Capturing Temporal Dynamics in Large-Scale Canopy Tree Height Estimation

Jan Pauls, Max Zimmer, Berkant Turan, Sassan Saatchi, Philippe CIAIS, Sebastian Pokutta, Fabian Gieseke

TL;DR: We estimate tree canopy height over time using Sentinel Satellite Time Series Data and a custom Deep Learning approach.

Abstract: With the rise in global greenhouse gas emissions, accurate large-scale tree canopy height maps are essential for understanding forest structure, estimating above-ground biomass, and monitoring ecological disruptions. To this end, we present a novel approach to generate large-scale, high-resolution canopy height maps over time. Our model accurately predicts canopy height over multiple years given Sentinel 2 time series satellite data. Using GEDI LiDAR data as the ground truth for training the model, we present the first 10 m resolution temporal canopy height map of the European continent for the period 2019–2022. As part of this product, we also offer a detailed canopy height map for 2020, providing more precise estimates than previous studies. Our pipeline and the resulting temporal height map are publicly available, enabling comprehensive large-scale monitoring of forests and, hence, facilitating future research and ecological analyses. For an interactive viewer, see https://europetreemap.projects.earthengine.app/view/europeheight.

Lay Summary:

Understanding forests is key to tackling climate change, but past tree height maps often lack detail or up-to-date information.

This study produced the first high-resolution (10-meter) maps of tree height across Europe that show changes from 2019 to 2022. Using satellite data from Sentinel-1 and Sentinel-2, we analyzed a full year of monthly images with a custom AI model — capturing fine forest details better than ever before.

The result: more accurate maps, especially for tall trees that store the most carbon. These time-based maps also help track forest changes like deforestation. Both the method and the maps are freely available to support climate research and forest monitoring.

Application-Driven Machine Learning: This submission is on Application-Driven Machine Learning.

Link To Code: https://github.com/AI4Forest/Europe-Temporal-Canopy-Height

Primary Area: Applications->Chemistry, Physics, and Earth Sciences

Keywords: canopy height, remote sensing, satellite, deep learning application, sentinel, gedi

Submission Number: 5057