Work

Partitioning and synergistic estimation of leaf area index in dryland forests using dual-sensor Sentinel 1 and 2 imagery

Year: 2026
Type: article
Abstract: Leaf Area Index (LAI) is a key biophysical parameter that describes forest canopy structure and function. While satellite-based approaches typically estimate total ecosystem LAI, partitioning it into forest overstory (LAIOverstory) and understory (LAIUnderstory) components remains challenging, particularly in dryland forests where open canopies and strong soil background effects influence remote sensing retrievals. A multi-sensor framework for estimating and partitioning LAI was developed and validated using Sentinel-1 (S1; SAR) and Sentinel-2 (S2; optical) imagery. The models were calibrated and validated against 117 optical (SunScan®) field measurements conducted during 2017-2023 across two long-term ecological research (LTER) sites with different overstory thinning treatments, along with 19 long-term monitoring (LTM) plots in east-Mediterranean Pinus halepensis and Pinus brutia forests, spanning an arid to dry-subhumid climatic range. Partial Least Squares Regression (PLS-R) models demonstrated strong predictive performance. Sentinel-2 predictors showed the highest sensitivity to LAIOverstory (R2 = 0.78, RMSE = 0.66). In contrast, Sentinel-1 C-band VH backscatter captured LAIUnderstory variability (R2 = 0.90, RMSE = 0.22). The combined S1 + S2 model provided the most consistent predictions, showing improved performance for both LAIOverstory (R2 = 0.85, RMSE = 0.53) and LAIUnderstory (R2 = 0.92, RMSE = 0.20), reflecting the complementary sensing mechanisms of optical and SAR observations. Thinning strongly influenced the vertical distribution of LAI in semiarid forest stands, with LAIUnderstory accounting for 16% to 45% of total LAI, depending on thinning intensity. Comparison with widely used satellite-derived LAI products (GCOM-C SGLI, PROBA-V, Sentinel-3 OLCI, MODIS, and SL2P at 10 m and 20 m) showed stronger agreement with effective PAI than with LAI, indicating systematic underestimation of ecosystem LAI in dryland forests. These findings highlight the importance of resolving vertical canopy structure for improving ecosystem monitoring and modeling in water-limited environments.
Source: Figshare
Language: English
Subfield: Ecology
Open Access status: green
OpenAlex ID: W7203773751
FWCI:
Cites:
Cited by:
Related to: 10