Aboveground carbon biomass estimate with Physics-informed deep network

TL;DR

This paper presents a physics-informed deep learning approach to accurately map aboveground carbon biomass across the US at high resolution, integrating multiple data sources and physical parameters.

Contribution

It introduces a novel deep neural network combining radar, hyperspectral imagery, and GPP data to improve AGB estimation accuracy over previous methods.

Findings

Deep learning model achieves lower RMSE than random forest.

Inclusion of GPP data reduces error and variability.

Model successfully detects AGB loss from wildfire impacts.

Abstract

The global carbon cycle is a key process to understand how our climate is changing. However, monitoring the dynamics is difficult because a high-resolution robust measurement of key state parameters including the aboveground carbon biomass (AGB) is required. Here, we use deep neural network to generate a wall-to-wall map of AGB within the Continental USA (CONUS) with 30-meter spatial resolution for the year 2021. We combine radar and optical hyperspectral imagery, with a physical climate parameter of SIF-based GPP. Validation results show that a masked variation of UNet has the lowest validation RMSE of 37.93 $\pm$ 1.36 Mg C/ha, as compared to 52.30 $\pm$ 0.03 Mg C/ha for random forest algorithm. Furthermore, models that learn from SIF-based GPP in addition to radar and optical imagery reduce validation RMSE by almost 10% and the standard deviation by 40%. Finally, we apply our model to measure losses in AGB from the recent 2021 Caldor wildfire in California, and validate our analysis with Sentinel-based burn index.