EUVIMEDEUVIMED entdecken
Crossref

Machine Learning Model for predicting NDVI in the Northwestern Himalayas

Deepak Kumar Raj, Gopikrishnan T

Journal of Mountain Research · 28.02.2026

DOI: 10.51220/jmr.v21-s1.16 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Mountain Research, online erschienen am 28.02.2026. Verfasst von Deepak Kumar Raj und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Vegetation patterns across the Northwestern Himalayan region of India —comprising Jammu & Kashmir, Ladakh, Himachal Pradesh, and Uttarakhand—are profoundly affected by climatic variables such as land surface temperature (LST), precipitation, and evapotranspiration (ETo). This study aims to identify the most effective machine learning (ML) algorithm for predicting the Normalized Difference Vegetation Index (NDVI) using integrated datasets of LST, ETo, and precipitation. Climatic and vegetation data spanning 2002 – 2022 were utilized to analyze the interdependence among these parameters. Precipitation data were retrieved from the Center for Hydrometeorology and Remote Sensing (CHRS), while MODIS products available through the Google Earth Engine (GEE) platform were used to derive LST and NDVI. Four ML algorithms—Random Forest (RF), Gradient Boosting Machine (GBM), Support Vector Regression (SVR), and K-Nearest Neighbors (KNN)—were trained and compared for NDVI prediction. Their performance was evaluated using multiple statistical indicators, including the coefficient of determination (R²), root mean square error (RMSE), mean absolute error (MAE), mean square error (MSE), and explained variance score (EVS). Among all models, the RF algorithm exhibited the highest predictive accuracy, with consistently high R² and low RMSE values, confirming its superior capability to capture vegetation variability in the complex Himalayan terrain. The correlation analysis revealed strong positive associations between ETo, NDVI, and LST, along with moderate correlations between NDVI and precipitation. Overall, the study underscores the interconnected nature of these climatic variables and highlights ETo as a key driver influencing vegetation productivity and surface temperature variations across the Northwestern Himalayas.

Zitieren

APA

Raj, D. K., & T, G. (2026). Machine Learning Model for predicting NDVI in the Northwestern Himalayas. Journal of Mountain Research. https://doi.org/10.51220/jmr.v21-s1.16

Vancouver

Raj DK, T G. Machine Learning Model for predicting NDVI in the Northwestern Himalayas. Journal of Mountain Research. 2026. doi:10.51220/jmr.v21-s1.16

BibTeX
@article{Raj2026,
  author = {Raj, Deepak Kumar and T, Gopikrishnan},
  title = {Machine Learning Model for predicting NDVI in the Northwestern Himalayas},
  journal = {Journal of Mountain Research},
  year = {2026},
  doi = {10.51220/jmr.v21-s1.16}
}
Kostenlos starten →