MERRA-2 FAQ: How can I subset and download a large amount of MERRA-2 data?

Use this Forum to find information on, or ask a question about, NASA Earth Science data.
Post Reply
GES DISCx - jimacker
Posts: 354
Joined: Mon Sep 30, 2019 1:57 pm America/New_York
Answers: 3
Has thanked: 2 times
Been thanked: 9 times

MERRA-2 FAQ: How can I subset and download a large amount of MERRA-2 data?

by GES DISCx - jimacker » Fri Jul 05, 2024 3:48 pm America/New_York

The GES DISC provides multiple services for subsetting and downloading MERRA-2 data on the Product Landing Page (see an example for M2T1NXSLV) for each data collection. Shown below are recommended methods:

If the user is interested in subsetting variables over a region:

Product landing page ---> Subset/Get Data ---> subsets using OPeNDAP
Output format: netCDF, ASCII
Characteristics: better performance

If the user is interested in subsetting variables over a region, plus the time of a day or vertical levels, or getting preprocessed data:

Product landing page ---> Subset/Get Data ---> GES DISC Subsetter
Output format: netCDF
Preprocessing options:
daily mean, minimum and maximum
horizontal regridding


How-to documents describing the use of the GES DISC Subsetter service can be found here:

How to use the Level 3 and 4 Subsetter and Regridder

How to Download MERRA-2 Daily Mean Data

If the user is interested in time series, please read the FAQ:

How do I obtain a time series for MERRA-2 data ?

If a user is interested in automating the data download process, here are some examples with sample Python scripts:

How to Use the Web Services API for Subsetting MERRA-2 Data

How to Use the Web Services API for Dataset Searching

How to Use the Web Services API for Subsetting


Additional information:
Please register and link to the GES DISC data system. Additional steps may be required ( particular the first time users) via using wget and other tools: Click here

Tips for successfully downloading a large amount of data files with "wget" :

reduce concurrent connections, by limiting the concurrent connections to 5.

try to add wget options

-t number ( or --tries=number ) : set number of tries

-w seconds ( or --waitretry=seconds ): wait the specified number of seconds between the retrievals of failed downloads

In case of missing some files due to various reasons, please try to download again with the option

-m (or -- mirror ) : turn on mirroring option

Filters:

Post Reply