Getting Started with GEDI Spaceborne Lidar for Ecosystem Applications — Training Repository

Overview & Purpose

This repository hosts the training webpage and hands-on tutorials to help users learn the fundamentals of the Global Ecosystem Dynamics Investigation (GEDI) spaceborne LiDAR, apply GEDI data to real-world ecosystems, and explore best practices for preparing context-aware, quality-filtered GEDI datasets.

These tutorials are designed for users new to spaceborne lidar (GEDI) who want to:

• Explore GEDI L2B (height and structural metrics)
• Estimate aboveground biomass using GEDI L4
• Compare GEDI observations with local, high-resolution lidar data
• Perform local biomass modeling combining R and Earth Engine (via JavaScript) scripts

They demonstrate end-to-end workflows: downloading/processing data, visualizing, analysis, and interpretation.

Citing & Credits

The training is developed by the NASA EarthRISE team.
You can also view a live version of the training site here:
nasa-earthrise.github.io/training_Getting_started_with_GEDI_spaceborne_lidar

If you use these tutorials or adapt parts thereof in your research or teaching, please cite:

Jiménez, S., Mayer, T., Pinto, N., Cooley, S., Healey, S., Christine Evans, Numata, I., Horn, K., West, D., Walker, K., Abramowitz, J., Cruz, S., Martin Arias, V., Pransky, L., Kruskopf, M., Yang, Z., Johnson, L., Fareed, N., d'Oliveira, M., … Billy Ashmall. (2025). NASA-EarthRISE/training_Getting_started_with_GEDI_spaceborne_lidar: v1.0.0 (First-release). Zenodo. https://doi.org/10.5281/zenodo.17353798

---

Table of Contents

1. Overview & Purpose
2. Citing & Credits
3. Repository Structure
4. Getting Started
5. Usage Guidelines & Tips
6. Contribution & Issues

---

Repository Structure

A few highlights:

• tutorials/ contains Jupyter notebooks (e.g. Exploring_Forest_Structure_with_GEDI_L2B.ipynb, Exploring_Biomass_with_GEDI_l4.ipynb) that walk you through using GEDI Level-2 and Level-4 products for forest-structure and biomass analysis with workflows combining R, Python, and Google Earth Engine (JavaScript).
• AOIs/ holds geospatial area-of-interest shapefiles or footprints for regional examples.
• docs/ and site files power the training webpage.
• images/ stores all figures used in the tutorials and web pages.
• append_field.txt is a helper file (e.g. appended field metadata) used by one or more notebooks.

Here is an outline of the main directories and files:

├── AOIs/ ← Areas of interest, boundary files, etc.

├── docs/ ← Documentation or site-generation files

├── images/ ← Images used in tutorials / site

├── tutorials/ ← Python notebooks, R scripts, and Google Earth Engine Scripts / hands-on modules

• Exploring_forest_structure_with_GEDI_L2B.ipynb ← Module 2 L2B vegetation structure access and preparation

• Exploring_biomass_with_GEDI_L4.ipynb ← Module 2 L4A & L4B biomass estimations access and preparation

• Comparing_GEDI_L2B_with_highres_lidar_sewanee.ipynb ← Module 2 correlate GEDI with airborne lidar

• Local_biomass_modelling/ ← Module 4 correlating biomass with local survey data and airborne lidar

  • Exercise 1 ← R scripts and supporting datasets for GEDI simulation

  • Exercise 2 ← R script for Random Forest Modelling

  • Exercise 3 ← Google Earth Engine javascript for biomass time series and CCDC land cover maps

├── README.md ← This file

├── append_field.txt ← (auxiliary file used in tutorials)

---

Getting Started

Take the Course via the Webpage

Take the self-paced course via the webpage and follow along the knowledge checks, external links, supporting information, and guided tutorials for multiple media learning content.

All Modules and Topics Overview

Overview of topics covered in each module, respective format(s), and any technical requirements are listed below. Additional prerequisites and background information will be detailed within each module.

Module 1

Introduction to Full Waveform Lidar introduces lidar data, its physical principles, and its sensitivity to biophysical parameters, and explores several U.S.-based and global applications using different lidar sensors. Exercises will solidify participants’ understanding of waveform lidar capabilities in terrestrial applications.

Level: introductory to intermediate.

Section	Topics	Format	Requirements
Introduction to GEDI Full Waveform Lidar for Terrestrial Applications	Fundamentals of Lidar Remote Sensing	Recorded lecture	Video and audio viewing
Introduction to GEDI Full Waveform Lidar for Terrestrial Applications	Full Waveform Lidar from the GEDI Mission	Recorded lecture	Video and audio viewing
Practical Applications with GEDI Full Waveform Data	Tutorial: A comprehensive Python notebook for processing and analyzing NASA GEDI L1B (Global Ecosystem Dynamics Investigation Level 1B) full waveform LiDAR data using Google Colab	Recorded demonstration and self-paced python tutorial	Google account and Drive

Module 2

A Deep Dive into GEDI details the GEDI mission, its data products, and case study applications. Exploratory scripts provide starting points for participants to evaluate many of GEDI’s data products by incorporating recommended filtering and pre-processing strategies.

Level: introductory to intermediate.

Section	Topics	Format	Requirements
Lasering in on the GEDI Mission	Why GEDI?	Self-paced lecture
Lasering in on the GEDI Mission	Navigating the GEDI Ecosystem	Self-paced lecture
Lasering in on the GEDI Mission	Tools for Navigating GEDI	Self-paced lecture
Lasering in on the GEDI Mission	The Building Block of Analysis	Self-paced lecture
Lasering in on the GEDI Mission	How GEDI Information Ecosystem Studies	Self-paced lecture
From Ground to Canopy	GEDI’s Elevation and Relative Height Metrics Explained	Self-paced lecture
From Canopy Layers to Vertical Profiles	Vegetation Structural Insights From GEDI	Self-paced lecture
From Canopy Layers to Vertical Profiles	Tutorial: Exploring Forest Structure with GEDI L2B in the Southeast	Self-paced python tutorial	Google account and Drive
From Canopy Layers to Vertical Profiles	Tutorial: Comparing L2B PAI with high-resolution lidar in the Sewanee Domain, Tennessee	Self-paced python tutorial	Google account and Drive
Above Ground Biomass with GEDI	GEDI’s Biomass Estimation Approach	Self-paced lecture
Above Ground Biomass with GEDI	Tutorial: Exploring Biomass with GEDI L4A and L4B in the Southeast	Self-paced python tutorial	Google account and Drive

Module 3

Biomass Change with OBIWAN discusses downstream GEDI derived products for carbon and biomass monitoring with Online Biomass Inference using Waveforms and iNventory (OBIWAN) Application Programming Interface (API). Participants will learn how biomass products are derived and their advantages and limitations. This module shows participants how to create dashboards and reporting systems for any geography covering any period from 1985 to the present.

Level: introductory to intermediate.

Section	Topics	Format	Requirements
From Estimating Biomass with GEDI to Estimating Biomass Change with OBIWAN API	GEDI mission biomass estimation: theory and products	Recorded lecture	Video and audio viewing
From Estimating Biomass with GEDI to Estimating Biomass Change with OBIWAN API	OBIWAN: estimating biomass change with GEDI and Landsat time series	Recorded lecture	Video and audio viewing
Customize OBIWAN through its API	Tutorial: Using the OBIWAN API in Alabama	Recorded demonstration of python and Google Earth Engine tutorials	Video and audio viewing

Module 4

Localized Forest Biomass Estimation will have advanced applications focused on GEDI derived products for monitoring forest disturbances and carbon dynamics. Dr. Izaya Numata (South Dakota State University) will present methods for forest biomass estimation using field, airborne, and satellite lidar (GEDI) data developed in Acre, Brazil. Participants will understand the advantages, limitations, and considerations for applying such methodology on their own.

Level: Intermediate to advanced.

Section	Topics	Format	Requirements
Forest Biomass Estimation Using Data from Field, Airborne and Spaceborne Lidar	Learn to develop local forest biomass model using field and remote sensing data and its application for AGB mapping	Self-paced lecture
Calibrating Field, Airborne and Spaceborne Lidar	Tutorial: Exercise1) GEDI waveform simulation with airborne lidar using rGEDI. Exercise 2) Development of local AGB model with field AGB and simulated ALS Relative Height metrics in R. Exercise 3) Above ground biomass change mapping in Google Earth Engine.	Self-paced RStudio and Google Earth Engine tutorials	Github account

Follow Along With or Adapt the Tutorials

This training contains modules and tutorials that build throughout the course. Below is a high-level overview:

Module, Focus Description / Key Outcomes

• Module 1 Introduction & lidar basics (Placeholder) Orientation, LiDAR fundamentals, GEDI mission overview
• Module 2 Forest structure (L2B) Work with GEDI L2B products — waveform, footprint metrics, canopy height, etc.
• Module 2 PAI, GEDI-LAS comparisons Compare GEDI metrics with local LiDAR or field data
• Module 2 Biomass estimation (L4) Use GEDI Level-4 biomass products, build biomass models
• Module 3 Biomass Change with OBIWAN
• Module 4 Advanced calibration for local biomass

Each notebook includes:

• Background context
• Step-by-step Python code
• Data download and processing instructions
• Visualizations (maps, graphs)
• Exercises or extension ideas
• Be sure to read the narrative around code cells — the explanations are integral to learning.

Before you begin EACH TUTORIAL, ensure you have the respective requirements for each tutorial:

• NASA EarthData Access account and login credentials or API token
• A working google colab notebook or jupyter notebook and Google account
• RStudio, Rtools, and respective defined libraries for Module 4 tutorials
• Google Earth Engine account using the Engine Code Editor
• Common geospatial/data libraries (e.g. numpy, pandas, geopandas, rasterio, matplotlib, etc.) are defined in the tutorials

Note: The specific versions used in the original tutorials may matter for reproducibility. If version info is available in the notebooks or associated metadata, use those.

We recommend using a virtual environment (venv, conda, etc.) to isolate dependencies.

Running the Tutorials

For colab notebooks:

1. Open a tutorial notebook, e.g. Exploring_forest_structure_with_GEDI_L2B.ipynb.
2. Go cell by cell (or “Run All”) to execute.
3. If prompted for credentials (Earthdata or Earth Engine), follow the instructions in cell comments.
4. Review outputs, figures, logs, intermediate files (if saved).

R + GEE (Local biomass modelling) tutorials, follow along the prompts from the presentation:

1. In R or RStudio, open the R scripts in Local_biomass_modelling/.
2. Set the working directory appropriately (point to where the .js file and data are).
3. If needed, edit the R script to point to your Python environment (via reticulate) or set paths.
4. Run the R script sequentially. It may call Earth Engine via R or call/submit the GEE JavaScript script.
5. For the JavaScript .js, you may run it in the GEE Code Editor (copy-paste)

Usage Guidelines & Tips

Data Access

• Many modules download GEDI data (L2 or L4). Ensure your internet connection is stable and that you have sufficient disk space.

Quality Filtering

• GEDI data include quality flags (e.g. sensitivity, beam type, etc. ). Tutorials emphasize applying filters to avoid spurious or low-confidence measurements.

Spatial Subsets / AOIs

• Where possible, focus on modest-sized areas (e.g. tens to a few hundreds of footprints) to speed execution and avoid memory bottlenecks.

Reproducibility

• Document all file paths and data transformations.

Extensibility

• You are encouraged to adapt notebook logic to your own regions (AOIs), combine with other datasets, and extend biomass/height modeling workflows.

Troubleshooting

• If you get dependency conflicts, try isolating in a fresh virtual environment by restarting or disconnect/reconnect the Runs.
• For large raster or geospatial operations, monitor memory use, delete old versions, especially if saved to Drive.
• If data downloads fail, check URLs/availability or replace with alternate mirrors.
• “Module not found” in Python --> Confirm you installed all packages in the active environment

Contribution & Issues

On the Discussions Forum, we welcome contributions, feedback, and requests such as:

• Filing GitHub issues for bugs, suggestions, broken links, or errata.
• Submitting pull requests with improvements, new modules, or fixes.
• Proposing new case studies or extension notebooks (e.g. for tropical forests, urban forests, change detection), especially via the pre and post questionnaires.
• Updating documentation or enhancing clarity of narrative text.