How to communicate during this workshop

Throughout this workshop, we will be using Slack to make sure folks are on the same page and getting the most out of our planned activities. If you have any difficulties before, during, or after the workshop please reach out the workshop organizers on Slack! Click here to join NGEE Arctic Modeling Workshop Slack (right-click, open in new tab)

Background on Docker

Docker is an open-source project that allows programmers to deploy their applications in a sandbox (called containers) to run on the host operating system (Linux). Docker allows programmers to package an application with all of its dependencies (code, data, etc) into a standardized unit that other users can work with despite running different types of operating systems. There is a lot more background information available on Docker at their website.

Some resources to learn more about Docker (right-click, open in a new tab):

Docker for beginners Docker for Dummies Helpful Docker commands for managing your Docker environment

How we will use Docker in this workshop:

We have designed the workshop materials and tutorials to use two main Docker images, one for running the ELM model, which is built on a precursor container where all of the required libraries for ELM are compiled (e.g. netCDF, HDF5, Python), and a second that is used to analyze ELM simulation output in Jupyter-Lab, as well as to configure custom parameter experiments. You can think about this as a "stack" of two different containers that can either be run in sequential order (i.e. run the ELM container, once complete run the jupyter-lab container to analyze the output) or can be used at the same time on a single computer. This allows a user to be running ELM simulations while also exploring the output of a previous simulation in Jupyter notebooks. Below we provide instructions for setting up and running ELM simulations in Docker as well as starting and using the Jupyter-Lab environment. Note: because we are using Docker, users do not need to install Python or any special libraries on their computers. Everything needed to run the model or launch jupyter-lab is contained within the two containers that you download when following the instructions below for setting up your Docker environment.

1. Setting up your Docker Environment for the NGEE-Arctic modeling workshop

Instructions for getting your local Docker environment setup are available here. We request that workshop attendees complete this homework prior to arriving at the workshop. If you run into problems, please reach out to the meeting organizers on Slack.

2. NGEE-Arctic ELM Docker Demos

This page includes a series of demonstration activities that we will be working through together during the workshop: Running ELM for NGEE Arctic Field Sites

3. How will we look at model output?

We will be using a series of Jupyter-Lab notebooks to dig through model output together. These are available here: Analyzing Output in Jupyter-Lab

4. Hands on Demos

• Restarting transient year simulations and Changing a parameter value
• AK K64G Demo Quick Start Guide 2c: Hydrology (soil-hydraulic parameters via texture and or OM)

5. Additional resources

• Using a host directory instead of a Docker volume to store model data and/or ELM output

• Jupyter Lab Container with Google Earth Engine You can use this alternative Jupyter-Lab environment to analyze ELM output and develop notebooks that can compare model output with GEE remote sensing layers

• Modifying soil physical properties in ELM

• Perturbing inputs from meteorological forcing files

• Scaling N or P deposition from forcing files

• Example ELM ensemble simulation