Hari Venkata Durga Rao Korada

Accurate surface water monitoring across India’s diverse landscapes is often hindered by the limitations of pixel-based classification and static indexing. This study introduces an object-based framework integrating DBSCAN clustering with supervised learning classification for automated water extraction using Sentinel-2 imagery. To ensure regional generalizability, the model was trained on a multi-site dataset spanning 20 distinct geographical locations across India. Out of 5 machine learning (ML)models, XGBoost outperformed other ML architectures, including Random Forest and state vector machine (SVM), by effectively minimizing loss in complex spectral zones. Methodological Benchmarking demonstrated that the proposed DBSCAN-XGBoost framework significantly surpassed both traditional NDWI thresholding and pixel-based XGBoost classification, achieving a superior overall accuracy of 97% and an F1-score of 0.96. Feature importance analysis via the Gain metric underscored the synergistic role of mean NDWI and mean NIR in refining decision boundaries. By shifting to an object-oriented paradigm, the proposed model substantially reduced “salt-and-pepper” noise and partially mitigated topographic shadow errors. This research presents a promising object-based framework that, after further multi-site validation, could support high-fidelity water auditing across diverse Indian landscapes.

Stream flow forecast and its inundation simulations prior to the event are an effective and non-structural method of flood damage mitigation. In this paper, a continuous simulation hydrological and hydrodynamic model was developed for stream flow forecast and for spatial inundation simulation in Brahmani–Baitarani river basin, India. The hydrologic modelling approach includes rainfall-run-off modelling, flow routing, calibration and validation of the model with the field discharge data. CARTOSAT Digital Elevation Model of 30 m resolution, land use/land cover derived from the Indian Remote Sensing Satellite (IRS-P6) AWiFS and soil textural data of the study area were used in the modelling to compute topographic and hydraulic parameters. The hydrological model was calibrated with the help of field observed discharge data of 2006 and 2009 and validated with the data of 2008 and 2011. From the results, it is found that computed discharges are very well matching well with the observed discharges. The developed model can provide the stream flow forecast with more than 30 h lead time. Possible flood inundations were simulated using hydrodynamic modelling approach. CARTO Digital Elevation Model of 10 m resolution, landuse and the computed flood hydrographs were used in inundation simulations.

The accurate assessment of soil erosion in the spatial environment can give valuable input in designing reservoirs and for preparing watershed management action plans. Rangit River in northeastern India is affected by severe soil erosion, causing reduction in its reservoir capacity. The main aim of the paper is to estimate the soil erosion in the Rangit catchment, India through a process-based soil erosion model that integrates remote sensing and field data in the geospatial domain, and also to identify the problem areas for further watershed management action plans. Various topographic parameters that were derived from land use, soil texture and a digital elevation model were used for the estimation of soil erosion. The Morgan, Morgan and Finny model was used to compute soil erosion. Based on the drainage pattern, the catchment was divided into eight sub-catchments for detailed study. Annual soil erosion was estimated for four years from 2002 to 2005 using the corresponding year’s meteorological data. Estimated soil erosion was compared with the field data in each year and found very good correlation.