Research update · 5 September 2026
The project now has three documented sets of terrain and land-surface inputs covering the full study area. They combine the checked Canadian elevation data and 2020 land cover with published seasonal surface parameters. These are inputs for the next regional wind-model test; no new wind or pollutant prediction is claimed by this update.
The proposed CALMET grid has 10,201 cells at 200 m spacing in WGS84 / UTM zone 10 north. Its 20.2 × 20.2 km boundary encloses the original 20 × 20 km study square. The original boundary remains separately identified; the extra margin is not presented as part of that original square.
Each cell retains the fractions of its different land-cover classes, including small patches and mixed shorelines. Fractions come from clipping the original 30 m land-cover cells to the new grid. Using only the class at a cell centre would lose some of that information. Terrain is a representative mean of 25 samples from the original CDEM per cell, with its CGVD28 elevation reference retained.
Surface roughness uses geometric weighting. Albedo, Bowen ratio, soil heat-flux parameter, anthropogenic heat and leaf-area index use arithmetic area weighting. These follow the published MAKEGEO method; the project has not claimed numerical equivalence to a particular MAKEGEO executable. The seasonal values come from B.C.’s July 2026 guidance. B.C. guideline, §4.4 and Table 4.8
| Case | Choice being examined |
|---|---|
| Summer table | Published Table 4.8 values, with mixed forest mapped to the table’s Mixed class |
| Mixed-forest alternative | Maps mixed forest to the alternative transitional class, because the acquired map does not establish forest density or condition |
| Shrub leaf-area alternative | Uses the shrub leaf-area value in the guideline’s example, which conflicts with its main summer table |
The choices are fixed before inspecting resulting winds. The mixed-forest alternative changes surface properties in 757 cells; the shrub alternative changes leaf-area index in 5,478 cells. Those counts describe input sensitivity, not a demonstrated change in wind or health risk.
Water remains a generic category; the raster does not determine the model’s marine-temperature treatment. A cell is assigned a dominant water class when at least half its area is water. Minority water still contributes to all six weighted properties. This rule classifies 66 mixed cells as land that a simple largest-category rule would classify as water.
These are computational and source-transcription checks. They do not establish present-day land-cover accuracy, observed seasonal conditions, final grid resolution or independent professional acceptance of the model.
The six adjoining B.C. WRF tiles for June 2023 have been acquired, inventoried and assembled into one regional input. It retains 733 hourly states on 600 native points and 40 vertical levels. Reverse checking reproduced all 4,398 original tile/hour hashes, with no weather values changed. Those checks establish record integrity; they do not establish wind accuracy or successful time-varying CALMET ingestion.
The setup test used an explicitly labelled vertical-grid fixture. The next wind experiment needs a justified vertical grid, terrain-adjustment settings and treatment of overwater temperatures. Current flare geometry and emissions remain separate requirements for a site-specific concentration calculation. The summer inputs do not substitute for winter snow, inversion or outflow evidence.
The package contains corrected GEO.DAT candidates, per-cell data, source-category fractions, parameter lineage and review summaries. It is a geophysical input package, not a complete model run. Generic urban heat values are not a Woodfibre industrial or flare heat inventory. The historical land cover and candidate seasonal choices need to remain visible when interpreting later results.
Land-cover source: NRCan 2020 Land Cover of Canada, under the Open Government Licence – Canada. Elevation source: the retained NRCan CDEM 092G tile and its previously checked provenance. The CALMET/MAKEGEO format and weighting references are documented by section and page in the parameter ledger; provider manuals, program source and executables are not redistributed in this package.