draft-readme.md

Overview

earthaccess is a python library to search, download or stream NASA Earth science data with just a few lines of code.

In the age of cloud computing, the power of open science only reaches its full potential if we have easy-to-use workflows that facilitate research in an inclusive, efficient and reproducible way. Unfortunately —as it stands today— scientists and students alike face a steep learning curve adapting to systems that have grown too complex and end up spending more time on the technicalities of the tools, cloud and NASA APIs than focusing on their important science.

During several workshops organized by NASA Openscapes the need to provide easy-to-use tools to our users became evident. Open science is a collaborative effort, it involves people from different technical backgrounds. Data analysis for the pressing problems we face cannot be limited by the complexity of the underlaying systems and thus providing easy access to NASA Earthdata is the main motivation behind this library.

Installing earthaccess

Install the latest release using conda

conda install -c conda-forge earthaccess

Using Pip

pip install earthaccess

Try it in your browser without installing anything!

Usage

With earthaccess we can login, search and download data with a few lines of code and even more relevant, our code will work the same way if we are running it in the cloud or from our laptop. earthaccess handles authentication with NASA's Earthdata Login (EDL), search using NASA's CMR and access through fsspec.

The only requirement to use this library is to open a free account with NASA EDL.

Authentication

Once you have an EDL account, you can authenticate using one of the following three methods:

Using a .netrc file
- Can use earthaccess to read your EDL credentials (username and password) from a .netrc file
Reading your EDL credentials from environment variables
- if available you can use environment variables EDL_USERNAME and EDL_PASSWORD
Interactively entering your EDL credentials
- You can be prompted for these credentials and save them to a .netrc file

import earthaccess

auth = earthaccess.login(strategy="netrc")
if not auth:
    auth = earthaccess.login(strategy="interactive", persist=True)

Once you are authenticated with NASA EDL you can:

Get a file from a DAAC using a fsspec session.
Request temporary S3 credentials from a particular DAAC (needed to download or stream data from an S3 bucket in the cloud).
Use the library to download or stream data directly from S3.
Regenerate CMR tokens (used for restricted datasets)

Searching for data

Once we have selected our dataset we can search for the data granules using doi, short_name or concept_id. If we are not sure or we don't know how to search for a particular dataset, we can start with the "searching for data" tutorial or through the Earthdata search portal. For a complete list of search parameters we can use visit the extended API documentation.

results = earthaccess.search_data(
    short_name='ATL06',
    version="005",
    cloud_hosted=True,
    bounding_box=(-10, 20, 10, 50),
    temporal=("2020-02", "2020-03"),
    count=100
)

Now that we have our results we can do multiple things, we can iterate over them to get HTTP (or S3) links; we can download the files to a local folder or we can open these files and stream their content directly to other libraries e.g. xarray.

Accessing the data

Option 1: Using the data links

If we already have a workflow in place for downloading our data, we can use earthaccess as a search-only library and get HTTP links from our query results. This could be the case if our current workflow uses a different language and we only need the links as input.

# if the data set is cloud hosted there will be S3 links available. The access parameter accepts "direct" or "external", direct access is only possible if you are in the us-west-2 region in the cloud.
data_links = [granule.data_links(access="direct") for granule in results]

# or if the data is an on-prem dataset
data_links = [granule.data_links(access="external") for granule in results]

Note: earthaccess can get S3 credentials for us, or auhenticated HTTP sessions in case we want to use them with a different library.

Option 2: Download data to a local folder

This option is practical if you have the necessary space available on disk, the earthaccess library will print out the approximate size of the download and its progress.

files = earthaccess.download(results, "./local_folder")

Option 3: Direct S3 Access - Stream data directly to xarray

This method works best if you are in the same region as the data (us-west-2) and you are working with gridded datasets (processing level 3 and above).

import xarray as xr

ds = xr.open_mfdataset(earthaccess.open(results, auth=auth), engine="scipy")

And that's it! Just one line of code, and this same piece of code will also work for data that are not hosted in the cloud, i.e. NASA hosted data sets.

More examples coming soon!

Compatibility

Only Python 3.8+ is supported.

Contributors

Contributing Guide

Welcome! 😊👋

Please see the Contributing Guide.

Project Board.

Glossary

Level of Support

This repository is not actively supported by NSIDC but we welcome issue submissions and pull requests in order to foster community contribution.

betolink/draft-readme.md