sparta-solar: Clear-sky Solar Irradiance with the SPARTA Radiative Transfer Model

A Python library to compute clear-sky solar irradiance at the ground surface using the Solar PArameterization of the Radiative Transfer of the Atmosphere (SPARTA) radiative transfer model, with built-in access to multiple atmospheric databases (Copernicus CAMS Radiative Service via SODA, hourly NASA MERRA-2 data via Google Earth Engine, and a daily dataset curated and maintained as part of the sparta-solar library on a dedicated Hugging Face dataset).

Licence: sparta-solar is distributed under CC BY-NC-SA 4.0. It is free to use, share and adapt for research and other non-commercial purposes, with attribution and under the same licence. Commercial use is not permitted without a separate agreement with the author.

Quick install

pip install sparta-solar

or with uv:

uv add sparta-solar

Note that the distribution is named sparta-solar, but the Python package is imported as spartasolar:

import spartasolar

Quick examples

At individual sites:

import pandas as pd
from spartasolar.atmosphere import merra2_daily

times  = pd.date_range("2020-06-15", periods=24, freq="h")
atmos  = merra2_daily.at_sites(
    times=times,
    latitude=36.72,  # or a sequence of latitudes...
    longitude=-4.42)  # ... and longitudes
result = atmos.compute()
print(result)

On a regular spatial grid:

import numpy as np

lats = np.arange(-60, 60.1, 1)
lons = np.arange(-90, 90.1, 1)
atmos  = merra2_daily.on_regular_grid(
    times=times,
    latitude=lats,
    longitude=lons)
result = atmos.compute()
print(result)

Documentation

Full documentation — installation guide, user guide, quick reference, and API reference — is available at https://jararias.github.io/sparta-solar

Citation

@article{ruizarias2023sparta,
  author  = {Ruiz-Arias, Jose A.},
  title   = {{SPARTA}: Solar parameterization for the radiative transfer of the cloudless atmosphere},
  journal = {Renewable and Sustainable Energy Reviews},
  volume  = {188},
  pages   = {113833},
  year    = {2023},
  doi     = {10.1016/j.rser.2023.113833}
}

License

CC BY-NC-SA 4.0 — free for research and other non-commercial use, with attribution and under the same licence. Commercial use is not permitted without a separate agreement with the author.