These are the tools I'd reach for when building climate software. It's not exhaustive. It's the set that covers most projects, from a weekend notebook to a production platform.
| Tool | Best for |
|---|
| Google Earth Engine | Fast exploration over huge catalogues, with compute included |
| Microsoft Planetary Computer | STAC catalogue + cloud-optimised data, Python-friendly |
| Copernicus Data Space Ecosystem | Official Sentinel data access |
| AWS Open Data | Sentinel, Landsat, and more on S3 |
See Remote Sensing with Machine Learning.
Python libraries
| Library | Use |
|---|
| xarray | Multi-dimensional climate data (NetCDF, Zarr) |
| rasterio / rioxarray | Reading and writing raster data |
| geopandas | Vector geospatial data |
| pystac-client | Searching STAC catalogues |
More in Best Python Libraries for Remote Sensing.
Weather and climate data
| Tool | Use |
|---|
| Open-Meteo | Weather forecasts and historical data via a simple API |
| Copernicus Climate Data Store | ERA5 and climate projections |
| NASA Earthdata | Precipitation, land, atmosphere products |
Emissions and energy
| Tool | Use |
|---|
| Electricity Maps | Grid carbon intensity, real-time and historical |
| Climate TRACE | Facility and sector emissions estimates |
| CodeCarbon | Estimating emissions from your own compute (AI energy) |
Geospatial app stack
| Tool | Use |
|---|
| PostGIS | Spatial database (tutorial) |
| MapLibre GL JS | Open-source web maps (dashboard guide) |
| deck.gl | Large-scale GPU visualisation (performance) |
| PMTiles / tippecanoe | Serverless vector tiles |
| TiTiler | Dynamic raster tiles from cloud-optimised GeoTIFFs |
