Result: A Platform for Large Scale Application of Remote Sensed Data to Wildland Fire Management

Spatial modelling and machine learning are powerful techniques that can be used to identify patterns within data and build complex relationships between response and predictor variables. While powerful, many of these techniques are computationally intensive and are not designed to fully leverage high performance computing resources, especially when used within a geospatial context. To fully leverage system resources, while facilitating various spatial, machine-learning, and statistical modelling workflows, we developed a Python-based processing library called raster_tools. The raster_tools library automates delayed reading and parallel processing through Dask and integrates seamlessly into popular spatial, machine learning, and visualisation libraries such as geopandas, rasterio, xarray, scikit-learn, xgboost, pygeos, shapely, matplotlib, plotly, folium, and many more. Combined, these open-source libraries provide users with free and powerful analytical capabilities that can be used at scale and that can dynamically display textual, tabular, spatial and graphical in a real-time fashion. In this paper, we will provide a brief overview of the raster_tools library and demonstrate how the described open-source stack can be used to perform GIS analyses using Jupyter-Lab in both a web and desktop environment.