The author developed a habitat suitability model to locate Boletus pinophilus (pine bolete) in the Leningrad Region. The methodology relies on the correlation between pine forests and sandy, acidic soils, specifically illuvial-ferruginous podzols.
Data sources included the 1988 soil map of the RSFSR (scale 1:2,500,000), the Copernicus Global Land Service (CGLS) forest cover layer, and OpenStreetMap. The author used a custom scraper based on the Chrome DevTools Protocol to extract soil data and processed GeoTIFF files via GDAL and rasterio.
The algorithm generated 3,265 potential zones covering 10,240 square kilometers. By applying a 60% sand content threshold from the SoilGrids dataset, the model identified 1,048 priority 'pine' zones totaling 4,348 square kilometers for field verification.
For mobile field use, the author created a web interface using OpenLayers. This engine was selected for its ability to dynamically transform projections, resolving a 18-kilometer coordinate shift between Web Mercator and Yandex map tiles at the 60th parallel.
The final pipeline includes automated testing and deployment via GitHub Actions. To optimize performance on mobile networks, the author used the mapshaper utility to tile large GeoJSON files into smaller, manageable segments.
