DOI: 10.21105/JOSS.02975
关键词: Statistical model 、 Raster graphics 、 Python (programming language) 、 Scale (map) 、 Computer science 、 Remote sensing 、 Spatial variability 、 Random effects model 、 Deforestation (computer science) 、 Spatial analysis
摘要: The forestatrisk Python package can be used to model the spatial probability of deforestation and predict future forest cover in tropics. data come from georeferenced raster files, which very large (several gigabytes). functions available process rasters by blocks data, making calculations fast efficient. This allows modeled over geographic areas (e.g., at scale a country) high resolution _ 30 m). offers possibility using logistic regression with auto-correlated random effects process. make possible structure residual variability process, not explained variables often large. In addition these new features, is open source (GPLv3 license), cross-platform, scriptable (via Python), user-friendly (functions provided full documentation examples), easily extendable (with additional statistical models for example). has been 2100 across humid