Remote Sensing - Based Estimation of Evapotranspiration Using the Triangle Method: Insights from Tropical Forests of Northeast India
Year: 2026
Type: article
Abstract: Understanding the interactions between LST, vegetation dynamics, and ET is critical in the context of climate change and its impacts on terrestrial ecosystems and the hydrological cycle. This study investigates the seasonal variability of ET and its relationship with LST and NDVI over forested regions of Lakhimpur, Dhemaji, and Dibrugarh districts in Assam. The analysis is specifically restricted to forest areas using a Dynamic World land use/land cover mask to ensure ecological consistency and minimize land cover heterogeneity. Evapotranspiration was estimated using the NDVI–LST triangle method, which derives evaporative fraction (EF) from the LST–NDVI feature space through the identification of dry and wet edges. Landsat 8 data were used to generate high-resolution NDVI and LST, while net radiation (Rn) was obtained from ERA5-Land and resampled to match Landsat spatial resolution. The results were further compared with widely used datasets, including MOD16A2, GLDAS, and FLDAS, to assess consistency and variability. The findings reveal clear seasonal patterns, with ET peaking during the monsoon (up to ~3.17 mm/day) and decreasing during pre- and post-monsoon periods. Forested areas exhibit relatively stable NDVI and lower LST, contributing to consistent and spatially coherent ET patterns. MOD16A2 significantly overestimates ET, while GLDAS and FLDAS provide more stable and physically realistic estimates. The approach demonstrates strong potential for monitoring seasonal ET dynamics in data-scarce regions. Future work should focus on integrating in-situ observations, improving methodological frameworks, and adopting multi-source data fusion techniques to enhance the accuracy and reliability of ET estimation for climate and water resource applications.
Source: Journal of Geomatics
Institutions: Indian Institute of Remote Sensing, Dibrugarh University
Language: English
Subfield: Global and Planetary Change
Field: Environmental Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7224134430
FWCI:
Cites:
Cited by:
Related to: