merradownload

A Python package for downloading, cleaning, and aggregating meteorological data from NASA's MERRA-2 database.

Adapted from Emily Laiken's merradownload, which builds on Jan Urbansky's opendap_download and the Open Power System Data notebook. Edited and maintained by Anemona.

Installation

pip install merradownload

Prerequisites: NASA Earthdata account

1. Register at https://urs.earthdata.nasa.gov/
2. Go to Applications → Authorized Apps → Approve More Applications
3. Approve NASA GESDISC DATA ARCHIVE

Quick start

Python API

from opendap_download.multi_processing_download import DownloadManager

dm = DownloadManager(
    username='your_earthdata_username',
    password='your_earthdata_password',
    links=['https://...'],       # list of OPeNDAP query URLs
    download_path='data/output', # where to save .nc4 files
)
dm.start_download(nr_of_threads=5)

CLI

merradownload \
  --username YOUR_USER \
  --password YOUR_PASS \
  --urls-file urls.txt \
  --output-dir data/output \
  --threads 5

urls.txt should contain one OPeNDAP query URL per line. See examples/merra_scraping.py for how to generate these URLs for a given location, year range, and variable set.

Example script

examples/merra_scraping.py shows a complete end-to-end workflow: coordinate translation → URL generation → download → xarray/pandas processing → CSV and plot outputs.

Dependencies

Processing dependencies (used in the example script only, not required by the core package):

pip install numpy pandas xarray netCDF4 matplotlib

Features

• Parallel downloads via ThreadPoolExecutor
• Automatic retry on HTTP 5xx errors (3 retries, exponential backoff)
• Skips already-downloaded files — safe to resume interrupted runs
• Progress bar
• Credential file support (yaml)