Improving Forest Loss Mapping In Nepal Using Landtrendr Time-Series And Machine Learning
--
https://doi.org/10.1016/j.rsase.2025.101864 <-- share paper
--
“HIGHLIGHTS:
ViT-based forest mask, multispectral ensemble LandTrendr and terrain shadow mask.
District-level RF/XGBoost model training with expert-weighted validation.
Outperformed GFC and REDD + AI benchmarks in accuracy and F1 performance.
RF excelled in High Mountains/Himalayas; XGBoost in the lower Mountain regions.
NBR contributed the most; snow-impacted forest loss uncertainty was observed.
ABSTRACT: Understanding forest disturbances is essential for effective conservation strategies. Given Nepal’s complex geography and forest ecology, change detection using Remote Sensing remains challenging, with limited time-series studies. This study introduces an enhanced LandTrendr (LT) workflow to improve forest loss mapping using medium-resolution imagery and machine learning. The approach includes: a) a Vision Transformers model (LiteForest-ViT) for semi-automated forest cover mask using Landsat 5, b) masking terrain shadows, c) ensemble of 7 spectral indices: NBR (Normalized Burn Ratio), NDVI (Normalized Difference Vegetation Index), TCA (Tasseled Cap Angle), TCB (Tasseled Cap Brightness), TCG (Tasseled Cap Greenness), EVI (Enhanced Vegetation Index), TCW (Tasseled Cap Wetness) with 6 LT-derived metrics for Random Forest (RF) and eXtreme Gradient Boosting (XGBoost) classification, d) expert-weighted district-level model selection tailored to regional heterogeneity, e) integration of multiple platforms for seamless processing, and f) MODIS-derived snow uncertainty loss estimation. The study spans (1995–2024) across Karnali, Bagmati, and Darchula. Results indicate RF edged XGBoost in the High Mountains and Himalayas, while XGBoost did better in the Siwalik and Middle Mountains. NBR was the most influential index regardless of model classifier and region. The algorithm achieved 0.90 overall accuracy, 0.74 kappa statistics, and 0.93 F1-score, exceeding GFC (Global Forest Change) and REDD + AI (CTrees) benchmarks. Overall, 7,870 ha of forest loss were detected, where ∼165 ha accounted for snow-impacted uncertain loss. While loss has decreased, continued disturbance underscores the significance of [their] findings to support REDD+ (Reducing Emissions from Deforestation and Forest Degradation) in the region…”
#Forestdisturbance #forest #disturbance #remotesensing #LandTrendr #workflow #timeseries #ViT #RF #XGBoost #GEE #Nepal #ForestNepal #spatial #GIS #mapping #earthobservation #landsat #Himalayas #mountains #alpine #vegetation #AI #multispectral #monitoring #spatialanalysis #spatiotemporal #loss #change #machinelearning #NDR #conservation #planning #policy #mitagion #ecology #Karnali #Bagmati, #Darchula #Siwalik #GlobalForestChange #Degradation