Research

Tropical forest disturbance monitoring

This research is mainly based on real-time monitoring of tropical forests by harmonizing Landsat and Sentinel-2 data. The key innovation combines change detection (BFAST Monitor) with machine learning (random forest) for earlier forest disturbance warnings.

Publication:
Sub-annual tropical forest disturbance monitoring… (Chen et al. 2021)

Carbon stock estimation in tropical secondary forests of Brazil

Using 29,000+ patches, this study uses GWR to estimate AGB and maps spatial variability in AGB recovery across Brazil.

Publication:
Revealing Spatial Variation in Biomass Uptake… (Chen et al. 2024)

Biomass and tree cover of regrowing forests in Brazil

Satellite analysis of 3,060 sites in Brazil reveals how regrowth is shaped by environmental and human pressures.

Publication:
Characterizing AGB and Tree Cover… (Chen et al. 2023)

Mapping mangroves using Sentinel-2

This study mapped mangroves in Dongzhaigang, China with Sentinel-2, CMRI, PCA, and random forest, achieving 90.47% accuracy.

Publication:
Mapping Mangrove Forests… (Chen 2020)