This repository provides a unified workflow of acquiring data from the Global Ecosystem Dynamics Investigation (GEDI) mission at a footprint level.

• Finder searches for all the orbits passing over a Region Of Interest (ROI) at the specified timestamp and outputs its download link from LPDAAC / ORNLDAAC
• Downloader downloads the entire granule from the link provided by the Finder, provided with authentication for EarthData
• Subsetter clips to the study area and selects all the available Science Dataset (SDS) data product variables to burn for each footprint, and outputs the subsetted orbit

Pipeline automates the previous modules into a single class / script.

1. The Finder searches NASA's data repository for all the available orbits that pass over the ROI (Rectangle) and a list of URLs is returned, containing the download links for the granules (A.).
2. The Downloader downloads each (entire) granule to a specified directory, as subsetting the granule before downloading is currently unavailable through the APIs provided.
3. After downloading, the subsetter first clips the footprints and the specified BEAMS that the ROI contains and then selects the specified SDS variables for each footprint, saving to a .gpkg file. This process is repeated every time a granule is downloaded. After subsetting, the original granule is deleted, as to save space.
4. After pipeline completion, the user is left with all the clipped orbits displayed by the Finder to the specified ROI (B.), saved as .gpkg files.
5. The user can manipulate, analyze and use each footprint as desired. The table displays the variables burnt into a single footprint (in this example, the data product is GEDI L2A Version 2) (C.).

Depending on the number of granules and the number of variables to subset, processing time may vary, so be patient.

Usually the user finds and downloads GEDI data products via a graphical user interface (GUI) using NASA EarthData Search or programmatically using NASA's Common Metadata Repository (CMR), and then subsets to a desired study area, using separate programs/scripts.

This pipeline automates such task by finding, downloading and subsetting all the granules provided by the Finder using NASA's CMR. There are repositories that automate the Finder and Subsetter modules, but need to be executed separately and with different program arguments. We believe that merging all of these separate frameworks eases some burden into launching different scripts, managing the data and processing, streamlining the workflow for Remote Sensing, Data Analysis and Data Science researchers, which enables a much more efficient analysis of GEDI data.

• The user must have a registered EarthData account before accessing any of these datasets.
• Python (>=3.12)

You can set up the virtual environment to run this project in two ways:

1. Python's venv and pip
2. Anaconda / Mamba Environment

1. Create a new virtual environment with name gedi-pipeline by running python3 -m venv ~/venvs/gedi-pipeline
2. Activate the newly created environment by source ~/venvs/gedi-pipeline/bin/activate (Check that python points to the new environment with which python3 or which pip after activating, it should print something like: ../venvs/gedi-pipeline/bin/python)
3. Install all the required packages for the pipeline to work with python3 -m pip install -r requirements.txt

You may create a Conda/Mamba environment specifically for this project, requiring a few extra steps than creating a venv.

1. Install Anaconda or Mambaforge (If installing on Windows, check the Add to PATH environment variable to work on a command-line interface).
2. Create a new conda environment called gedi-pipeline with pip installed by running conda create --name gedi-pipeline python=3.12 pip (or switch conda with mamba if Mambaforge is installed).
3. Activate the newly created environment with conda activate gedi-pipeline or mamba activate gedi-pipeline and check if python points to the activated environment path with which python3
4. Update the existing environment with environment.yml provided in the repository with conda env update -f environment.yml. Be patient, it should take a while.

You can execute the pipeline script by running python3 gedi_pipeline.py. Additional commands must be provided for the pipeline to work, such as GEDI product, version, start date and end date query and the output directory. You can run python3 gedi_pipeline.py --help for more information.

• GEDI L1B Geolocated Waveform Data Global Footprint Level - GEDI01_B
• GEDI L2A Elevation and Height Metrics Data Global Footprint Level - GEDI02_A
• GEDI L2B Canopy Cover and Vertical Profile Metrics Data Global Footprint Level - GEDI02_B
• GEDI L4A Footprint Level Aboveground Biomass Density - GEDI04_A

For each GEDI data product, you can specify which version you want to download: version '001' or version '002'.

This project is in its early stages so any contributions are welcome with a well documented/explained issue and implementation!

We would like to thank the University of Maryland and NASA's Goddard Space Flight Center for their relentless work on the GEDI mission.

The GEDI Pipeline is (very) inspired by other repositories, which we would like to thank for their contributions for the GEDI Project:

• NASA's GEDI Data Resources: https://github.com/nasa/GEDI-Data-Resources
• pyGEDI: https://github.com/EduinHSERNA/pyGEDI
• Also from this specific answer to a forum post: https://forum.earthdata.nasa.gov/viewtopic.php?t=591
• rGEDI: https://github.com/carlos-alberto-silva/rGEDI

This work was conducted within the framework of the 3D-SMOS project (Combining LiDAR, radar, and multispectral data to characterize the three-dimensional structure of vegetation and produce land cover maps), financially supported by the Directorate-General for Territory (DGT) with funds from the Recovery and Resilience Plan (Investimento RE-C08-i02: Cadastro da Propriedade Rústica e Sistema de Monitorização da Ocupação do Solo).

Dubayah, R., Blair, J.B., Goetz, S., Fatoyinbo, L., Hansen, M., Healey, S., Hofton, M., Hurtt, G., Kellner, J., Luthcke, S., & Armston, J. (2020) The Global Ecosystem Dynamics Investigation: High-resolution laser ranging of the Earth’s forests and topography. Science of Remote Sensing, p.100002. https://doi.org/10.1016/j.srs.2020.100002

Hancock, S., Armston, J., Hofton, M., Sun, X., Tang, H., Duncanson, L.I., Kellner, J.R. and Dubayah, R., 2019. The GEDI simulator: A large-footprint waveform lidar simulator for calibration and validation of spaceborne missions. Earth and Space Science. https://doi.org/10.1029/2018EA000506

Silva,C.A; Hamamura,C.; Valbuena, R.; Hancock,S.; Cardil,A.; Broadbent, E. N.; Almeida,D.R.A.; Silva Junior, C.H.L; Klauberg, C. rGEDI: NASA's Global Ecosystem Dynamics Investigation (GEDI) Data Visualization and Processing. version 0.1.10, accessed on February. 23, 2024, available at: https://CRAN.R-project.org/package=rGEDI

Corado, L., Godinho, S. GEDI-Pipeline. version 0.1.0, available at: https://github.com/leonelluiscorado/GEDI-Pipeline