PSE, the Planetary Sensing Engine, is open source
PSE fuses Earth observation data from six sources behind one API, and the code is open under Apache 2.0. One request instead of six formats.
Six sources, one model
Earth observation data is scattered. Temperature comes from ERA5, solar irradiance from the Global Solar Atlas, imagery from Sentinel-2, infrastructure maps from OpenStreetMap, weather from Open-Meteo and socioeconomic indicators from the World Bank. Each has its own format, coordinate system and timing.
PSE brings them into one common model and serves the result through a single API. Where two sources measure the same thing, it blends them by reliability and recency instead of picking one.
Every value traced to its source
Every response carries its provenance, so each data point can be followed back to where it came from and how it was transformed. The output fits the Python tools climate and Earth observation researchers already use.
Open for anyone to build on
The code is on GitHub under Apache 2.0, free to use and extend. PSE is one of the open code bases behind Northflow's Earth observation work, alongside CERES, LITHOS and climval.
