top of page


Predicting land cover using GEDI-L2A tree canopy data for New York State

Frank Seidl, Luke Kiernan, Keavin Moore, Nicholas Barvinok, Noah Rahman


The tree canopy height and vertical structure of a given region can provide important implications for researchers and developers concerned with climatological trends over time. This can provide information to describe a link between important human factors, such as urban population and wildfire persistence, through an area’s land cover. We created a classification model that can predict land cover within New York state (e.g., urban-developed, forested, wetland, barren) by combining the GEDI-L2A tree canopy dataset and MRLC land cover dataset of the contiguous United States. We hope to use our best-fit model to predict land usage classifications for another region of interest, such as wildfire-prone California, or the dense forests of northern Europe.

Screen Shot 2022-06-03 at 11.31.35 AM.png
github URL
bottom of page