NSIDC-Data-Tutorials

Summary

This combined repository includes tutorials and code resources provided by the NASA National Snow and Ice Data Center Distributed Active Archive Center (NSIDC DAAC). These tutorials are provided as Python-based Jupyter notebooks that provide guidance on working with various data products, including how to access, subset, transform, and visualize data. Each tutorial can be accessed by navigating to the /notebooks folder of this repository. Please see the README files associated with each individual tutorial folder for more information on each tutorial and their learning objectives. Please note that all branches outside of Main should be considered in development and are not supported.

Tutorials

ICESat-2_Cloud_Access Notebooks

These notebooks demonstrate how to search and access ICESat-2 from the NASA Earthdata Cloud:

Accessing and working with ICESat-2 Data in the Cloud

Originally presented to the UWG (User Working Group) in May 2022, this tutorial demonstrates how to search for ICESat-2 data hosted in the Earthdata Cloud and how to directly access it from an Amazon Web Services (AWS) Elastic Compute Cloud (EC2) instance using the earthaccess package.

Plotting ICESat-2 and CryoSat-2 Freeboards

This notebook demonstrates plotting ICESat-2 and CryoSat-2 data in the same map from within an AWS ec2 instance. ICESat-2 data are accessed via "direct S3 access" using earthaccess. CryoSat-2 data are downloaded to our cloud instance from their ftp storage lcoation and accessed locally.

Processing Large-scale Time Series of ICESat-2 Sea Ice Height in the Cloud

This notebook utilizes several libraries to performantly search, access, read, and grid ATL10 data over the Ross Sea, Antarctica including earthaccess, h5coro, and geopandas. The notebook provides further guidance on how to scale this analysis to the entire continent, running the same workflow from a script that can be run from your laptop using Coiled.

MEaSUREs

Download, crop, resample, and plot multiple GeoTIFFs

This tutorial guides you through programmatically accessing and downloading GeoTIFF files from the NSIDC DAAC to your local computer. We then crop and resample one GeoTIFF based on the extent and pixel size of another GeoTIFF, then plot one on top of the other.

We will use two data sets from the NASA MEaSUREs (Making Earth System data records for Use in Research Environments) program as an example:

SnowEx_ASO_MODIS_Snow

Snow Depth and Snow Cover Data Exploration

Originally demonstrated through the NASA Earthdata Webinar "Let It Snow! Accessing and Analyzing Snow Data at the NSIDC DAAC" on May 6, 2020, this tutorial provides guidance on how to discover, access, and couple snow data across varying geospatial scales from NASA's SnowEx, Airborne Snow Observatory, and Moderate Resolution Imaging Spectroradiometer (MODIS) missions. The tutorial highlights the ability to search and access data by a defined region, and combine and compare snow data across different data formats and scales using a Python-based Jupyter Notebook.

ICESat-2_MODIS_Arctic_Sea_Ice

Getting the most out of NSIDC DAAC data: Discovering, Accessing, and Harmonizing Arctic Remote Sensing Data

Originally presented during the 2019 AGU Fall Meeting, this tutorial demonstrates the NSIDC DAAC's data discovery, access, and subsetting services, along with basic open source resources used to harmonize and analyze data across multiple products. The tutorial is provided as a series of Python-based Jupyter Notebooks, focusing on sea ice height and ice surface temperature data from NASA’s ICESat-2 and MODIS missions, respectively, to characterize Arctic sea ice.

ITS_LIVE

Global land ice velocities. The Inter-mission Time Series of Land Ice Velocity and Elevation (ITS_LIVE) project facilitates ice sheet, ice shelf and glacier research by providing a globally comprehensive and temporally dense multi-sensor record of land ice velocity and elevation with low latency. Scene-pair velocities were generated from satellite optical and radar imagery.

The notebooks on this project demonstrate how to search and access ITS_LIVE velocity pairs and provide a simple example on how to build a data cube.

IceFlow

Caution

The IceFlow notebooks and supporting code have some known problems and users should exercise caution. It is likely that users will run into errors while interacting with the notebooks. Requests for ITRF transformations are not currently working as expected. We recommend users look at the corrections notebook for information about how to apply ITRF transformations to data themselves. IceFlow is currently under maintenence, and we hope to resolve some of these issues soon.

Harmonized data for pre-IceBridge, ICESat and IceBridge data sets. These Jupyter notebooks are interactive documents to teach students and researchers interested in cryospheric sciences how to access and work with airborne altimetry and related data sets from NASA’s IceBridge mission, and satellite altimetry data from ICESat and ICESat-2 missions using the NSIDC IceFlow API

Usage with Binder

The Binder button above allows you to explore and run the notebook in a shared cloud computing environment without the need to install dependencies on your local machine. Note that this option will not directly download data to your computer; instead the data will be downloaded to the cloud environment.

Usage with Docker

On Mac OSX or Linux

Install Docker. Use the left-hand navigation to select the appropriate install depending on operating system.
Download the NSIDC-Data-Tutorials repository from Github.
Unzip the file, and open a terminal window in the NSIDC-Data-Tutorials folder's location.
From the terminal window, launch the docker container using the following command, replacing [path/notebook_folder] with your path and notebook folder name:

docker run --name tutorials -p 8888:8888 -v [path/notebook_folder]:/home/jovyan/work nsidc/tutorials

Example:

docker run --name tutorials -p 8888:8888 -v /Users/name/Desktop/NSIDC-Data-Tutorials:/home/jovyan/work nsidc/tutorials

Or, with docker-compose:

docker-compose up

If you want to mount a directory with write permissions, you need to grant the container the same permissions as the one on the directory to be mounted and tell it that has "root" access (within the container). This is important if you want to persist your work or download data to a local directory and not just the docker container. Run the example command below for this option:

docker run --name tutorials -e NB_UID=$(id -u) --user root -p 8888:8888 -v  /Users/name/Desktop/NSIDC-Data-Tutorials:/home/jovyan/work nsidc/tutorials

The initialization will take some time and will require 2.6 GB of space. Once the startup is complete you will see a line of output similar to this:

To access the notebook, open this file in a browser:
        file:///home/jovyan/.local/share/jupyter/runtime/nbserver-6-open.html
    Or copy and paste one of these URLs:
        http://4dc97ddd7a0d:8888/?token=f002a50e25b6f623aa775312737ba8a23ffccfd4458faa6f
     or http://127.0.0.1:8888/?token=f002a50e25b6f623aa775312737ba8a23ffccfd4458faa6f

If you started your container with the -d/--detach option, check docker logs tutorials for this output.

Open up a web browser and copy one of the URLs as instructed above.
You will be brought to a Jupyter Notebook interface running through the Docker container. The left side of the interface displays your local directory structure. Navigate to the work folder of the NSIDC-Data-Tutorials repository folder. You can now interact with the notebooks to explore and access data.

On Windows

Install Docker.
Download the NSIDC-Data-Tutorials repository from Github.
Unzip the file, and open a terminal window (use Command Prompt or PowerShell, not PowerShell ISE) in the NSIDC-Data-Tutorials folder's location.
From the terminal window, launch the docker container using the following command, replacing [path\notebook_folder] with your path and notebook folder name:

docker run --name tutorials -p 8888:8888 -v [path\notebook_folder]:/home/jovyan/work nsidc/tutorials

Example:

docker run --name tutorials -p 8888:8888 -v C:\notebook_folder:/home/jovyan/work nsidc/tutorials

Or, with docker-compose:

docker-compose up

If you want to mount a directory with write permissions you need to grant the container the same permissions as the one on the directory to be mounted and tell it that has "root" access (within the container)

docker run --name tutorials --user root -p 8888:8888 -v C:\notebook_folder:/home/jovyan/work nsidc/tutorials

The initialization will take some time and will require 2.6 GB of space. Once the startup is complete you will see a line of output similar to this:

To access the notebook, open this file in a browser:
        file:///home/jovyan/.local/share/jupyter/runtime/nbserver-6-open.html
    Or copy and paste one of these URLs:
        http://(6a8bfa6a8518 or 127.0.0.1):8888/?token=2d72e03269b59636d9e31937fcb324f5bdfd0c645a6eba3f

If you started your container with the -d/--detach option, check docker logs tutorials for this output.

Follow the instructions and copy one of the URLs into a web browser and hit return. The address should look something like this:

http://127.0.0.1:8888/?token=2d72e03269b59636d9e31937fcb324f5bdfd0c645a6eba3f

You will now see the NSIDC-Data-Tutorials repository within the Jupyter Notebook interface. Navigate to /work to open the notebooks.
You can now interact with the notebooks to explore and access data.

Usage with Mamba/Conda

Note: If we already have conda or mamba installed we can skip the first step.

Install mambaforge (Python 3.9+) for your platform from mamba documentation
Download the NSIDC-Data-Tutorials repository from Github by clicking the green 'Code' button located at the top right of the repository page and clicking 'Download Zip'. Unzip the file, and open a command line or terminal window in the NSIDC-Data-Tutorials folder's location.
From a command line or terminal window, install the required environment with the following commands:

Linux

mamba create -n nsidc-tutorials --file binder/conda-linux-64.lock

OSX

mamba create -n nsidc-tutorials --file binder/conda-osx-64.lock

Windows

mamba create -n nsidc-tutorials --file binder/conda-win-64.lock

You should now see that the dependencies were installed and our environment is ready to be used.

Activate the environment with

conda activate nsidc-tutorials

Launch the notebook locally with the following command:

jupyter lab

This should open a browser window with the JupyterLab IDE, showing your current working directory on the left-hand navigation. Navigate to the tutorial folder of choice and click on their associated *.ipynb files to get started.

Tutorial Environments

Although the nsidc-tutorial environment should run all the notebooks in this repository, we also include tutorial-specific environments that will only contain the dependencies for them. If we don't want to "pollute" our conda environments and we are only going to work with one of the tutorials we recommend to use them instead of the nsidc-tutorial environment. The steps to install them are exactly the same but the environment files are inside the environment folders in each of the tutorials. e.g. for ITS_LIVE

cd notebooks/itslive 
mamba create -n nsidc-itslive --file environment/conda-linux-64.lock
conda activate nsidc-itslive
jupyter lab

This should create a pinned environment that should be fully reproducible across platforms.

NOTE: Sometimes Conda environments change (break) even with pinned down dependencies. If you run into an issue with dependencies for the tutorials please open an issue and we'll try to fix it as soon as possible.

Credit

This software is developed by the National Snow and Ice Data Center with funding from multiple sources.

License

This repository is licensed under the MIT license.

nsidc/NSIDC-Data-Tutorials