain
                                        TERRAD       varies     Requires expert judgment. See Table 6.1
 features
 Distance from a sfc stn at which
 the obs and 1st guess field are          R1         varies     Requires expert judgment. See Table 6.1
 equally weighted.
 Distance from an upper air stn
 at which the obs and 1st guess           R2         varies     Requires expert judgment. See Table 6.1
 field are equally weighted.
                                                                Change this value if IPROG=1 (CSUMM model winds are used in
 Relative weighting of the
                                        RPROG          0        the Step 1 wind field). CSUMM output is rarely used and not
 prognostic wind field data.
                                                                recommended.
 Maximum acceptable
 divergence in the divergence           DIVLIM       5x10-6     Not used since IKINE=0
 min. procedure.
 Maximum # of iterations in the
                                         NITER         50       Not used since IKINE=0
 divergence min. procedure.
                                                                2 passes in the lowest layer, and 4 passes in the higher layers.
 # of passes in the smoothing
                                      NSMTH (NZ)    2,4,4,4,…   More passes will smooth the final wind field and may be
 procedure.
                                                                needed in complex terrain - but rarely altered
 Max # of stns used in each layer
 for the interpolation of data to       NINTR2         99       Use default.
 a grid point
                                                                If Froude no. < CRITFN, wind has an uphill component and
 Critical Froude number.                CRITFN         1        direction is changed to be tangent to the terrain. If Froude
                                                                no. > CRITFN, no adjustment is made.
 Empirical factor controlling the
                                        ALPHA         0.1       Not used since IKINE=0
 influence of kinematic effects.
 Multiplicative scaling factor for
 extrapolation of surf obs to         FEXTR2 (NZ)   NZ * 0.0    Seldom used and not used when IEXTRP = ± 4.
 upper layers.




Version: July 2026                                                                                                   Page 61
                                  British Columbia Air Quality Dispersion Modelling Guideline



                                                      0       Usually not used. Use barriers to block out a certain station
 Number of barriers to
                                         NBAR                 effects. Barriers can extend from the surface layer to user-
 interpolation of the wind fields.                 depends    defined upper layer limit.
                                                              Used only if NBAR > 0. User defined switch to control vertical
 Level (1 to NZ) up to which
                                         KBAR       varies    extent of barriers. Requires careful examination of the
 barriers apply.
                                                              resulting wind field at each level.
                                        XBBAR
                                        YBBAR
 X and Y coordinates of barriers.                   varies    Use only if NBAR > 0 to define the coordinates of the barrier.
                                        XEBAR
                                        YEBAR
                                                              Surface temperatures computed internally using hourly surface
 Diagnostic module surface
                                        IDIOPT1       0       observations or NWP model output. (1) read pre-processed
 temperature option.
                                                              values from DIAG.DAT.
 Diag module sfc station to use
                                        ISURFT        -1      (-1) 2-D spatially varying surface temperatures
 for the sfc temp (stn ID).
                                                              (0) Lapse rate computed internally from twice-daily upper air
 Diagnostic module domain-
                                        IDIOPT2       0       observations or NWP model output. (1) read pre-processed
 averaged lapse rate option.
                                                              values from DIAG.DAT.
 Diagnostic module upper air stn
 (stn ID) to use for lapse rate to       IUPT         -1      (-1) 2-D spatially varying potential temperature lapse rate
 use.
 Depth through which the
 domain-scale lapse rate is              ZUPT        200      Units: Meters. Only used if IDIOPT2 = 0.
 computed.
 Initial guess field wind                                     (0) Computed internally from observations or NWP output
                                        IDIOPT3       0
 components                                                   wind fields. (1) read pre-processed values from DIAG.DAT
 Upper air station to use for                                 Use 3-D initial guess fields. Used only if IDIOPT3 = 0 and
                                       IUPWND         -1
 domain-scale winds.                                          NOOBS = 0.
 Bottom and top of layer
 through which the initial guess       ZUPWND       1, 1000   Units: Meters. Used only if IDIOPT3 = 0
 winds are computed.
 Observed surface wind
                                                              Read wind speed and wind direction from a surface data file
 components for wind field              IDIOPT4       0
                                                              (SURF.DAT). DIAG.DAT not used.
 module.
 Observed upper air wind                                      Read wind speed and wind direction for an upper air data file
                                        IDIOPT5       0
 components                                                   (UP1.DAT, UP2.DAT, etc.). DIAG.DAT not used.
 Use Lake Breeze Module?                LLBREZE       F       No, do not use Lake Breeze Module.
 # of boxes defining region.             NBOX         0       Used only if LLBREZE = T.
 X Grid line 1 defining the region
 of interest                             XG1
                                                      0       Used only if LLBREZE = T. One set per box.
 X Grid line 2 defining the region       XG2
 of interest
 Y Grid line 1 defining the region
 of interest                             YG1
                                                      0       Used only if LLBREZE = T. One set per box.
 Y Grid line 2 defining the region       YG2
 of interest
 X Point defining the coastline                               Beginning x coordinate (km) of user defined coastline. Used
                                        XBCST         0
 (straight line)                                              only if LLBREZE = T. One for each box.
 Y Point defining the coastline                               Beginning y coordinate (km) of user defined coastline. Used
                                        YBCST         0
 (straight line)                                              only if LLBREZE = T. One for each box.



Version: July 2026                                                                                                Page 62
                                  British Columbia Air Quality Dispersion Modelling Guideline



 X Point defining the coastline                              Ending x coordinate (km) of user defined coastline. Used only if
                                        XECST         0
 (straight line)                                             LLBREZE = T. One for each box.
 Y Point defining the coastline                              Ending y coordinate (km) of user defined coastline. Used only if
                                        YECST         0
 (straight line)                                             LLBREZE = T. One for each box.
 Number of stations in the                                   Surface and upper air stations in a box. Used only if LLBREZE =
                                         NLB          0
 region                                                      T. One for each box.
 Station ID’s in the region.
                                       METBXID
 Surface stations first, then                         0      Used only if LLBREZE = T. One set per box.
                                        (NLB)
 upper air stations.




Version: July 2026                                                                                              Page 63
                             British Columbia Air Quality Dispersion Modelling Guideline



     DISPERSION MODELLING

        SIZE OF DOMAIN

The model domain will generally be greater for tall stacks with buoyant emissions where a domain of 50 by 50 km centred
on the stack may be required due to the large area affected by the emissions. For shorter stacks, a smaller domain may be
appropriate (e.g., 10 km by 10 km). Some recommended ways to establish the domain are:

        As a starting point, establish the domain on the basis of the isopleth resulting from the project only case that
         represents 10% of the ambient air quality objective.
        For AERSCREEN and AERMOD, consider the terrain in defining the domain boundaries (e.g., elongated
         domain for valleys defined by bounding mountain ridges).
        Consider sensitive receptor areas (e.g., a hospital, recreation area or neighbourhood) or areas of interest
         such as nearby residents/communities where interest in the predictions may be high.
        Consider other emission sources that need to be included in the modelling such as sources that contribute to
         baseline (see Section 8.1) whether they currently exist or could be built in the future.
        CALPUFF domain should be big enough to capture potential recirculation of pollutants.

        RECEPTOR SPACING

Receptors are the locations within the model domain where the concentration/deposition predictions are calculated. The
location and number of receptors must be judiciously selected in order to achieve a balance between enough receptors to
resolve maximum concentrations and too many receptors where computer processing times and output files become
unreasonable.

Consider the following receptor spacing as a minimum:

        20 m receptor spacing along the plant boundary (as defined in Section 7.3)
        50 m spacing within 500 m of source
        250 m spacing within 2 km of source
        500 m spacing within 5 km of source
        1000 m spacing beyond 5 km of source

The above are for general guidance for a single point source or for multiple sources that are close and will depend on
several factors such as source types, the distance to plant boundary, etc. For example, area sources such as an open pit
mine, the “source” may cover a large irregularly shaped area, so the initial receptor spacing as recommended above may
have to be adjusted based on the facility boundary and the shape of the area source.

Finally, receptor spacing may also depend on whether there are specific areas where predictions are of interest such as
populated areas where higher resolution may be required even though the concentration/deposition gradients are low.
These are considered to be sensitive receptors (see Section 7.4)




Version: July 2026                                                                                           Page 64
                             British Columbia Air Quality Dispersion Modelling Guideline



        PLANT BOUNDARY

Model output is typically compared to ambient air quality objectives. Although the areas of applicability of the BC ambient
air quality objectives are not defined, often they are applied to areas where there is public access (i.e., beyond the plant
boundary). The plant boundary is determined to be the following:

        The facility fence line or the perimeter of disturbed area that defines where public access is restricted.
        If a facility is located within another larger facility boundary, the plant boundary is the boundary of the
         encompassing facility.
        If a public access road passes through the plant, the plant boundary is the perimeter along the road
         allowance.
There may be areas outside the plant boundary such as over a large body of water where public access is likely to be
very rare. In this case, the human and environmental risk is low even though these areas may have predicted
contaminant concentrations. The applicability of the model output and ambient air quality objectives for such areas is
a matter to be discussed with the Ministry.

        SENSITIVE RECEPTORS

It may be desirable to predict the concentrations (maximums or some percentile) at sensitive receptors such as individual
residences, residential areas, schools, hospitals, commercial day care and seniors’ centres, campgrounds, parks,
recreational areas, and sensitive ecosystems. AERMOD and CALPUFF have the ability to output concentrations at specified
receptors in addition to a grid of receptors, and AERSCREEN is able to output concentrations at specified distances in
addition to gridded distances.

        FLAG POLE RECEPTORS

Dispersion models allow the user to select the receptor height, called flagpole receptors. Under most situations,
ground-level (i.e., a receptor height of 0 m) concentration predictions are sufficient since the concentration profile
within the first few metres of the ground is expected to be fairly homogeneous due to mechanical mixing. For very
short stacks (i.e., a few metres), the difference in flagpole receptor height between 0 and 1.5 m may lead to large
differences in the predicted concentrations in areas near the source.

Elevated flagpole receptors may be required to determine concentrations for treetop concentrations in order to
assess the potential for vegetation effects, or to estimate the exposure to humans where 1.5 m flag pole receptors
would be considered to be breathing height. Flagpole receptors should be identified in the Dispersion Modelling Plan.

        BUILDINGS

If emissions occur close to buildings, the downwind dispersion can be affected by the flow deformation and enhanced
turbulence created as air moves over and around the building. For example, building downwash may occur where
emissions from stacks on or near buildings may be drawn in the cavity zone behind the building (and re-circulate
resulting in high concentrations) and/or caught in the wake turbulence, mixing the plume to the ground rapidly
resulting in high ground-level concentrations.

All of the Guideline models can handle the effects resulting from a single or several buildings with different
dimensions and orientations.



Version: July 2026                                                                                             Page 65
                             British Columbia Air Quality Dispersion Modelling Guideline


For AERSCREEN, if downwash from only one building is to be accounted for building downwash, the following
parameters must be provided:

        orientation of maximum building horizontal dimension relative to north (0-179 degrees)
        building height
        maximum horizontal building dimension
        minimum horizontal building dimension
        angle from north of stack location relative to building centre (0-360 degrees)
        distance between stack and building centre

Building Profile Input Program for PRIME (BPIPPRM) can be used to prepare downwash related input for the Plume
rise Model Enhancements (PRIME) building downwash algorithm. BPIPPRM can determine whether a stack is
subjected to wake effects from a structure(s), and calculate building heights (BH) and projected building widths (PBW)
for cases when the plume is affected by building wakes.

In multiple building situations, BPIPPRM determines building separation distances and will fill in the gap between the
buildings under specific circumstances if they are “sufficiently close”. With the addition of more buildings and stacks,
a maze of influence zones results, and BPPPRM automates these calculations for these complicated situations.

BPIPPRM requires the following information:

        X and Y coordinates for all building corners and stacks,
        Height of all stacks and buildings, and for buildings with more than one height or roofline, then each tier
         height with respect to the building base elevation, and
        Base elevations of all stacks.

BPIPPRM produces a file for AERSCREEN, AERMOD and CALPUFF if downwash is to be treated. For AERSCREEN, the
output from BPIPPRM can be read directly by specifying the filename of an existing BPIPPRM output file. For AERMOD
and CALPUFF, the output file must by copied and pasted into the required Input Group for source and building
information.

In addition to PRIME, CALPUFF also includes the ISC model downwash treatment. Evaluations of both methods for
buildings with different aspect ratios (building width/building height) show that the ISC treatment performs better for
buildings with aspect ratios > 5 (squat buildings). For CALPUFF applications, the ISC treatment is recommended if
aspect ratios > 5 (see Table 6.1). When the ISC downwash option is selected for CALPUFF, only the HEIGHT and WIDTH
output parameters from BPIPPRM are required as input into the Point Source Parameters (Input Group 13).

        AERMOD: REGULATORY DEFAULT SETTINGS (CORE MODEL OPTIONS)

The regulatory default option is controlled from the MODELOPT keyword on the CO pathway. As its name implies, this
keyword controls the selection of modelling options. Unless specified otherwise through the available keyword
options, it is recommended that the following AERMOD default options are selected:

        Use the elevated terrain algorithms.
        Use stack-tip downwash (except for building downwash cases).
        Use the calms processing routines.



Version: July 2026                                                                                          Page 66
                             British Columbia Air Quality Dispersion Modelling Guideline



        Use the missing data processing routines.
        Use a 4-h half-life for exponential decay of SO2 for urban sources.

Recommendations in Section 8.2 should be followed for NOx to NO2 conversion. The parameters used to specify
options on the MODELOPT keyword are “secondary keywords” that are descriptive of the option being selected. For
example, to ensure that the regulatory default options are used for a particular run, the user would include the
secondary keyword “DFAULT” on the MODELOPT input.

        CALPUFF: INPUT GROUP 2 AND 12 SWITCH SETTINGS

In order to promote consistency in the application of CALPUFF, similar to the switch setting recommendations for
CALMET provided in Section 6.4.3, the following section provides guidance for CALPUFF Input Group 2 and 12 Switch
Settings.

The switches/user defined parameters are shaded according to the following:

    Black = do not touch (if changed, provide justification)
    Dark Grey = recommended default
    Light Grey = expert judgment required
    White = not used




Version: July 2026                                                                                    Page 67
                                 British Columbia Air Quality Dispersion Modelling Guideline


Table 7.1           Recommendations for Key CALPUFF (Version: 6.42, Level: 110325) Model Options in Input Group 2
and 12

                                      Black (do not touch) Dark Grey (recommended default)
                                              Light Grey (expert judgement required)


  Option                                Parameter       Value         Explanation & Justification
  Vertical distribution used in the
                                           MGAUSS            1        Gaussian.
  near field.
  Terrain adjustment method.               MCTADJ            3        Partial plume path adjustment.
  Subgrid scale complex terrain                                       Usually 0, but allows for CTDM-like treatment of
                                           MCTSG             0
  module flag.                                                        subgrid scale hills
                                                                      No slug model. However it is the recommended
  Near-field puffs modelled as                                        approach for area sources with receptors in the very
                                           MSLUG             0
  elongated?                                                          near field or for episodic time-varying emissions such as
                                                                      accidental releases.
  Transitional Plume Rise modelled?        MTRANS            1        Transitional rise computed.
                                                                      (1) Stack-tip downwash modelled, particularly if ratio of
                                                                      stack gas exit velocity to wind speed is < 1.5.
  Stack-tip downwash?                       MTIP           1 or 2     (2) No stack tip downwash for flares if pseudo-stack
                                                                      parameters are calculated using the AERflare/ABflare
                                                                      spreadsheet.
  Method selected to compute
  plume rise for point sources not          MRISE            1        Briggs plume rise.
  subject to downwash.
  Method used to simulate building                                    (2) PRIME method. (1) ISC type method if building
                                           MBDW            1 or 2
  downwash?                                                           aspect ratios of W/H are > 5 (see Section 7.6)
                                                                      No vertical wind shear. Variable flow in the vertical is
  Vertical wind shear above stack
                                           MSHEAR            0        based on upper air data. If used (1) a power law wind
  top modelled in plume rise?
                                                                      speed extrapolation is applied above stack top.
                                                                      No puff splitting for short-range modelling. In long
  Puff splitting allowed?                  MSPLIT            0
                                                                      range transport, puff splitting may be necessary.
                                                                      (0) No chemical transformation (6) transformation
  Chemical Transformation Scheme.          MCHEM           0 or 6     internally calculated using updated RIVAD scheme with
                                                                      ISOROPPIA equilibrium. Computationally intensive.
                                                                      Used only if MCHEM = 6 or 7. Transformation rates and
                                                                      wet scavenging coefficients adjusted for in-cloud
  Aqueous phase chemistry flag            MAQCHEM            1
                                                                      aqueous phase reactions. Consult Ministry if an
                                                                      alternative scheme is requested.
                                                                      Used only if MAQCHEM = 1. MLWC = 1 is
                                                                      recommended if gridded cloud liquid water content
  Liquid Water Content flag                 MLWC             1        data are available (obtained by CALMET from
                                                                      MM5/WRF output and stores in an auxiliary
                                                                      “calmet.dat.aux” file).
                                                                      (0) no, (1) if wet deposition modelled. Important for
  Wet removal modelled?                     MWET           0 or 1     long range transport but may be used for near-field in
                                                                      appropriate.
                                                                      (0) no, (1) if dry deposition modelled. Important for
  Dry deposition modelled?                  MDRY           0 or 1     long range transport but may be used for near-field in
                                                                      appropriate.



Version: July 2026                                                                                                  Page 68
                                 British Columbia Air Quality Dispersion Modelling Guideline



                                                                     (0) turns off plume tilt, recommended for small
                                                                     particles. (e.g. combustion size particles less than 10
 Gravitational settling (plume tilt)?     MTILT         0 or 1       µm)
                                                                     (1) yes, recommended for very large particles with
                                                                     substantial gravitational settling effects.
 Methods used to compute the
                                                                     (2) turbulence based dispersion coefficients is
 horizontal and vertical dispersion       MDISP           2
                                                                     recommended.
 coefficients.
 Sigma measurements (σv/σθ, σw)                                      When measured sigmas are available, use observed
 from PROFILE.DAT used to               MTURBVW           3          σv/σθ, and σw from the PROFILE.DAT file to calculate σy
 compute σy, σz                                                      and σz. Used only if MDISP =1 or 5.
 Backup method to compute                                            Used only if MDISP=1 or 5. Backup method is PG-based
 dispersion when measured                 MDISP2          3          dispersion coefficients for RURAL areas when
 turbulence data are missing.                                        turbulence data are missing
 Method used for Lagrangian time                                     (0) Lagrangian time scale (617.284s). Only used when
                                         MTAULY           0
 scale for σy                                                        MDISP = 1 or 2.
 Advective-Decay timescale for
                                         MTAUADV          0          (0) no turbulence advection.
 turbulence
                                                                     (1) use standard CALPUFF subroutines. (2) use
 Method used to compute                                              AERMOD algorithms to compute turbulence profiles.
                                         MCTURB           1
 turbulence σv and σw profiles                                       Model evaluations have shown both options have
                                                                     similar performance.
 PG σy and σz adjusted for
                                         MROUGH           0          (0) adjustment for surface roughness is not needed.
 roughness
 Partial plume penetration of                                        (1) Evaluate partial plume penetration into elevated
                                         MPARTL           1
 elevated inversion?                                                 inversions applied to point sources.
                                                                     Model partial plume penetration into elevated
 Partial plume penetration from
                                        MPARTLBA          1          inversions. Important for very hot buoyant area sources
 buoyant area sources
                                                                     such as forest fires.
 Strength of temp inversion                                           (0) computed from default gradients and upper air
 provided in PROFILE.DAT extended         MTINV           0          data. Otherwise (1) if PROFILE.DAT contains detailed
 records                                                             temperature profiles
 Probability Distribution Function
                                                                     (1) only if MDISP = 2 (turbulence based dispersion
 used for dispersion under                MPDF          0 or 1
                                                                     coefficients). Otherwise (0)
 convective conditions
                                                                     (0) do not use, however may be used for applications
 Sub-grid TIBL module used for                                       located along a coastline. If used (1) a coastline file
                                         MSGTIBL        0 or 1
 shore line                                                          (COASTLN.DAT) must be prepared to specify the
                                                                     location of the land-water boundary.
 Boundary conditions                                                 (0) boundary conditions are not used. (1) when
                                          MBCON           0
 (concentration) modeled?                                            boundary conditions may be important
                                                                     (0) FOG model not run. (1) if condensed (visible) plume
 Configure for FOG Model output?          MFOG            0
                                                                     assessment is required.
 Test options specified to see if
                                                                     (0) No checks made. (1) to check for conformity with
 they conform to regulatory               MREG            0
                                                                     US EPA Long Range Transport Guidance
 values?
                                                     σv =.2 for A,   For applications where calm wind and stagnation
 Minimum turbulence velocities,
                                          SVMIN      B, C, D, E or   events are significant, set SVMIN = 0.2 to better
 sigma v and sigma w for each                              F
                                          SWMIN                      represent lateral spread of the plume. Leave SWMIN as
 stability class over land and water
                                                     σw = default    default value.




Version: July 2026                                                                                                 Page 69
                                 British Columbia Air Quality Dispersion Modelling Guideline



        POST PROCESSING

           ADDING BASELINE AIR QUALITY CONCENTRATIONS

Although it is useful to know the predicted incremental air quality contribution of the source, it is the cumulative air
quality that is of importance. This is especially important when comparing model predictions to ambient objectives.

The cumulative air quality is given by:

Cumulative = Baseline + Predicted Increment (contribution from the source(s) modelled)

“Baseline” is the concentration due to emissions from both natural and human-caused sources. In other words, it is
the result of the contribution from all sources except the source(s) being modelled. The baseline may be determined
from air quality monitoring data or may be estimated from modelling other contributing sources or a combination of
both.

Choosing the appropriate baseline concentration can be critical in assessing overall air quality. In order of priority, the
information sources used to establish the baseline concentration level are:

           A network of long-term ambient monitoring stations near the source under study;
           Long-term ambient monitoring at a different location that is adequately representative; and
           Modelled baseline.

Given the importance of the baseline level, it is recommended that the Dispersion Modelling Plan includes a
description of the method to establish baseline.


            SITE SPECIFIC MONITORING DATA

The baseline concentration can be determined through the use of existing ambient monitoring data collected at
the site. Current and historical air quality monitoring data at various sites across BC are available online from the
BC Air Data Archive website 47.

To obtain validated ambient air quality data measured in the Lower Fraser Valley, an email request should be sent to:
AQinfo@metrovancouver.org. MV data may also be available on the Ministry website; however, the MV station data
available from this site may not be validated and/or up-to-date.


            MONITORING DATA FROM A DIFFERENT LOCATION

If there are no representative ambient data available for the site, it is possible to use data from another location on
the condition that the alternate site is located in a similar meteorological and air quality regime. Factors to be
considered are:

           Differences in geophysical characteristics (meteorology, topography, surface features).
           Differences in emissions types (point sources, roads, etc.).



47
     http://envistaweb.env.gov.bc.ca/


Version: July 2026                                                                                            Page 70
                              British Columbia Air Quality Dispersion Modelling Guideline



        Differences in emission changes (rapid urban growth, forest fires, nearby temporary sources, etc.).
        Distance from sources that could influence the monitor.

Finally, a review of the instrumentation and the data collection protocols used in the off-site program is recommended.


         ESTABLISH A MONITORING PROGRAM

Establishing a monitoring program in order to determine existing levels of air quality should be considered for the
following reasons:

        To establish a baseline in situations where existing levels are critical to the assessment of air quality in an
         area.
        To conduct post-construction assessment, after a source is operational, to evaluate the air quality
         assessment and/or to determine compliance with ambient air quality objectives.

Although a fully instrumented, continuous monitoring station operating for at least a year is ideal, establishing baseline
could be determined from a short-term study or passive monitoring. Details of any new ambient monitoring program
need to be reviewed with and accepted by the Ministry.


         SELECTING A BASELINE LEVEL USING MONITORING DATA

When representative monitoring data are available, the next step is to select an appropriate baseline level. Typically a
single value is chosen as baseline, which is assumed to apply for every hour of the year and for every location within the
modelling domain. Choosing an appropriate single value from a year’s worth of monitoring data is problematic given the
variation in concentrations that will occur in time and space over the period.

The value selected for baseline depends on the purposes of the modelling assessment. If a conservative estimate of the
potential cumulative air quality changes due to the source is desired (when determining compliance with ambient
objectives/guidelines, worst-case analysis, potential risk exposure estimates), a conservative value should be used to
establish baseline levels.

To select a baseline level, the following is recommended:

        If the monitoring data are deemed representative of the area under consideration, the baseline should be
         based on the most recent year of hourly monitoring data. At least one year of data is required, and a data
         record that is 75% complete in each quarter of the year is recommended.
        Select baseline levels for time averages that correspond to the modelled time averages. For example,
         cumulative air quality compared to a 24-h average ambient objective would require a 24-h average model
         prediction and a 24-h average baseline level.
        The monitoring data can be pre-screened to exclude any periods when an intermittent source has a
         significant influence on the monitor.
              o   For example, if there are episodes where wildfire smoke is the major contributor, such periods could
                  be eliminated from the data only if it is supported by analysis of the air quality and meteorological
                  record.
        For Level 1 Assessments, select the maximum measured concentration (100th percentile) from the screened
         or unscreened data set as the baseline level.




Version: July 2026                                                                                              Page 71
                             British Columbia Air Quality Dispersion Modelling Guideline



        For Level 2 and 3 Assessments, baseline levels for 1-h cumulative prediction values should be determined as
         follows,
             o    For SO2, use the 99th percentile of daily maximum 1-h values.
             o    For NO2, use the 98th percentile of daily maximum 1-h values.
             o    For other pollutants, use the 98th percentile from the screened hourly data set.
        For Level 2 and 3 Assessments, methods for developing refined 1-h cumulative predictions for NO2 are
         provided in the GUIDANCE FOR NO2 DISPERSION MODELLING IN BRITISH COLUMBIA.
        For Level 2 and 3 Assessments, use the 98th percentile of the 24-h (daily) values from the screened data as
         the baseline level for a 24-h cumulative prediction.
        For Level 2 and 3 Assessments, methods for developing refined 24-h cumulative predictions of mining-
         related fugitive TSP, PM10, and PM2.5 are provided in the GUIDANCE FOR DISPERSION MODELLING OF
         MINING-RELATED FUGITIVE PARTICULATE MATTER EMISSIONS IN BRITISH COLUMBIA.
        For an annual average cumulative prediction, use the annual mean calculated from the screened or
         unscreened hourly data over a year.
        If there is more than one representative monitoring site, follow the previous steps to establish baseline for
         each site for the different averaging periods. The representative baseline value would be the arithmetic
         average of the 1-h, 24-h and annual average values obtained for each site.
        In figures and/or tables that present the modelling results, provide the total air quality (baseline plus the
         increment due to the source(s) modelled) and the modelled incremental concentration (due to the source(s)
         modelled).


         MODELLING BASELINE (CONTRIBUTING SOURCES)

If there are no representative ambient monitoring data, a model can be used to estimate the baseline concentration for
sources where emissions can be quantified. In addition, for practical reasons such an approach would be limited to areas
with a limited number of contributing sources. The decision to include or exclude sources that may or may not contribute
to the baseline concentration depends in part on the scale of the modelling domain. Which sources to include can be
guided by following these recommended steps:

       Run a Screening Model (i.e. AERSCREEN) for each source in order to determine the contribution of each
        source to the concentrations in the area around the source and assess their relative importance.
        Consider the surrounding terrain where a source may be close to the source under consideration but is in an
         adjacent valley separated by high ridges where its contribution to the area of interest is minimal.
        Final decisions on what sources to include to produce a modelled baseline concentration may require an
         iterative approach.

The guidance that is outlined in Section 3.3 regarding source emission rates for model inputs also applies to those
sources included in the modelling to estimate baseline.


         FUTURE BASELINE

There may be situations where the current emissions that contribute to the air quality in the area of interest are
expected to change dramatically in the future (new sources or decommissioning of existing sources). In this case, the
current baseline may not be an appropriate starting point. Estimating the future baseline could be informed by
dispersion modelling only if there is certainty in the emission characteristics from the new or decommissioned sources.



Version: July 2026                                                                                          Page 72
                                British Columbia Air Quality Dispersion Modelling Guideline


The question of whether the future baseline needs to be considered should be identified in the Dispersion Modelling
Plan.

        NO TO NO 2 CONVERSION

The Ministry has released updated guidance on the prediction of NO2 concentrations from nitrogen oxide emissions.
For assumptions and methods recommended for NO to NO2 conversion, refer to GUIDANCE FOR NO2 DISPERSION
MODELLING IN BRITISH COLUMBIA.

Total oxides of nitrogen (NOx) are comprised of nitric oxide (NO) and nitrogen dioxide (NO2). The concentration of NO2 in
the exhaust of typical combustion sources is generally in the order of five to 10% of the NOx concentration.
Transformation of NO to NO2 continues in the atmosphere due to reaction with atmospheric ozone and other volatile
organic compounds.

Given that NO2 is the contaminant of interest due to its effects on human health and the environment, the conversion
for NO to NO2 must be accounted for by a chemical transformation treatment built into the model itself, and/or
through a conversion factor applied to the modelled NOx concentration.

The recommended methods for obtaining the NO2 concentration from the predicted NO concentration are based on a
step-wise approach that involves different conversion treatments which are summarized as follows:

             100% conversion. If there are exceedances, use one of three methods described next.
             If there are adequate (at least one year) hourly NO and NO 2 monitoring data, use the ambient ratio
              method.
             If adequate monitoring data are not available, use the ozone limited method.
             If AERMOD is used, apply the plume volume molar ratio method.


            100% CONVERSION

For any of the models recommended in this Guideline, the most conservative approach is to assume that all of the NO
is converted immediately to NO2. If there are exceedances of the ambient NO2 objective, then the following alternate
methods can be used.


            AMBIENT RATIO METHOD

The Ambient Ratio Method (ARM) is a conversion factor based on measurements of NO and NO 2. If there are at least
one year of representative ambient hourly NO and NO2 monitoring data, an empirically derived conversion formula
can be applied. Alternatively, as a simple alternative to the full conversion curve, the monitoring data may be used to
create a single conversion factor that is based on a range of expected concentrations. The approach should be outlined in
the Dispersion Modelling Plan and approved by the Ministry.

An illustration of the method is provided in RWDI (2005) but is outlined here:

           Plot all the 1-h observations of NO2/NOx (y axis) vs NOx (x axis). The scatter will be limited to a distinct
            envelope (see example, Figure 8-1).
           Estimate up to what NOx concentration the NO2/NOx ratio should be 1. This is the lower NOx limit.




Version: July 2026                                                                                                 Page 73
                            British Columbia Air Quality Dispersion Modelling Guideline



       Estimate the NOx concentration where NO2 = 0.1 NOx. This is the upper NOx limit.
       Fit an exponential equation of the form to the upper envelope of the scatter:

                                                  NO2/NOx = a NOxb

        Where a and b are selected through an iterative process to produce a curve that fits the upper bound of the
        envelope of the scatter (see example, Figure 8.1).

       The formula should not be used for concentrations below the lower NO x limit in order to ensure that the
        calculated NO2/NOx ratio never exceeds unity.
       Check the equation: make sure that NO2 does not decrease with an increase in NOx.
       To account for baseline, follow Section 8.1.4 to establish the baseline 1-h and annual average NOx. Add the
        modelled NOx to this baseline to obtain a total 1-h NOx. Apply the ARM curve to obtain the total NO2
        concentration.




Figure 8.1       Example: Dependence of NO2/NOx Ratio to 1-h Average NOx Concentration, (RWDI, 2005)


        OZONE LIMITING METHOD

If adequate hourly NO and NO2 monitoring data are not available, the Ozone Limiting Method (OLM) can be applied with
the approval of the Ministry. The NO2 concentration can be determined using the following equation (Cole &
Summerhays, 1979):

                             NO2 = 0.1* NOx + the lesser of (O3 or 0.9 NOx) + baseline NO2




Version: July 2026                                                                                       Page 74
                              British Columbia Air Quality Dispersion Modelling Guideline


The initial 10% conversion of NO to NO2 (the first term in the equation) may not be appropriate for all types of NOx
sources. In this case, it is acceptable to use an initial conversion value that better corresponds to the source type. This
should be identified with suitable justification in the Dispersion Modelling Plan.

The above equation assumes the concentrations for all constituents are in vol/vol units (e.g., ppm). For modelling
purposes, NOx is the predicted concentration as NO2 since the NOx emission rate is generally expressed as NO2 equivalent.
If there are multiple plumes that have merged together, apply OLM to the combined plume NOx concentration. Otherwise
for plumes that do not merge, apply OLM to each plume separately.

The O3 concentration used for the OLM is based on the maximum hourly O3 concentration measured from one-year of
representative monitoring data (i.e., typically the closest station if it represents the conditions that the plume would
experience). Select the maximum hourly O3 value (100th percentile) from this data set for the input O3. Care is required in
selecting the appropriate O3 value, as it may already be depleted (or scavenged) by local NOx sources.


         PLUME VOLUME MOLAR RATIO METHOD (AERSCREEN AND AERMOD ONLY)

For AERSCREEN and AERMOD applications use the Plume Volume Molar Ratio Method (PVMRM) for NO x conversion
(Hanrahan, 1999). This method limits the conversion of NO to NO2 based on the amount of O3 available in the plume
and accounts for the changing plume volume due to dispersion. The method also accounts for merging plumes in
multi-source modelling scenarios. There are different input options for baseline O3, the NO2/NOx equilibrium ratio,
and the in-stack NO2/NOx ratios. The following are recommended input options when selecting this method:

        For baseline O3 , use a single value based on the maximum hourly O3 concentration measured from one-year
         of representative monitoring data. Care is required in selecting the appropriate O3 value, as it may already be
         depleted (or scavenged) by local NOx sources.
        NO2/NOx equilibrium ratio = 0.90 unless information is available to justify using a different value.
        NO2/NOx in-stack ratio = 0.10 unless information is available to justify using a different value.


         RIVAD/ARM3 AND RIVAD/ISORROPIA CHEMICAL TRANSFORMATION

The CALPUFF chemical conversion modules (i.e., MESOPUFF II, RIVAD/ISORROPIA) are used for acid deposition and
secondary PM2.5 estimates. In addition, the RIVAD/ARM3 and RIVAD/ISORROPIA chemical transformation schemes are
able to provide predicted NO2 concentrations as part of the output suite of parameters. However, the use of these
modules directly for local and regional scale NO2 estimates has not been evaluated. Until more evidence emerges on the
use of CALPUFF for this specific purpose, the Ministry does not recommend the use of RIVAD/ARM3 and
RIVAD/ISORROPIA for NO/NO2 conversion.

        DISPERSION MODEL OUTPUT

If the model output is to be submitted to the Ministry for review and acceptance, the recommended information
should include sufficient information (tables, figures) that address the objectives of the study, the methodology and
other supporting information to prove that the model has been applied properly and the model output can be used to
inform decision makers. It is crucial that documentation includes sufficient methodology such that the reviewer can
understand the assumptions and steps involved in the work. Recommended documentation for Level 1, 2 and 3
Assessments are defined below.




Version: July 2026                                                                                                 Page 75
                             British Columbia Air Quality Dispersion Modelling Guideline



         LEVEL 1 ASSESSMENTS

The following documentation is recommended for Level 1 Assessments:

        Description of emission sources (locations, elevations).
        Stack dimensions, exit parameters, and emission rates that characterize the emissions.
        Receptor grid resolution and domain size.
        Description of topography, land use and sensitive receptors in model domain.
        Building dimensions.
       Identification of the screening model, assumptions and switches/options used.
       Maximum concentration of contaminants predicted including location and corresponding meteorological
        conditions.
       Baseline concentration of the contaminants of concern.
       Input files (e.g. AERSCREEN.INP, terrain data file, and external file that provides the values of surface
        characteristics).
       Printout of screening model output.


         LEVEL 2 OR 3 ASSESSMENTS

The following documentation is recommended for Level 2 or 3 Assessments:

Site Description

        A site plan showing location and elevation of emission sources and buildings.
        Description of topography and land use in model domain and geophysical data used in assessment (map
         showing contours, residential areas, roads, prominent geographic features).


Emissions Description

        Stack dimensions and exit parameters that characterize the emissions.
        Emission rates used in assessment.


Meteorological Data Description

        Description of meteorological data used in assessment, reasons for their use, and presentation of wind roses
         (seasonal and annual).
        If CALMET applied, examples of meteorological data (winds, temperature, stability) in space and time.
        Any pre-processing utilities and assumptions applied to prepare the data set.
        Description of meteorological conditions leading to air quality episodes.
        If NWP model output used, description of QA/QC tests undertaken to assure the quality of the output.




Version: July 2026                                                                                           Page 76
                            British Columbia Air Quality Dispersion Modelling Guideline


Model-Related Information

        Identification of model used for assessment, stating any assumptions and modifications and identifying
         settings used in the model.
        Receptor grid resolution and size.
        Description of QA/QC procedures used to ensure model input (e.g. CALMET-generated wind fields) is correct,
         and model behaviour is reasonable.


Specific Output for All Pollutants Modelled (not all required, but depends on objectives of modelling study)

        Baseline concentration and an explanation as to why these baseline values were chosen.
        Figures showing the isopleths of maximum (for all averaging times of concern) and annual average predicted
         concentrations overlaid on a map of the model domain. Map must show the terrain, identify the sources
         modelled and all sensitive receptors.
        Exceedance frequencies above a specified threshold concentration.
        For multi-year dispersion modelling, map showing isopleths of maximum concentrations (including 1-h, 24-h,
         and annual average) over the model period (e.g. 3 years) as well as each individual year and including all
         receptors modelled.
        Concentrations based on different emission scenarios (permitted, normal operation and other operating
         conditions as specified).
        Model output may include scenarios of existing sources, individual new sources, project case and cumulative.
        Figures showing isopleths of 98th or 99th percentile values. This may be based on hourly concentrations or
         daily maximum hourly values.
        Time series of model output for additional sources for a receptor located at an existing monitoring site for
         the same monitoring period. Model predicted concentrations should be added to monitored concentrations
         to determine the change in air quality at that receptor.



         COMPUTER FILES

Electronic copies of input and critical output files must be compiled and available upon request by the Ministry.




Version: July 2026                                                                                         Page 77
                               British Columbia Air Quality Dispersion Modelling Guideline



       QUALITY ASSURANCE AND QUALITY CONTROL

Good modelling practice involves the examination of the input files to ensure that specific data treatments have been
applied properly (missing data, calms, formats and units). However, even with error-free input files, there is no
assurance that the model output will be correct. For example, CALPUFF has user-adjustable parameters, where expert
judgment is required to set the right combination of these parameters in order to improve model realism. A review of
the output is essential as it can indicate whether the model is behaving as expected under different circumstances.
Spotting odd model behaviour and errors through this type of analysis comes from experience as well as expert
training.

Section 9.1 provides guidance on the application of QA/QC for CALPUFF as this model can involve more professional
judgment for its proper application than the other recommended models. A discussion on model performance and
uncertainty can be found in Section 9.2.

           CALMET/CALPUFF QA/QC PROCESS

Much of the quality assurance for CALPUFF depends on following the recommendations for preparation of the input
data and the appropriate selection of model parameters as outlined here and the CALPUFF User’s Guide48. Further
efforts for quality control are recommended in the following sections. They are not exhaustive, nor are all of them
necessary. They focus on CALMET, as this component of the CALPUFF modelling system involves the greatest amount
of data preparation, option choices and user-defined parameters.


            CALMET/CALPUFF QA FILES
A good starting point is to review the QA files produced by CALMET/CALPUFF to plot and check the locations of the
grid, NWP grid points, and source locations as these reflect the locations actually seen by CALMET/CALPUFF, including
any undetected errors in the input files (blanks, comma instead of period, wrong UTM zone, etc.).

The following sections provide more detailed QA/QC procedures on the input data and output files.


            CALMET INPUT DATA
           Plot the terrain and land use from the GEO.DAT input files to ensure they match with other maps of the area.
           Plot the locations of the meteorological observation stations to check whether they are located properly in
            the horizontal and vertical.
           Compare all the CALMET-ready input files with the raw data to ensure no errors in data conversion to
            CALMET-ready files (reformatting, unit conversions, etc.).
           Compare each month of CALMET input meteorological files with each other to ensure all parameters are
            consistent from month to month.
           For meteorological files, plot wind roses and frequency distributions of the various input meteorological
            parameters and check for reasonableness given the location and geophysical characteristics. If NWP model
            output is used, the QA/QC process is provided in Section 6.1.




48
     https://www.calpuff.org/calpuff/download/CALPUFF_UsersGuide.pdf


Version: July 2026                                                                                           Page 78
                             British Columbia Air Quality Dispersion Modelling Guideline



        For the CALPUFF.INP file, review all source information (values, formats, units) associated with Input Group
         13-16 of the CALPUFF.INP file to ensure emission information is correct.
        Plot the source locations to ensure that they are located properly and ensure that their vertical location
         (stack base relative to terrain height for that location) is correct.
        For the CALPUFF.INP file, review locations (horizontally and vertically) of all specified receptors.


         CALMET OUTPUT DATA
        Select a few representative periods where thermally driven flows (i.e., upslope and drainage flows) would be
         expected. Plot the wind vector fields at various levels to confirm that the wind fields are reasonable given the
         terrain and the meteorological conditions.
        Plot the frequency distribution of surface wind speeds for different locations in the domain and at the
         surface station locations and check for reasonableness (compare with observations, consider the location,
         and what might be expected given the topography).
        Plot annual and seasonal surface wind roses for different locations as well as the surface station locations
         and check for realism (compare with observations, consider the location, and what might be expected based
         on topography).
        For different 24-h periods within a summer and winter season, plot a surface, mid-level and upper-level wind
         field every hour for a 24-h period with light winds and stable conditions. Check for reasonableness of the
         wind fields in the domain (extent of terrain effects and the appropriateness of the settings that require
         expert judgment).
        Plot time series of average surface temperature by month for the source location as well as surface station
         locations. Compare with observations/climate normals. Check for reasonable monthly variation for the given
         locations.
        Plot time series of average surface temperature by hour-of-day for the source location as well as surface
         station locations. Compare with observations/climate normals. Check for reasonable diurnal variation for the
         given locations.
        Plot time series of average precipitation by month (if precipitation is an input) for one location as well as
         surface station locations. Compare with observations. Check for reasonable monthly variation for the given
         locations.
        Plot the frequency distribution of mixing heights for different locations. Check for reasonableness.
        Plot a time series of mixing heights for a 24-h summer and winter period during a light wind, and a clear sky
         period. Examine the diurnal behaviour for reasonableness.
        Although the Guideline recommends the AERMOD type dispersion option for local scale applications (on the
         order of <50 km) in CALPUFF, it is a good idea to also produce P-G stability class and plot the frequency
         distribution for the source location as well as surface station location. Compare to the airport observation P-
         G class frequency distribution (if available). Check for reasonableness for the given locations.
        If NWP model output is used, examine CALMET-generated wind fields for a 24-h period of light winds, and
         clear skies at surface, mid and upper levels with and without NWP output and check for reasonableness.

        MODEL PERFORMANCE AND UNCERTAINTY

Although a simple comparison between model predicted concentrations and observations may indicate that the
model is over or under-predicting by some factor, model predicted concentrations cannot be adjusted by this factor




Version: July 2026                                                                                              Page 79
                             British Columbia Air Quality Dispersion Modelling Guideline


to “make the results closer to reality”. Proper model evaluation requires considerable expertise in order to interpret
the performance of a model, and as such “adjustment factors” cannot be applied to the output.

Measurements and model predictions can be compared in a variety of ways, each providing a different perspective on
model performance. A model may show good competency in certain predictions (e.g., maximum concentrations) but
poor in others (e.g., the frequency of concentrations above a certain threshold). Although it is common to blame the
model for poor performance, there are other reasons that are often not considered. These include:

        Uncertainties in the input values of the known conditions (e.g., poor quality or unrepresentative
         meteorological, geophysical and source emission data).
        Errors in the measured concentrations that are used to compare with model predictions.
        Inadequate model physics and formulation.

The sources of uncertainty in the model predictions listed above are considered “reducible”. That is, they can be
controlled or minimized by collecting the proper input data, preparing the input files correctly, checking and re-
checking for errors, correcting for odd model behaviour, ensuring that the errors in the measured data are minimized
and applying better model physics. However, even if the reducible uncertainty is eliminated (i.e., perfect input with
the perfect model), there is always “inherent uncertainty” in model predictions that cannot be eliminated.

Dispersion models use known conditions (as defined by wind speed, stability, emission rate, etc.) to produce
concentrations for this specific event. Even though much is known about the processes involved in dispersion, there
are unknown conditions (turbulent processes that are not completely known and cannot be resolved by
measurements and known science). These unknown conditions may vary even under seemingly identical repetitions
of an event (under what seems to be exactly the same wind speed, stability, emission rate, etc.) and can result in
differences between the model prediction and observations for that event. Such differences could be due to “inherent
uncertainty”, something that would occur even if the inputs, models and modellers were all “perfect”. See EPA (2024)
for a more complete rendering of this concept.

The U.S. EPA has developed the following general statements on model performance (US Environmental Protection
Agency, 2024). Such statements must be considered in their totality (i.e., quoting one of the statements in isolation of
the others does not provide a complete picture).

        Models are more reliable for estimating longer time-averaged concentrations than for estimating short-term
         concentrations at specific locations.
        The models are reasonably reliable in estimating the magnitude of highest concentrations occurring
         sometime, somewhere in the area. For example, errors in highest estimated concentrations of ± 10 to 40%
         are found to be typical (assuming appropriate inputs).
        Estimates of concentrations that occur at a specific time and site are poorly correlated with actual observed
         concentrations (paired in space and time) and are much less reliable.
        The above poor correlations between paired concentrations at fixed stations may be due to reducible
         uncertainties (i.e., error in plume location due to input wind direction error can result in large differences) or
         un-quantified inherent uncertainties. Such uncertainties (which can be on the order of 50% for the maximum
         concentrations) do not indicate that an estimated concentration does not occur, only that the precise time
         and locations are in doubt.

It is recognized that model performance will vary depending on the application. For example, in some cases a model
may consistently over-predict, and in others, it may consistently under-predict. Although this behaviour may change
depending on the application specifics, it is not necessary to conduct a full model performance evaluation every time



Version: July 2026                                                                                            Page 80
                            British Columbia Air Quality Dispersion Modelling Guideline


a model is used. To do so properly involves considerable effort and regulatory agencies (including the Ministry)
assume the results of carefully controlled tests conducted and/or reviewed by the U.S. EPA provide insight on how
well they will perform in a variety of other applications.

The goal of the Guideline is to minimize the reducible error by providing direction on input data, the appropriate
choice and application of models, and directions on preparing model input files and settings. There is considerable
reliance on the training and experience of the modeller to ensure each of these steps is done properly. If the guidance
provided here (i.e., the 11 Steps to Good Modelling Practice) is followed and there are no errors in the selection, set
up and application of the model, model performance is expected to be similar to the performance under U.S. EPA test
situations. Under these conditions, the predictions from models can be viewed as the “best estimate” available to
inform decision making.




Version: July 2026                                                                                        Page 81
                                British Columbia Air Quality Dispersion Modelling Guideline



     SPECIAL TOPICS

        FLARING

The flaring of sour gas (i.e., gas streams with H2S) results in emissions of SO2. Continuous flaring with relatively
constant emissions and can be treated by models as a continuous point source. However, flaring can be short-term
with variable emissions over hourly or sub-hourly periods and as such adjustments are required to account for these
situations. The following sections provide modelling guidance for these different types of flaring.


         FLARE SOURCE PARAMETERS

The Guideline Screening model AERSCREEN has an explicit treatment of flare stacks. AERSCREEN requires:

        emission rate of SO2

        flare stack height

        total heat release rate

        radiative heat loss fraction

The use of 55% of the radiative heat loss fraction is recommended although there is evidence (Guigard, Kindzierski, &
Harper, 2002) that this value for these types of flares is very conservative.

The Guideline Refined model AERMOD does not have an explicit treatment of flares. Instead, flares can be modelled
as point sources with modified stack parameters (i.e. pseudo-stack parameters) which are selected to ensure the
correct plume rise is calculated. Use the following formulas to compute the effective stack height and effective stack
diameter:

        𝐻𝑒 = 𝐻𝑠 + (4.56 ∗ 10−3 ) ∗ 𝑄0.478

                              𝑇 ∗𝑄∗(1−𝐹)
                     𝑠
         𝐷𝑒 = 0.0122√𝑔∗(𝑇 −𝑇 )∗𝑉 𝑠   𝑎   𝑠


Where: He = effective stack height (m)
         De = effective stack diameter (m)
         Hs = physical stack height (m)
         g = acceleration due to gravity = 9.81 m/s2
         Q = total heat release (cal/s)
         F = radiative heat loss fraction (%)
         Vs = effective stack exit velocity (m/s)
         Ts = effective stack exit temperature (K)
         Ta = ambient temperature (K)

The following is recommended for calculation of pseudo-stack parameters for AERMOD:

        Obtain the total flare heat release, Q (in cal/s), by summing the heat of combustion of the individual flared
         gas components based on the volume flared in one second.



Version: July 2026                                                                                          Page 82
                            British Columbia Air Quality Dispersion Modelling Guideline



       Set effective stack exit velocity to 20 m/s. Contact the Ministry for a different value.

       Set effective stack exit temperature to 1273 K. Contact the Ministry for a different value.

       Ambient temperature is typically set to 293 K. Contract the Ministry for a different value.

       Set radiative heat loss fraction to 55%.

Although the latest version of CALPUFF (version 7.0.0) has a flare source type, it is recommended that flares be
modelled by CALPUFF as point sources with those pseudo-stack parameters which are calculated by the formulas
above.


        INTERMITTENT EMISSION: ABNORMAL (UNSCHEDULED) FLARING

A gas processing facility can flare source gas under abnormal operations: by-pass flaring for plant safety during
process upsets, maintenance flaring, or flaring under start-up or shut down conditions. Although such emissions are a
relatively rare occurrence, they are an expected part of the operation of a gas processing facility and as such it is
important to understand the air quality implications of these different abnormal emission scenarios. Although the
probability of abnormal flaring occurring at the same time as poor dispersion conditions may be very small, modelling
the worst-case air quality scenario is recommended to show due diligence to protect the environment.

The following is recommended for situations where the emissions associated with these situations can be quantified:

       Establish the realistic emission scenarios associated with abnormal emissions (based on the
        physics/chemistry of the process) and their anticipated duration and frequency of occurrence based on the
        potential abnormal operation scenarios (example: start up, shut down).

       Apply AERSCREEN to determine the potential worst-case concentration for these scenarios.

       If AERSCREEN-predicted maximum 1-h SO2 concentrations exceed 450 µg/m3, additional refined dispersion
        modelling (AERMOD or CALPUFF) is required.

       For refined modelling, the following is recommended:

             o   If the flaring is relatively constant during the flaring period, this type of flaring should be modelled
                 as a continuous source with a representative, constant emission rate and at least one year of
                 meteorological data.

             o   If the flaring is not constant during the flaring period, then follow the steps in the next section for
                 transient emission periods.

       If the predicted 1-hr SO2 concentrations exceed 450 µg/m3, then either:

             o   The heating value should be adjusted by adding fuel gas until the ambient guideline is met, or

             o   Indicate the potential for exceedances by assessing the probability of the meteorological conditions
                 associated with the predicted worst-case air quality. If this is a planned event (such as scheduled
                 maintenance) then the period of flaring can be based on avoiding the worst-case meteorological
                 conditions or during periods when the likelihood of such conditions is extremely low.




Version: July 2026                                                                                            Page 83
                             British Columbia Air Quality Dispersion Modelling Guideline



         INTERMITTENT, TRANSIENT RELEASES: SCHEDULED PIPELINE BLOW -DOWN FLARING

There are situations when flaring occurs under controlled or scheduled situations such as pipeline blow-down flaring.
In these situations a flare stack may be used to depressurize a pipeline (for such scheduled activities such as pipeline
repair, testing, abandonment, cleaning) and the flow rate of gas to the flare will be transient (i.e., it will decrease over
time). Modelling these scenarios can be difficult since parameters such as duration and flow rates will vary depending
on the nature of the event and the design of the flare. In addition, it is not clear what period of the blow-down will
result in high ground level concentrations. At the beginning the SO 2 emissions are the greatest, however the plume
rise is the highest. At the end, although the SO2 emissions are low, the plume rise is also low. Thus different periods
need to be modelled to determine which part of the blow-down period will yield the maximum ground level
concentration. In these cases the following is recommended:

        Given that the dispersion modelling is for scheduled release situations where gas volumes are known,
         calculate the temporal blow-down profile (emission vs time).

        For multi-hour transient emissions (i.e., blow-downs):

              o   Determine hourly average emission rates.

              o   Run AERSCREEN for each hourly emission segment. If predicted maximum 1-h concentrations for
                  any of those segments exceed 450 µg/m3, additional refined dispersion modelling (AERMOD or
                  CALPUFF) is required.

              o   For refined modelling, assume a continuous source with emission rates varying on an hourly basis
                  and at least one year of meteorological data.

        For transient emissions (i.e., blow-downs) periods less than one hour:

              o   Determine the duration (in minute) of the blow-down.

              o   For screening modelling, follow the recommendations below:

                       ▪    Divide the blow-down profile into three n-minute average emission segments.

                       ▪    Run AERSCREEN for each n-minute emission segment.

                       ▪    Multiply the predicted maximum 1-h concentration by n/60 to arrive at the adjusted 1-hr
                            concentration. If the adjusted maximum 1-h concentrations from any of those segments
                            exceed 450 µg/m3, the SO2, additional refined dispersion modelling (AERMOD or CALPUFF)
                            is required.

              o   For refined modelling, assume a continuous source with emission rates varying on a sub-hourly basis
                  and at least one year of meteorological data.

        For those emission segments where exceedances are predicted, then either:

              o   adjust the heating value by adding fuel gas until the ambient guidelines is met, or

              o   indicate the potential for exceedances by assessing the probability of the “worst-case”
                  meteorological conditions associated with these exceedances. For transient emissions that are
                  scheduled, the time of release can be based on avoiding the worst-case meteorological conditions
                  or during periods when the likelihood of such conditions is extremely low.




Version: July 2026                                                                                             Page 84
                             British Columbia Air Quality Dispersion Modelling Guideline



         WELL TEST FLARING

Based on BC Energy Regulator’s Flaring and Venting Reduction Guideline49, the operator must evaluate ambient air
quality impacts before flaring sour gas if the gas contains ≥ 1 mole percent H₂S or if the sulphur emission rate is one
tonne per day during the event.

Depending on the maximum predicted concentration, the operator must submit a Level 1 Assessment and in some
situations, a Level 2 or 3 Assessment. These are described in the following sections.


10.1.4.1 LEVEL 1 ASSESSMENT

AERSCREEN (with default heat loss fraction of 55% and user-specified surface characteristics) must be used for the
Level 1 Assessment. A Level 2 or 3 Assessment is required if:

        Well-test flaring is planned between April and September inclusive and AERSCREEN predicts an exceedance
         of 450 µg/m3.

        Well-test flaring is planned between October and March inclusive and AERSCREEN predicts an exceedance of
         900 µg/m3.

Options for reducing the air quality impact can be considered such as: reducing the flaring rate and/or increasing the
flare stack height, delay until winter (October - March) or directing the flow into a pipeline. These options can be
assessed by re-doing the Level 1 Assessment.


10.1.4.2 LEVEL 2 ASSESSMENT

AERMOD or CALPUFF must be used for the Level 2 Assessment. The area to be modelled should have a minimum
radius of 10 km from the flare site. Either polar or Cartesian coordinate systems can be used.

        If the Cartesian coordinate system is used, gridded receptors should be placed at no more than 100 m apart
         for the first 4 km from the flare, and at no more than 250 m apart (for extreme complex terrain) for
         distances from 4 to 10 km from the flare site.

        For the polar coordinate system, gridded receptors should be placed at no more than ten degree intervals
         and at no more than 100 m apart for the first 4 km, and at no more than 250 m apart for distances from 4 to
         10 km from the flare site. The source of terrain information must be specified.

If predicted maximum 1-hr SO2 concentrations exceed 450 µg/m3, the operator may wish to redesign the test
parameters (e.g., modify the flare rates, increase the stack height, and wait until winter season).


10.1.4.3 HUMAN HEALTH: ASSESSMENT, MITIGATION AND MONITORING

In order to address the risk to human health based on the Level 2 or 3 Assessments, if the maximum predicted 1-h SO2
concentration exceeds 900 µg/m3 at any residences, then:



49
  https://www.bc-er.ca/files/operations-documentation/Energy-Resource-Activity-Operations-Manual/Supporting-
Documents/flaring-and-venting-reduction-guideline.pdf


Version: July 2026                                                                                           Page 85
                             British Columbia Air Quality Dispersion Modelling Guideline



        The applicant must submit a plan to continuously monitor ambient SO 2 levels at the affected residences
         throughout the flaring operation, and

        The operator must indicate that flaring will be suspended if ambient hourly levels exceed 900 µg/m3 at the
         residential monitoring location.


10.1.4.4 VEGETATION: ASSESSMENT, MITIGATION AND MONITORING

In order to address vegetation effects, the operator must submit predictions for potential “visible foliar injury”:

        If well-test flaring is planned between April and September inclusive and the Level 2 or 3 maximum predicted
         SO2 concentration exceeds 450 µg/m3.

        If well-test flaring is planned between October and March inclusive and the Level 2 or 3 maximum predicted
         SO2 concentration exceeds 900 µg/m3.

“Visible foliar injury” is determined using the concentration versus duration equations developed by Legge (1995) (see
Table 10.1). The duration values in the Legge (1995) vegetation effects equations are based on four categories:

        start of growing season (spring)

        growing season (summer)

        dormant stage (winter)

        nighttime (spring and summer)

Table 10.1        The Legge SO2 Concentration Criteria for the Onset of Acute Visible Foliar Injury

                         Number of
                                              Apr – Jun      Jul – Sep     Apr – Sep      Oct – Mar
                     consecutive hours
                                              (daytime)     (daytime)     (nighttime)     (all hours)
                     with concentration
                                               (µg/m3)        (µg/m3)       (µg/m3)        (µg/m3)
                     above given level
                             1                  1306          1741           4724            7086
                              2                  832          1110           3025            4538
                              3                  639           852           2331            3496
                              4                  530           707           1937            2906
                              5                  459           612           1678            2517
                              6                  408           543           1493            2239
                              7                  369           491           1352            2028
                              8                  338           451           1241            1861
                              9                  313           417           1150            1725
                             10                  292           390           1075            1612
                             11                  275           366           1011            1516
                             12                  260           346            956            1434

The vegetation sensitivity to SO2 is dependent on the season and time of day that influences when vegetation is
susceptible. Table 10.1 shows that calculated hourly concentration (µg/m3) required for each group of consecutive



Version: July 2026                                                                                           Page 86
                              British Columbia Air Quality Dispersion Modelling Guideline


hours (e.g., for the daytime period in May, hourly concentrations over 459 µg/m3 for five consecutive hours can result
in visible foliar injury).

For application of the foliar damage assessment, receptors should be set at the height of the vegetation canopy in the
region of the flare. Typically, this would be a flagpole height of 15 m.

Along with this analysis, based on the predicted hourly SO 2 the operator must submit:

        Figures that show the spatial distribution of the number of hourly exceedances above concentrations of 900,
         1300, 1700, 4700 and 7000 µg/m3,

        A figure that shows the maximum concentration predicted at each receptor point, and

        Any areas where visible foliar injury is predicted to occur.

With regards to the physical environs around the flare, the operator must also:

        Submit a map showing the elevation contours and location of any residence or areas that may have people
         (i.e., camp sites, coal mines),

        The elevation and location (either lat./long. or UTM coordinates) of the flare site,

        The expected flare date, flow rate, stack height, gas composition,

        A wind rose of the meteorological data used, and

        Emission parameters (including pseudo-stack parameters) used in the modelling.

If there is a potential of acute visible foliar injury, then the operator must provide:

        The meteorological conditions to avoid during flaring, which may cause the acute visible foliar injury.

Where potential for significant vegetation damage has been predicted prior to the test, then:

        Subsequent re-modelling of the well test using the site-specific meteorological and well-test data is required.

         STAGNATION CONDITIONS

Light wind conditions can lead to high concentrations as emissions can build up in an area or transport slowly
downwind with very little mixing and impinge on the side of a hill, resulting in high concentrations in a localized zone.

Stagnation occurs when a light wind or no-wind condition (meteorological calm) and variable wind directions for these
periods persist for many hours or days. This commonly occurs in BC valleys, especially during the winter. With stagnation,
contaminant dispersion is minimal due to limited turbulence and transport of the emissions from the area (accumulation),
as well as contaminant recirculation.

If a site is subject to frequent and prolonged stagnations, straight-line Gaussian models such as AERMOD have limited
value since calms treatment methods (wind speed = 0.0) eliminate these important periods (see Section 5.8.2)
Furthermore, causality effects are not handled which are important during light wind, variable wind direction periods. In
such cases consider the following:




Version: July 2026                                                                                            Page 87
                              British Columbia Air Quality Dispersion Modelling Guideline



        Although AERMOD has a low wind speed limit of approximately 0.3 m/s which should capture many low wind
         periods, the anemometer starting threshold may be much higher, leading to frequent periods of zero wind
         speeds and many skipped hours.

        Use CALPUFF since it can treat zero wind speeds and causality explicitly.

        Use NWP output to provide wind speeds for periods where measurements indicate a calm.

        SHORE / COASTAL EFFECTS

Near the shore of a large body of water, or along the coast, complex winds and turbulence are generated as a result of
the difference in surface roughness and differential heating between land and water. A sea breeze (onshore breeze)
can develop during the day, and at night, the opposite can occur (a land breeze or offshore breeze). The localized
wind flow and turbulence can have significant effects on the dispersion of emission sources.

Coastal fumigation can occur when a tall stack located on a shoreline emits a plume towards the land into a stable (or
neutral) air regime and is then intercepted by a thermal internal boundary layer over the land (TIBL). These boundary
layers over the land create considerable convective or mechanical mixing and can generate high ground-level
concentrations as the plume is pulled towards the ground.

In situations where these shore/coastal effects are important, CALPUFF in Hybrid mode is recommended. The NWP
model output blended with observations in CALMET is an effect way to include the 3-D sea and land breeze effects in
the meteorological fields. In order to properly treat the coastal effects, overwater meteorological data are required
(air-sea temperature different, relative humidity, air temperature, and overwater mixing height). In addition, if the
effects of the TIBL on the plumes are occurring at scales smaller than the CALMET grid spacing, then a sub-grid scale
treatment can be invoked. This treatment requires the X, Y coordinates of one or more coastlines in an optional file
called COASTLN.DAT. The purpose of this file is to better resolve the relationship between the coastline and source
locations during periods conducive to onshore fumigation events. Further instructions on the treatment of coastline
effects and the required data are found in TRC (2011b).

        HORIZONTALLY ORIENTED STACKS AND STACKS WITH RAIN CAPS

If there are horizontal stacks or rain caps on a point source stack, there is effectively no vertical velocity of the effluent
although the plume may still rise due to buoyancy if the effluent is warmer than the ambient air. In the case of a rain
cap, the exit velocity may even be negative (i.e., downward oriented) and the plume will start to rise from a lower
point than the actual stack exit.

AERSCREEN and AERMOD can handle this situation explicitly through the selection of options, POINTCAP and
POINTHOR for treating capped and horizontal plumes respectively. The source parameters are input as if it were a
vertically oriented stack and the model applies adjustments internally to account for these types of orientations. For
plumes with little or essentially no buoyancy, users can specify a stack gas exit temperature = 0.0 K which
automatically sets the exit temperature to the ambient temperature.

CALPUFF can also handle these sources through the use of the adjustable vertical momentum flux factor (FMFAC) for
point sources with constant emissions which ranges from 1 (corresponding to a vertically oriented stack) to 0
(corresponding to a horizontal or capped stack with no vertical momentum). If time varying point source emissions
are applied, in the PTEMARB.DATE file, set TIDATA(7) (the vertical momentum flux) = 0.




Version: July 2026                                                                                              Page 88
                             British Columbia Air Quality Dispersion Modelling Guideline



        LIQUID STORAGE TANKS

There are two types of liquid storage tanks. One has a fixed roof where most vapours are emitted through a single
vent to top of the tank. The other type has a floating roof, where most of the emissions occur from between the wall
and the floating cover i.e., through the seal between the two.

For these two types, the following is recommended:

        For fixed roof tanks, treat the vent as a point source and consider the tank as a building where downwash
         effects are included.

        For floating roof tanks, treat the emissions as at least 8 point sources (where the emissions are distributed
         equally) positioned equidistant around the perimeter of the tank, and consider the tank as a building where
         downwash effects are included.

Note that if the emissions from these sources have no plume rise, so treating them as point sources as recommended
above means that both the momentum and buoyancy should be set to zero through the inputting of the following:

        exit velocity = 0.001 m/s

        exit diameter = 0.001 m

        exit temperature = ambient

These apply for all the models recommended in this Guideline. For AERSCREEN and AERMOD, if the stack gas exit
temperature = 0 K, the exit temperature is automatically set equal to the ambient temperature.

        PLUME CONDENSATION (FOGGING) AND ICING

For fog assessments where the distances of a condensed visible plume or where areas of potential icing need to be
identified, CALPUFF with the FOG algorithm is recommended (MFOG=1). The FOG module in CALPUFF has two modes:
the Plume mode (provides the length and height of the visible plume due to condensation) and the Receptor mode
(indicates whether there is ground based fog or ice occurring at each receptor). To assess the length and height of
visible plumes, CALPUFF is used with a single location hourly “ISCMET” meteorological data file that includes hourly
wind speed, direction, temperature, stability class, mixing height, etc.

There are two source emission preprocessors: one for cooling tower plumes (CTEMISS) and one for combustion
turbines (FGEMISS). These pre-processors create CALPUFF-ready PTEMARB.DAT variable water vapour emission files
as input to the CALPUFF-ISC fog model. In addition, these preprocessors can work only with the Schulman-Scire
building downwash algorithm.

        WET AND DRY SULPHUR AND NITROGEN DEPOSITION

CALPUFF is recommended for regional scale estimates of sulphur and nitrogen deposition. There are several chemical
transformation options available, and CALPUFF V6.42 includes the more advanced RIVAD/ISORROPIA chemical
mechanism. Based on two plume studies, Scire, et al. (2012) reports that it performs better, avoids the large over-
prediction bias in nitrate concentrations associated with the old methods, and has improved NOx and sulphate
transformation rates.




Version: July 2026                                                                                         Page 89
                              British Columbia Air Quality Dispersion Modelling Guideline


For this reason, the Guideline recommends the RIVAD/ISORROPIA mechanism (MCHEM=6) for regional scale wet and
dry deposition estimates. As such, Table 10.2 provides the corresponding options and settings for this selection.

If aqueous phase chemistry is selected (MAQCHEM=1) then MLWC=1 is recommended in order to get full 3-
dimensional, gridded cloud liquid water content from MM5/WRF through CALMET. CALMET must be set up to output
additional auxiliary “calmet.dat.aux” file.

Finally, the post processor POSTUTIL has been updated to include the ISORROPIA gas/particle partitioning scheme.
For the RIVAD/ISORROPIA chemistry treatment (MCHEM=6), then in POSTUTIL select MEQPHASE=2 (ISORROPIA
scheme for nitrate partitioning).

Table 10.2        Recommended CALPUFF Input Group 11a, 11b

          RIVAD/ISOROPPIA Chemistry Parameters (TRC Environmental Corporation, 2010) (Lawrence, 2012)

                                  Black (do not touch) Dark Grey (recommended default)
                           Light Grey (expert judgement required) White (user specified values)


 Option                                Parameter       Value      Explanation & Justification

 Ozone data input option                  MOZ            0        (0) use a monthly background ozone value
                                                                  For MOZ = 0 specify 12 ozone values for each month in
 Monthly ozone concentrations            BCKO3
                                                                  ppb (default: 12*80.0)
 Monthly ammonia data input                                       (0) use a monthly background ammonia value for all
                                         MNH3            0
 option                                                           layers
 Ammonia vertical averaging option     MAVGNH3           1        Not used if MNH3 = 0
                                                                  used only if MCHEM = 1, 3, 6, or 7 and MNH3 = 0
 Monthly ammonia concentrations         BCKNH3                    Specify 12 ammonia values for each month in ppb
                                                                  (default: 12*10.0)
 Nighttime SO2 loss rate in %/hour       RNITE1         0.2       Used for day and night for MCHEM = 6 or 7. Use default
 Nightime NOx loss rate in %/hour        RNITE2         2.0       Used only if MCHEM = 1
 Nightime HNO3 loss rate in %/hour       RNITE3         2.0       Used only if MCHEM = 1
                                                                  Used only if MCHEM = 6 or 7 and MAQCHEM = 1
                                                                  (0) monthly background H2O2 value
 H2O2 data input option                 MH2O2            0
                                                                  (1) read hourly H2O2 concentrations from the H2O2.DAT
                                                                  data file
                                                                  Used only if MAQCHEM = 1 and either MH2O2 = 0 or
 Monthly H2O2 concentrations in                                   MH2O2 = 1 and all hourly H2O2 data missing
                                        BCKH2O2
 ppb                                                              Specify 12 H2O2 values for each month in ppb (default
                                                                  12*1.)


       POINT SOURCE PLUME: SECONDARY PM 2. 5 AND O 3

Predicting secondary PM2.5 and O3 has been the exclusive domain of regional photochemical grid based models, where
models such as CMAQ are applied over large domains that include urban areas that include thousands of sources of
various types. However, for assessing the contribution of a single point source (or sources in a limited area), predicting
the secondary pollutants is problematic since the plumes from these sources cannot be resolved by the grid. As a way to



Version: July 2026                                                                                              Page 90
                              British Columbia Air Quality Dispersion Modelling Guideline


treat this, plume-in-grid (PIG) methods have been used, so sub-grid processes such as dispersion and chemical
transformation can be applied.

There are other methods that can treat point-source plume chemical transformation. For example, CALPUFF (Beginning
with Version 6.0) includes the RIVAD/ISORROPIA chemical mechanism (the same mechanism used in CMAQ), which
provides the ability to calculate secondary PM2.5 for regional scale applications.

Currently there are no methods/models that are preferred for this purpose. If this is a critical issue in the assessment,
methods should be discussed with the Ministry and included in the Dispersion Modelling Plan.




Version: July 2026                                                                                             Page 91
                              British Columbia Air Quality Dispersion Modelling Guideline



     MODEL RESOURCES

         SUMMARY OF BC MINISTRY-RECOMMENDED ACCEPTANCE ITEMS

Throughout this the Guideline, there are references made to the need for Ministry review and feedback on certain
proposed choices and actions. These are summarized here.

1.   Use of an alternate model or modifications to a Guideline model (Section 2.3.1 and 2.3.2)

2.   Dispersion modelling plan should include:

         general approach (Section 1.5)
         project description and geographic setting
         model selected (Section 2), model domain (Section 6.4.2 and 7.1), model switch settings (Section 6.4.3, 7.7
          and 7.8) and receptor grid (Section 7.2 and 7.4)
         planned model output (Section 8.3)
         method used to establish source emission rates (Section 3.3 and 3.4)
         method used to establish baseline concentrations (Section 8.1)
         building information for downwash if applicable (Section 7.6)
         the source of the geophysical data and data treatment (Section 4)
         meteorological data input sources (Section 5), including NWP model output if applicable and how they are
          used in AERMOD and CALPUFF if applicable. (Section 6.1 and 6.3)
         if applicable, the method used to determine NO2 concentrations (Section 8.2)
         if applicable, the method used for chemical transformation and its detailed inputs (e.g., ammonia, ozone,
          hydrogen peroxide concentrations, and nighttime loss and formation rates for nitrates and sulphates)
         table of particle emission size/density distribution if non-recommended particle size distributions are used
          for particle deposition (Section 3.7)
         procedures used for QA/QC of model input and output (Section 9)

3.   Any new monitoring program, ambient (Section 8.1.3) or meteorological (Section 5.1).

4.   Any well-test flaring application (Section 10.1.4)




Version: July 2026                                                                                          Page 92
                             British Columbia Air Quality Dispersion Modelling Guideline



     REFERENCES

Barclay, J., & Scire, J. (2011). Generic Guidance and Optimum Model Settings for the CALPUFF Modeling System for
          Inclusion into the “Approved Methods for the Modelling and Assessments of Air Pollutants in NSW, Australia.
          Sydney: Prepared for NSW Office of Environment and Heritage. Retrieved from
          https://www.epa.nsw.gov.au/sites/default/files/calpuffmodelguidance.pdf

Brook, J., Zhang, L., Di-Giovanni, F., & Padro, J. (1999). Description and evaluation of a model deposition velocities for
          routine estimates of air pollutant dry deposition over North America. Part I: model development.
          Atmospheric Environment, 33 , 5037-5051.

Byun, D., & Schere, K. L. (2006). Review of the Governing Equations, Computational Algorithms, and Other
         Components of the Models-3 Community Multiscale Air Quality (CMAQ) Modeling System. Applied
         Mechanics Reviews, 59, 51-77 .

Cole, H. S., & Summerhays, J. E. (1979). A Review of Techniques Available for Estimation of Short-Term NO2
          Concentrations. Journal of the Air Pollution Control Association, 29(8), 812-817.

Environmental Research and Technology. (1987). User’s Guide to the Rough Terrain Diffusion Model (RTDM). Rev. 3.20.
        ERT Document No. P-D535-585. Environmental Research and Technology Inc., Concord, MA. (Docket No. A-
        80-46, IVD-5).

Exponent. (2011). CALPUFF Modeling System Version 6 User Instructions. Retrieved from
       https://www.calpuff.org/calpuff/download/CALPUFF_Version6_UserInstructions.pdf

Guigard, S. E., Kindzierski, W. B., & Harper, N. (2002). Heat Radiation from Flares. Alberta Environment, Science and
         Technology Branch, Edmonton, Alberta.

Hanrahan, P. L. (1999). The Plume Volume Molar Ratio Method for Determining NO2/NOx Ratios in Modelling. Part II:
       Evaluation Studies. Journal of the Air and Waste Management Association, 49, 1332-1338.

Lawrence, B. (2012). Guidance Document: Plume Dispersion Modelling for Newfoundland and Labrador. Retrieved
       from https://www.gov.nl.ca/eccc/files/env-protection-science-gd-ppd-019-2.pdf

Legge, A. (1995, February 11). An Environmentally Significant Flare Event from a Sour Gas Processing Plant: Proposed
         Quantitative Definitions. Prepared for Westcoast Energy Inc.

Levelton Consultants Ltd. (2002). Research on the Applicability of Modelled Site Specific Meteorological Data to Well
         Test Flaring Assessments in BC. BC Oil and Gas Commission.

McEwan, B., & Murphy, B. (2004). Use of High Resolution Numerical Meteorological Fields with the CALPUFF
      Modelling System: An Analysis of RAMS and MC2 Fields over Kamloops B.C. B.C. Ministry of Water Land and
      Air Protection. SENES Consultants Ltd. Retrieved from https://www2.gov.bc.ca/assets/gov/environment/air-
      land-water/air/reports-pub/calpuff.pdf

Metro Vancouver. (2012). Station Information: Lower Fraser Valley Air Quality Monitoring Network. Retrieved from
        https://metrovancouver.org/services/air-quality-climate-action/Documents/lower-fraser-valley-air-quality-
        monitoring-network-station-information-2012.pdf




Version: July 2026                                                                                            Page 93
                             British Columbia Air Quality Dispersion Modelling Guideline


New South Wales Department of Environment and Conservation. (2005). Approved Methods for the Modelling and
       Assessment of Air Pollutants in New South Wales. Retrieved from
       https://www.epa.nsw.gov.au/sites/default/files/ammodelling05361.pdf

Oke, T. (1987). Boundary Layer Climates. London: University Press, Cambridge.

RWDI. (2005). Roberts Bank Container Expansion Project, Deltaport Third Berth. Technical Volume 8, Air Quality Health
        Risk Assessment, Appendix D (CALPUFF Dispersion Model).

Scire, J. S., Strimaitis, D. G., & Yamartino, B. J. (2000). A User’s Guide for the CALPUFF Dispersion Model (Version 5).
           Retrieved from https://calpuff.org/calpuff/download/CALPUFF_UsersGuide.pdf

Scire, J. S., Strimaitus, D. G., & Wu, Z. X. (2012). New Developments and Evaluations of the CALPUFF Model. Retrieved
           from
           https://gaftp.epa.gov/Air/aqmg/SCRAM/conferences/2012_10th_Conference_On_Air_Quality_Modeling/Pre
           sentations/3-5-CALPUFF_Improvements_Final.pdf

TRC Environmental Corporation. (2010). CALPUFF Chemistry Updates: User's Instructions for API Chemistry Options.
        Retrieved from
        https://www.calpuff.org/calpuff/download/Mod64_Files/UsersInstructions_UpdatedChemistry.pdf

US Environmental Protection Agency. (2000). Meteorological Monitoring Guidance for Regulatory Modeling
        Applications. Retrieved from https://www.epa.gov/sites/default/files/2020-10/documents/mmgrma_0.pdf

US Environmental Protection Agency. (2024). AERMOD Implementation Guide. Retrieved from
        https://gaftp.epa.gov/Air/aqmg/SCRAM/models/preferred/aermod/aermod_implementation_guide.pdf

US Environmental Protection Agency. (2024). Guideline on Air Quality Models. U.S. Federal Register. Retrieved from
        https://www.epa.gov/system/files/documents/2024-11/appendix_w-2024.pdf

US Environmental Protection Agency. (2024). User`s Guide for the AERMOD Meterological Preprocessor (AERMET).
        Retrieved from https://gaftp.epa.gov/Air/aqmg/SCRAM/models/met/aermet/aermet_userguide.pdf

Zhang, L., Brook, J. R., & Vet, R. (2003). A revised parameterization for gaseous dry deposition in air-quality models.
         Atmospheric Chemistry and Physics, 33, 2067-2082.

Zhang, L., Moran, M. D., Makar, P. A., Brook, J. R., & Gong, S. (2002). Modelling gaseous dry deposition in AURAMS: a
         unified regional air-quality modeling system. Atmospheric Environment, 36, 537-560.




Version: July 2026                                                                                             Page 94
                             British Columbia Air Quality Dispersion Modelling Guideline



APPENDIX A: DISPERSION MODELLING PLAN

This Dispersion Modelling Plan is available from the Ministry website.



GENERAL

Date:

Facility Name, Company, Location (Lat, Long):

Air Quality Consultant and Contact Name:

Ministry Contact Name:

Level of Assessment (1, 2 or 3) and also provide rationale for the proposed level of assessment:




Does this plan follow a modelling approach that is similar to the approach taken in a previous air quality assessment
already reviewed and accepted by the Ministry? If so, provide the project name and Ministry contact:




PROJECT DESCRIPTION AND GEOGRAPHIC SETTING

Provide an overview of the project, including process description and the purpose of the dispersion modelling study.




Provide a description of the following:

        Terrain characteristics within domain: flat terrain or complex terrain (i.e., will complex flow need to be
         considered?)
        Dominant land cover: urban, rural, industrial, agricultural, forested, rock, water, grassland



DISPERSION MODEL

Selected Dispersion Model:

        List model(s) and version to be used (see Section 2).




Version: July 2026                                                                                           Page 95
                            British Columbia Air Quality Dispersion Modelling Guideline



       Specify any non-guideline models or versions (i.e., beta-test versions) planned for use (Section 2.3.1). Provide
        rationale.



       If modifications to any of the models are planned, provide a description and the rationale (Section 2.3.2).



Default Switch Settings

       For AERMOD identify any switch settings that will be different than the recommended defaults (Section 7.7).
        Provide rationale.



       For CALMET/CALPUFF identify any key switch settings in CALMET and CALPUFF that will be different from the
        “black (do not touch)” defaults as per Tables 6.2 and 7.1. Provide rationale.



       If the CALMET model is used, provide:
        o a CALMET domain map that also shows the locations of surface meteorological stations and upper air
             stations

        o    anticipated grid resolution:________ (m)

        o    number of grids in X and Y direction (NX = ______ ,NY = _______)

        o    vertical levels (m): _____,_____,_____,_____,_____,_____,_____,_____,_____,_____,____



AERMOD and Receptors

If the AERMET/AERMOD model is used, provide the following:

       proposed receptor grid spacing (see Section 7.2)
       an AERMET/AERMOD domain map that shows the locations of surface meteorological stations, upper air
        stations and receptor grid
       anticipated sensitive receptors (see Section 7.4) and also indicate them on the domain map (if applicable)
       receptor (flagpole) height (m) (see Section 7.5)



CALPUFF and Receptors

If the CALPUFF model is used, provide the following:

       proposed receptor grid spacing (see Section 7.2)
       a map of the CALPUFF domain and receptor grid



Version: July 2026                                                                                         Page 96
                              British Columbia Air Quality Dispersion Modelling Guideline



        anticipated sensitive receptors (see Section 7.4)) and also indicate them on the CALPUFF domain map (if
         applicable)
        receptor (flagpole) height (m) (see Section 7.5)



PLANNED MODEL OUTPUT: AIR QUALITY ASSESSMENT NEEDS

Output Requirements for Model Demonstrations

What model output is required for decision makers and stakeholders? (i.e. what is the purpose of the assessment?).
Circle as appropriate.

        Air Quality: concentrations, depositions, visibility, fogging, icing, other (specify)


Tables and Figures for Level 1 Assessment:

        maximum concentration of contaminants predicted including location and corresponding meteorological
         conditions
        printout of AERSCREEN model output


Tables and Figures for Level 2 and 3 Assessments (see detailed list in Section 8.3.2):

        spatial distribution maps of air quality parameters (maximums, exceedance frequencies, annual averages)
        tables of maximum short and long term average air quality parameters (locations and associated
         meteorological conditions)
        tables of air quality parameters at select receptors of interest (maximums, frequency distributions)
        tables of air quality parameters under abnormal emission situations (upsets, start-up)
        output spatial scale: near-field (<10 km), local (<50 km), regional (>50 km)
        special output required for vegetation, health risk or visibility assessments
        other (specify):



EMISSION SOURCES AND CHARACTERISTICS

Provide a map showing the source locations, buildings, and facility fence line.

Model Emission Scenarios

        If applicable, describe the different model emission scenarios required for the assessment if multiple options
         are under consideration. For example, different source characteristics (stack dimensions, emission rates),
         averaging periods (e.g., 24-hr, annual), or source arrangements (locations, types, buildings) may need
         separate modelling runs to examine the air quality implications of different scenarios.



        If separate modelling scenarios are presented to evaluate alternative control technologies or mitigation
         measures, provide a detailed description of each scenario, including the derivation of the assumed control



Version: July 2026                                                                                         Page 97
                             British Columbia Air Quality Dispersion Modelling Guideline


          efficiencies (Section 3 and 4.1, GUIDANCE FOR DISPERSION MODELLING OF MINING-RELATED FUGITIVE
          PARTICULATE MATTER EMISSIONS IN BRITISH COLUMBIA).



         For mining-related fugitive PM assessments, describe the representative and worst-case operating scenarios,
          as applicable, including operating restrictions, mutually exclusive activities, moving sources, applicable
          controls, and natural mitigation assumptions (Sections 3, GUIDANCE FOR DISPERSION MODELLING OF
          MINING-RELATED FUGITIVE PARTICULATE MATTER EMISSIONS IN BRITISH COLUMBIA).



Contaminants Emitted for Each Emission Scenario

Provide the following details of the sources to be modelled:

                           Specify Source, Type, Contaminants (extend Table as necessary)

 Source      Type:                       Contaminants      Basis of Emissions (Section 3.3)
             Point (P), Area (A),        (SO2, NO2,
             Line (L), Volume(V), etc.   PM2.5*. . .)
             Indicate type
                                                           ___approved/proposed emission limits
                                                           ___manufacturer specifications
                                                           ___emission factors
                                                           ___CEM
                                                           ___modelled emission rates
                                                           ___stack sample
                                                           ___other (specify)
                                                           ___approved/proposed emission limits
                                                           ___manufacturer specifications
                                                           ___emission factors
                                                           ___CEM
                                                           ___modelled emission rates
                                                           ___stack sample
                                                           ___other (specify)
                                                           ___approved/proposed emission limits
                                                           ___manufacturer specifications
                                                           ___emission factors
                                                           ___CEM
                                                           ___modelled emission rates
                                                           ___stack sample
                                                           ___other (specify)

* for PM emissions indicate whether it is filterable, or filterable + condensable, or if unknown (see Section 3.6)




Version: July 2026                                                                                           Page 98
                             British Columbia Air Quality Dispersion Modelling Guideline


For any mining-related fugitive PM source not addressed in the GUIDANCE FOR DISPERSION MODELLING OF MINING-
RELATED FUGITIVE PARTICULATE MATTER EMISSIONS IN BRITISH COLUMBIA, describe:

        the source and justify the model representation and assumptions used (Section 2.3, GUIDANCE FOR
         DISPERSION MODELLING OF MINING-RELATED FUGITIVE PARTICULATE MATTER EMISSIONS IN BRITISH
         COLUMBIA).




For mining-related blasting emissions, describe:

        the method used to convert emissions per blast to short-term and annual-average model emission rates for
         TSP, PM10, and PM2.5 (Section 2.1, GUIDANCE FOR DISPERSION MODELLING OF MINING-RELATED FUGITIVE
         PARTICULATE MATTER EMISSIONS IN BRITISH COLUMBIA).




        the method used to determine release height and initial dimensions, including key assumptions and the
         treatment of multiple blast holes (Section 2.3.3, GUIDANCE FOR DISPERSION MODELLING OF MINING-
         RELATED FUGITIVE PARTICULATE MATTER EMISSIONS IN BRITISH COLUMBIA).




For mining-related fugitive PM emissions, provide supporting information to justify any proposed control efficiencies
(e.g., those for unpaved roads) that are not specified in the GUIDANCE FOR ESTIMATING FUGITIVE PARTICULATE
MATTER EMISSIONS FROM MINING ACTIVITIES IN BRITISH COLUMBIA. Acceptable justification may include one or
more of the following:

        recognized guidance documents, standards, or technical literature
        manufacturer-guaranteed control efficiencies or design specifications
        site-specific measurement data (e.g., measurements collected before and after implementation of the
         control technology)
        engineering calculations that demonstrate the expected emission reductions




Sensitivity Analysis for Source Location

Where a source may occur at different locations, describe the proposed approach for identifying the worst-case
location. For blasting at discrete locations within a larger operating area, this should include a source-location
sensitivity analysis for the applicable pollutants and averaging periods (Sections 2.3.3, GUIDANCE FOR DISPERSION
MODELLING OF MINING-RELATED FUGITIVE PARTICULATE MATTER EMISSIONS IN BRITISH COLUMBIA)



Version: July 2026                                                                                        Page 99
                            British Columbia Air Quality Dispersion Modelling Guideline


Source Emission Rate Variability

Do emissions have sub-hourly or sub-daily variation (e.g., blow-down flares with high emission peaks during the hour,
quarry operations that only take place during a specific range of hours in a day, wind erosion from storage piles and
exposed areas)? If so, describe the approach to assess air quality implications of those sub-hourly or sub-daily high
emission peaks.



For wind erosion from storage piles and exposed areas, describe the wind-speed and disturbance-frequency
assumptions. If CALPUFF's six wind-speed categories are used, explain how the emission rate for each PM fraction (i.e.,
TSP, PM₁₀, and PM₂.₅) and the emission-rate scaling factor for each wind-speed category will be derived. (Sections 2.1
and 2.3.4, GUIDANCE FOR DISPERSION MODELLING OF MINING-RELATED FUGITIVE PARTICULATE MATTER EMISSIONS
IN BRITISH COLUMBIA).




Describe the approach to assess air quality implications under the 25, 50, 75% emission scenario. See Section 3.4.2.




If there are batch processes, provide a temporal emission profile (emission rate vs time) for each batch process.




Describe anticipated abnormal emission scenarios (e.g., start-up, shut-down, maintenance of control works) and their
anticipated frequency of occurrence. See Section 3.4.3.




BASELINE CONCENTRATION

       Indicate method used to determine baseline concentrations for each pollutant (Section 8.1):

        _____monitoring data (Section 8.1.1 and 8.1.2)

        _____establish monitoring program (Section 8.1.3)

        _____modelled sources (Section 8.1.5)

        _____other method (describe)




Version: July 2026                                                                                       Page 100
                           British Columbia Air Quality Dispersion Modelling Guideline



       If existing monitoring data to be used, complete the following Table:

                                     Representative Air Quality Measurements

 Station Name (Lat./Long./ or          Period of Record (start/end          Contaminants Measured
 indicate on map)                      date)




       If baseline concentrations are anticipated to change in the future due to planned significant reductions or
        increases in emissions, provide a description of how these will be accounted for (e.g., construction of a
        nearby new facility or the planned decommissioning of a currently operating facility) and the uncertainties
        involved in estimating future emissions.



       For NO2 models, provide a description of how NO2 chemistry, location and proximity of urban regions relative
        to the modelled source, and proximity of nearby large industrial or transportation sources of NOX are
        considered when selecting the baseline dataset (Section 3.3.2, GUIDANCE FOR NO2 DISPERSION MODELLING
        IN BRITISH COLUMBIA).



       For NO2 models, if refined baseline options are proposed (Section 3.3.2, GUIDANCE FOR NO2 DISPERSION
        MODELLING IN BRITISH COLUMBIA), show the baseline value(s) in the form of each sequential step (e.g.,
        show the 98th percentile of daily 1-hour maximum and the 98th percentile of monthly hour-of-day values if
        proposing to use the Monthly Hour-of-Day option).



       If the Monte Carlo method (Section 3.3.2.1, GUIDANCE FOR NO2 DISPERSION MODELLING IN BRITISH
        COLUMBIA) is applied for NO2 baseline, submit the computer code used to generate results.




Version: July 2026                                                                                      Page 101
                          British Columbia Air Quality Dispersion Modelling Guideline



      For models with mining-related fugitive particulate matter emission sources, specify:
           o The TSP, PM10, and/or PM2.5 monitoring dataset used, if applicable, and why it is representative of
              the facility site (Sections 2.7.2, GUIDANCE FOR DISPERSION MODELLING OF MINING-RELATED
              FUGITIVE PARTICULATE MATTER EMISSIONS IN BRITISH COLUMBIA).
           o The method used to determine the TSP, PM 10, and/or PM2.5 baseline concentrations and combine
              them with modelled concentrations. Provide rationale for any alternative method used to
              determine baseline or total concentrations (Sections 2.7.2 and 2.7.3, GUIDANCE FOR DISPERSION
              MODELLING OF MINING-RELATED FUGITIVE PARTICULATE MATTER EMISSIONS IN BRITISH
              COLUMBIA).



BUILDING DOWNWASH

      Potential for building downwash. Please provide rationale if building downwash is not modelled.



      If building downwash included, provide a site map to indicate buildings to be processed by BPIP-PRIME, and
       complete the following Table:

  Source Height        Distance from the      Building Length      Building Height          Building Width
  (m)                  Source to the          (m)                  (m)                      (m)
                       Nearest Building
                       (m)




GEOPHYSICAL DATA INPUT

Topography and Land Use Data

      Terrain data (specify source of data) and an elevation map for the model domain:




Version: July 2026                                                                                   Page 102
                              British Columbia Air Quality Dispersion Modelling Guideline



          Land use data (specify source of data) and a land use map for the model domain:




Surface Characteristics

For AERSCREEN, provide seasonal values of surface characteristics (surface roughness, albedo and Bowen ratio) for
input to MAKEMET.



For Level 2 and 3 Assessments, indicate if recommended seasonally varied surface characteristics (surface roughness,
albedo, Bowen ratio, etc.) (see Section 4.3 and 4.4) are used for the dispersion modelling study. If not, provide the
proposed surface characteristics and the rationale.



METEOROLOGICAL DATA INPUT (FOR LEVEL 2 AND 3 ASSESSMENTS ONLY)

Surface Meteorological Data

If surface observation data are used, provide a map with the location of each surface meteorological station identified
and also provide the following:


     Surface Met Data and    Data Source           Period of Record     % of Wind    Anemometer       Parameters
     Location (lat/long or   The Ministry, MV,     (start/end data) 2   Speeds =     Height (m)
     indicate on map)        MSC, Site Specific,                        0.0 3
                             Other (specify) 1




1. If data from a non - Ministry, MV or MSC station are planned to be used, follow guidance in Section 5.2.3
2.
   For data completeness and data filling, follow guidance in Section 5.5
3.
   For light and no wind conditions, follow guidance in Section 5.6




Version: July 2026                                                                                       Page 103
                                British Columbia Air Quality Dispersion Modelling Guideline


Upper-Air Meteorological Data

If upper air meteorological data are used provide the following:


Station         Period of Record (start/end date) 1                     Distance between the Upper Air Station
Name                                                                    and Project (km)




1.
     For data completeness and data filling, follow guidance in Section 5.5.

NWP Model Output

If NWP output (different than the provincewide WRF output) used provide the following:

           Mesoscale Meteorological Model (Name\Version\Model Configuration):
           Model Output Provider:
           Domain (attach a map showing the horizontal extent):
           Horizontal and Vertical Grid Resolution and Height of Each Vertical Level:
           Data Period (start/end date):
           Four Dimensional Data Assimilation is applied (Yes or No):


NWP model output use (circle one below for the selected dispersion model):

           AERMET/AERMOD:
            o Extract pseudo surface station and pseudo upper air sounding (as input to AERMET), or
            o Create .SFC and .PFL files (AERMOD-ready files, skip AERMET)

           CALMET:
            o NWP only, or
            o Surface station and NWP, or
            o Surface station, upper air sounding, and NWP, or
            o Other (specify):




Version: July 2026                                                                                        Page 104
                            British Columbia Air Quality Dispersion Modelling Guideline


TREATMENTS

NO to NO2 Conversion (Section 3.2, GUIDANCE FOR NO2 DISPERSION MODELLING IN BRITISH COLUMBIA)

Identify the method to be used. Please note that the results of total conversion must be presented as part of all model
reports, regardless of the conversion method selected for the project.



_____Total Conversion

_____Ambient Ratio Method

       Indicate which NO/NO2 dataset is used for the ARM2 curve (AERMOD screening ARM2 curve, BC ENV-
        developed category curve, or single site representative of project site) and explain the basis for selecting the
        dataset.
             o If a single site dataset is used, provide the dataset and completeness statistics (e.g., number of years,
                 percent complete per quarter).
       If CALPOST is used, provide the 24 values used for the step function.

_____OLM:

       Indicate which O3 dataset is used and explain the basis for selecting the O 3 dataset.
             o If a single site representative hourly O3 dataset corresponding to the meteorological period is used,
                  specify the method of data substitution used for addressing data gaps, provide the dataset, and
                  include the completeness statistics (e.g., number of years, percent complete per quarter).
       If non default equilibrium ratios are used, specify and provide rationale.
       Specify and provide rationale for in-stack ratio(s) used. If multiple NOX sources are modelled, provide
        justification for how the ISR(s) is/are selected.

_____PVMRM (for AERSCREEN and AERMOD only):

       Indicate which O3 dataset is used and explain the basis for selecting the O 3 dataset.
             o If a single site representative hourly O3 dataset corresponding to the meteorological period is used,
                 specify the method of data substitution used for addressing data gaps, provide the dataset, and
                 include the completeness statistics (e.g., number of years, percent complete per quarter).
       If non default equilibrium ratios are used, specify and provide rationale.
       Specify and provide rationale for each in-stack ratio used.



Specify the considerations given to ambient concentrations, characteristics of modelled sources, and availability of
relevant monitoring data when selecting the NO2 modelling method indicated above.




Version: July 2026                                                                                        Page 105
                              British Columbia Air Quality Dispersion Modelling Guideline


Chemical Transformation:

        Specify transformation method and provide details on inputs if secondary PM2.5, acid deposition or visibility
         effects are to be estimated. Depending on the transformation method, this could include ammonia, ozone,
         hydrogen peroxide concentrations, nighttime loss and formation rates for nitrates and sulphates.



Open Pit Retention

        If pit retention is applied in CALPUFF modelling, describe how the PM₁₀ and TSP escape fractions will be
         estimated and applied to the modelled emissions. If an alternative pit-retention method is proposed,
         describe and justify its use (Sections 2.5, GUIDANCE FOR DISPERSION MODELLING OF MINING-RELATED
         FUGITIVE PARTICULATE MATTER EMISSIONS IN BRITISH COLUMBIA).



Particle Deposition:

        If particle size distributions other than those recommended in Section 3.6 of the BRITISH COLUMBIA AIR
         QUALITY DISPERSION MODELLING GUIDELINE are proposed, provide the particle-size ranges, geometric
         mean diameters, mass fractions, and the basis for the selected values. Where particle density is used to
         derive effective CALPUFF particle-size parameters for heavy metals, provide and justify the assumed density.



        If AERMOD is proposed for mining-related fugitive PM sources, provide and justify the particle-deposition
         parameters for each applicable source or source group (Appendix A, GUIDANCE FOR DISPERSION
         MODELLING OF MINING-RELATED FUGITIVE PARTICULATE MATTER EMISSIONS IN BRITISH COLUMBIA).



Stagnation:

        Provide an estimate of the frequency of stagnation based on local meteorological data if available. If
         AERMOD is proposed, provide methodology on how stagnation periods will be treated (see Section 10.2).



Shore/Coastal Effects:

       If included, indicate whether sub-grid-scale Thermal Internal Boundary Layer option is selected along with
        the required input coastline coordinate data (see Section 10.3).



Plume Condensation (Fogging) and Icing:

        Indicate if this will be included (Section 10.6).




Version: July 2026                                                                                        Page 106
                             British Columbia Air Quality Dispersion Modelling Guideline


QUALITY MANAGEMENT PROGRAM

Model Input Data

Indicate the tests that will be undertaken to assure the quality of the inputs.



For the geophysical input data:

        contour plot of topography
        plots of land use and land cover

For the meteorological data:

        wind rose (annual and/or seasonal)
        frequency distribution of surface wind speeds
        average hourly temperature plot (annual and/or seasonal)

If NWP output is used, describe the tests undertaken to assure the quality of the output (Section 6.1)

        wind rose at selected locations and heights (annual and/or seasonal)
        average hourly temperature plot at selected locations and heights (annual and/or seasonal)
        wind field plots for selected periods that indicate topographic influences such as channeling and thermally
         generated flows



Model Output Data

For CALMET/CALPUFF applications, provide a list of the tests conducted to confirm the quality of the model output
(intermediate pre-processing files and concentration/deposition predictions).

With respect to the pre-processed files that are prepared for CALPUFF input, there are several tests listed in Section
9.1.1 and 9.1.2 to check the output from the pre-processing utility programs to confirm that they have been properly
processed. These are related to checking:

        terrain, land use
        sources (locations and elevation) and emission characteristics
        meteorological data (locations) and tests to confirm proper processing of the raw meteorological data (units,
         parameters)
        receptor locations and elevations

For CALMET output there are several tests listed in Section 9.1.3 to test the quality of the generated meteorological
fields. These are related to reviewing the following:

        wind field maps (surface and different elevations) for select periods where topographic influences
         (channeling, thermally driven flows) would be evident
        wind roses at selected locations and elevations (annual, seasonal)




Version: July 2026                                                                                       Page 107
                           British Columbia Air Quality Dispersion Modelling Guideline



       frequency distributions of various meteorological parameters (annual, seasonal) such as PG-stability class
        and mixing heights
       plots of hourly average parameters such as temperature, mixing height, and precipitation at key locations
        (seasonal and annual)



Model Performance Evaluation

For Level 3 Assessments, indicate whether an assessment of model performance will be conducted as quality
assurance for the project (Section 4.3, GUIDANCE FOR NO2 DISPERSION MODELLING IN BRITISH COLUMBIA). If not,
provide rationale.



Note: The Ministry may require all computer files associated with the modelling to be submitted upon request.

If any of the advanced processing methods like Monte Carlo are applied, the computer code used to generate results
should also be included in the submittal.



MINISTRY REVIEW OF PLAN AND REVISIONS

A modelling plan can change over the course of developing the air quality assessment so acceptance of the initial
submission of the plan is based on the best information provided to date. Changes to the plan (additions,
modifications) should be noted and agreed to with the Ministry as necessary. An updated Dispersion Modelling Plan
may be necessary.

Ministry Acceptance of Original Plan (Name):_________________________

Date:____________________




Version: July 2026                                                                                      Page 108
