Loading...
EOSDIS provides access to a wide range of Earth observation datasets, including satellite imagery, climate data, land cover, oceanography and atmospheric data.
NASA Earthdata authentication is required. Engineers use the `earthaccess` Python library for cloud-optimized S3 access to EOSDIS datasets without downloading. `xarray` and `netCDF4` process the HDF5/NetCDF scientific data formats into analysis-ready multi-dimensional arrays.
NASA EOSDIS datasets are the backbone of Earth observation AI. Train deep learning models for cloud cover segmentation, fire detection, and flood mapping using NASA satellite imagery. The breadth of sensors and time coverage (1970s to present) enables AI climate change attribution studies at global scale.
# pip install earthaccess xarray
import earthaccess
# Authenticate with NASA Earthdata credentials
earthaccess.login(strategy="environment") # set EARTHDATA_USERNAME / PASSWORD
results = earthaccess.search_data(
short_name="MOD11A1", # MODIS Land Surface Temperature
temporal=("2024-01-01", "2024-01-07"),
bounding_box=(-10, 35, 30, 60) # Europe
)
print(f"Found {len(results)} granules")
files = earthaccess.open(results[:2])Official dataset source
More datasets used by Python data engineers.
NOAA platform provides access to a vast collection of climate-related datasets, including historical weather data, climate observations, satellite imagery and climate model outputs.
The NCEI, part of NOAA, provides access to a wide range of environmental datasets, including climate data, weather observations, oceanographic data and geophysical data.
GLDAS provides datasets on land surface conditions, including soil moisture, temperature, precipitation and other hydrological variables, derived from satellite and ground-based observations.