Skip to content

Machine Learning-Based Prediction of Soil Moisture in Sikkim's High-Rainfall Zones Using Multimodal Remote Sensing Data

Sep 2026 · Journal of Agricultural Engineering · 0 citations · 24 references

TL;DR

The strong predictive performance demonstrated the effectiveness of integrating high-resolution Sentinel-2 imagery data with ANN for accurate, scalable, and efficient soil moisture estimation in high-relief, data-sparse environments.

Abstract

Soil moisture is an important variable influencing agricultural productivity, hydrological processes, and land management, particularly in high-rainfall regions such as the North Eastern Hill (NEH) States of India. Although conventional soil moisture measurement techniques provide reliable observations, they are time-consuming, labour-intensive and limited in spatial coverage, restricting their applicability for regional-scale monitoring. Remote sensing integrated with machine learning provides a promising alternative for generating spatially continuous and timely soil moisture estimates. This study aimed to predict surface soil moisture in the Ranipool-Rumtek administrative region of Sikkim, India, using multimodal remote sensing data and machine-learning techniques. Soil moisture was measured at 85 locations using the gravimetric method and these same locations were used as ground truthing sites. Multi-temporal imagery from Landsat-8 and Sentinel-2 was processed to derive vegetation and moisture-related indices, i.e., Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Normalized Difference Moisture Index (NDMI), Normalized Shortwave-infrared Difference Soil Moisture Index (NSDSI3), Land Surface Temperature (LST), Moisture Stress Index (MSI), and Vegetation Supply Water Index (VSWI). These indices were used as predictor variables to develop artificial neural network (ANN), support vector machine (SVM), and multiple linear regression (MLR) models. The results revealed that the ANN model developed using Sentinel-2-derived indices exhibited the highest predictive accuracy, achieving values of coefficient of determination (R2) as 0.82, Root Mean Square Error (RMSE) as 6% and Mean Absolute Error (MAE) of 4.6%. The performance of the Sentinel-2-derived ANN model was found to be better than that of the Landsat-8-based ANN model as well as the SVM and MLR models developed using Sentinel-2 data. The strong predictive performance demonstrated the effectiveness of integrating high-resolution Sentinel-2 imagery data with ANN for accurate, scalable, and efficient soil moisture estimation in high-relief, data-sparse environments. The proposed approach provides a robust framework to support precision agriculture, irrigation scheduling, hydrological modelling, and drought and flood monitoring.

View source

Similar papers

Open access Aug 2026

Non-Growing Season Surface Soil Salinity Estimation: Integrating Multi-Source Remote Sensing Data and Convolutional Neural Network Models in Arid Agricultural Areas

Soil salinization reduces crop productivity and threatens agricultural sustainability in arid regions. Reliable estimation of farmland soil salinity is therefore essential for salinization monitoring and land management. During the non-growing season, limited crop cover increases soil surface exposure. Existing studies...

Wan-Zhi Zhou, Xinjun Wang, Wen-Li Dong et al. · 0 citations
Open access Oct 2026

Cross-Regional Digital Soil Mapping of OpenLandMap-Derived USDA Soil Texture Classes Using Remote Sensing and Machine Learning

Results show that the Landsat-based spectral and terrain features, when used in conjunction with ensemble learning, could assist cross-regional soil texture classification for the USDA, however, class imbalance, spectral overlap between soil texture classes, temporal variability, and resolution mismatch between satelli...

Umair Maqsood, Fatima Iftikhar, Kashif Sattar et al. · 0 citations
Open access Sep 2026

Machine learning-based prediction of soil aggregate stability for sustainable land management in the eastern Himalayan region of India

Plot-scale rainfall-induced soil erosion data are scarce in the eastern Himalayas of India. Soil aggregate stability (SAS), commonly quantified by mean weight diameter (MWD), is a key indicator of erosion susceptibility. As laboratory determination of MWD is labor-intensive, machine learning (ML)-based pedotransfer...

Suryakant Gupta, Prachi Yadav, Jagabandhu Panda et al. · 0 citations
Open access Sep 2026

Downscaling SMAP Soil Moisture to 1 km with Machine Learning and MODIS Data for Agricultural Drought Assessment in Békés County, Hungary

Accurate mapping of Soil moisture (SM) is essential for effectively monitoring agricultural drought. However, the coarse spatial resolution of passive microwave products, including the 9 km Soil Moisture Active Passive (SMAP) retrievals, limits their effectiveness at regional and local scales. To address this limitatio...

Mahrokh Shafiei, István Waltner, Z. Vekerdy et al. · 0 citations
Open access Sep 2026

A novel hybrid remote sensing–machine learning framework for multiday soil moisture forecasting

The proposed method performed better than the conventional LSTM algorithm in all forecasting scenarios and showed robust performance even at a 7-day forecasting lead time, showing promise for applications in short-range soil moisture prediction and environmental monitoring studies.

S. Samadianfard, E. Khajeh, Neda Beirami et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.