Assessing land use/land cover change detection of north-eastern watersheds of Kashmir valley using GIS and remote sensing techniques

Abstract

Land use changes in the Kashmir Himalayas must be mapped and monitored for long-term development and efficient planning. This work uses geospatial technologies like Remote Sensing and Geographic Information System to track changes in Land cover trends in four main watersheds in the Kashmir Valley's north-eastern Himalayas from 2003 to 2013. Land cover maps were created using images from the Landsat-5 Thematic Mapper. The watershed's Land use/Land cover maps were created using supervised classification utilizing the Maximum Likelihood Classifier (MLC). Over the previous decade, the land use/land cover in the study watersheds has undergone a series of intricate changes as a result of deforestation, climate change, and agroforestry growth. A total of eleven major LULC classifications were discovered, indicating that forests are the most common land use in all four watersheds. Forest cover, river beds, water bodies, non-perennial snow, and glaciers have all decreased significantly, whereas Scrubland, horticulture, rock mass, built-up, barren land, and agriculture have all increased significantly, except for Sindh watershed, where Scrubland class has decreased by 5.97% from 2003 to 2013. The study's methodology and conclusions point to crucial policy implications for long-term LULC management in the Kashmir Himalayas' Madhumati, Arin, Sindh, and Lidder watersheds.

Towards Improving Sustainable Water Management in Geothermal Fields: SVM and RF Land Use Monitoring

“…The SVM algorithm uses the parameters cost c, hyperparameter sigma, number of support vectors, and training error. In contrast, the RF algorithm uses the parameters mtry, number of trees, number of variables tried at each split, and OOB estimate of error rate [52][53][54][55]. The model is then evaluated using a confusion matrix and Cohen's Kappa to assess accuracy [56,57].…”