Generate regional preciptitation, temperature, and solar radiation data that is spatially correlated.
Calibrates on NASA POWER
import datetime as dt
import geopandas as gpd
import sgen
# Regional geometry in WGS84 projection
region = gpd.read_file('region.geojson').geometry[0]
# Builds generator at 0.5 deg resolution and calibrates on NASA POWER historical archive.
generator = sgen.build_spatial_weather_generator(region)
start_date = dt.date(2020, 1, 1)
num_days = 2000
# Outputs xarray on region in WGS84 projection with daily ppt, max and min temp, and srad
# on a 0.5 deg grid
generated_weather = generator.simulate_weather(start_date, num_days)Based on the following papers by Wilks:
Multisite generalization of a daily stochastic precipitation generation model (1998) DOI: 10.1016/S0022-1694(98)00186-3
Simultaneous stochastic simulation of daily precipitation, temperature and solar radiation at multiple sites in complex terrain (1999) DOI: 10.1016/S0168-1923(99)00037-4