NASA Earthdata Cloud Data Access Guide

The NASA Earthdata Cloud is NASA’s cloud-based archive of Earth observations, hosted on Amazon Web Services (AWS). The NSIDC DAAC is actively migrating data products into the Earthdata Cloud.

The NSIDC Data Search tool lists which NSIDC DAAC data products are already available in the cloud.

This guide explains how to access NSIDC DAAC data in the Earthdata Cloud. You can:

Direct Cloud Access (S3) – Work with data directly in the cloud without downloading everything first.
Local Download – Download cloud-hosted data to your own computer or storage system.

You’ll also find information about customizing datasets, troubleshooting errors, and advanced workflows.

Getting Started

What is the NASA Earthdata Cloud?

The Earthdata Cloud is a cloud-based archive of NASA Earth science data. Instead of only storing data on on-premises servers, NASA is moving holdings to the cloud for:

Faster access close to computing resources.
Scalable storage for growing satellite missions.
New workflows that allow users to analyze data “in place” without downloading terabytes locally.

Learn more: FAQs About NSIDC DAAC's Earthdata Cloud Migration

Choosing the Right Access Method

Here are common ways to work with Earthdata Cloud data:

Browser Download → Easiest option for small datasets.
Command Line (wget, curl) → Efficient for batch downloads.
Python (earthaccess) → Flexible scripting, works locally or in the cloud.
Direct S3 access (Jupyter on EC2 or AWS CLI on EC2) → For cloud-native analysis without moving large files.

Tip: If you’re just getting started, try browser downloads or the earthaccess Python library. Move to direct S3/EC2 workflows when you’re ready to analyze large volumes of data.

Accounts, Authentication, and Tokens

To access NASA Earthdata Cloud data, you’ll need NASA Earthdata Login (EDL).

1. Create an account: Earthdata Login Registration.

2. Generate a token in your EDL profile settings (optional, for API use)

3. Use your login credentials with tools like:

wget / curl → prompt for username + password or use .netrc.
See: Creating a .netrc file for Earthdata Login
earthaccess Python library → handles login seamlessly.
AWS CLI with temporary credentials → needed for S3 access.
See: Accessing NSIDC DAAC S3 Data with Temporary AWS Credentials

Local Download (to Your Computer)

Choose this approach if you want to store and process data on your own computer.

Option 1: Browser-Based (No Coding Required)

If you prefer a graphical interface, use one of these web-based tools:

Earthdata Search - Filter and customize data before downloading files.
HTTPS File system (CMR Virtual Directory) - Browse folders by year/month/day and download individual files directly.

See the full guide: Downloading NSIDC DAAC Earthdata Cloud Data Using a Web Browser

Option 2: Command Line Tools

If you're comfortable with the terminal:

Use wget or curl to download known files or run batch download with HTTPS URL lists.
See: Creating a .netrc file for Earthdata Login for automated login
Create .txt files of HTTPS (or S3) URLs for efficient scripting.
See: Creating Text Files of HTTPS and S3 URLs for Earthdata Cloud Data Access
Combine the CMR Virtual Directory with wget for directory-level downloads.
PODAAC Data Subscriber – A Python-based utility that supports spatial and temporal filtering, Though designed for PODAAC, it can be adapted for NSIDC and other DAACs.

Step-by-step instructions: Downloading Data from Earthdata Cloud to Your Local Computer Using the Command Line

Option 3: Python (`earthaccess` Library)

The earthaccess Python library provides a streamlined way to:

Authenticate with your Earthdata Login.
Search across NASA DAAC collections.
Download data locally or into cloud compute environments (e.g., Jupyter on EC2)

Guide: Downloading NSIDC DAAC Earthdata Cloud Data Using earthaccess

Direct Cloud Access (S3)

Instead of downloading, you can work directly with NSIDC DAAC holdings in Amazon S3, the object storage system used by NASA Earthdata Cloud. This approach lets you stream data into memory and analyze it in place, avoiding large transfers to your local computer.

Understanding S3 Buckets and URLs

An S3 bucket is like a top-level folder.
Each dataset file is an object with a unique key (path).
You’ll see URLs like:
- HTTPS: https://data.nsidc.earthdatacloud.nasa.gov/...
- S3: s3://nsidc-cumulus-prod-protected/...

Learn more:

See: Understanding S3 Buckets and Paths for a deeper explanation of bucket structure, object keys, and how to interpret S3 vs. HTTPS URLs.
See: Accessing NSIDC DAAC S3 Data with Temporary AWS Credentials

Working in EC2

Launch an EC2 instance (Amazon virtual machine) near the data.
Use AWS CLI for direct S3 access or install JupyterLab for notebook-based workflows.

Help Articles:

Cost Management Tips

Choose smaller instance types unless you need high compute power.
Always stop or terminate instances when not in use.
Minimize egress charges by keeping analysis in the cloud instead of downloading massive datasets.

Data Customization (Harmony)

Select NSIDC DAAC datasets can be customized using NASA's Harmony Service.

Beginner: Customize data directly in Earthdata Search
Programmatic option:
- NASA Openscapes tutorial - Subset ICESat-2 and SMAP data using harmony-py.
- NSIDC Harmony API Quickstart Guide - explains how to use the API to request customized data with specific temporal and spatial filters. It covers topics such as formatting API endpoints, submitting jobs to Harmony, monitoring job status, and downloading processed data.

Troubleshooting

403/404 errors → Check credentials or token expiration.
Expired temporary credentials → Refresh via Earthdata Login.

Resources

NASA Earthdata Cloud Cookbook
General Information about the Earthdata Cloud: Earthdata Cloud Cookbook - Cheatsheets, Guides and Slides
NSIDC FAQ: Earthdata Cloud Migration
Community forums and support
- Earthdata Forum