Abstract
Satellite remote sensing and geographical information system (GIS) have been used successfully to monitor and assess the land use and land cover (LULC) dynamics and their impacts on people and the environment. LULC change detection is essential for studying spatiotemporal conditions and for proposing better future planning and development options. The current research analyzes the detection of spatiotemporal variability of spate irrigation systems using remote sensing and GIS in the Khirthar National Range, Sindh Province of Pakistan. We use Landsat images to study the dynamics of LULC using ArcGIS software and categorize five major LULC types. We obtain secondary data related to precipitation and crop yield from the provincial department of revenue. The maximum likelihood supervised classification (MLSC) procedure, augmented with secondary data, reveals a significant increase of 86.25% in settlements, 83.85% in spate irrigation systems, and 65% in vegetation, and a substantial negative trend of 39.50% in water bodies and 20% in barren land during the period from 2013 to 2018. Our study highlights an increase in settlements due to the inflow of local population for better means of living and an increase in spate irrigation systems, which indicates the water conservation practices for land cultivation and human purpose lead to the shrinkage of water bodies. The confusion matrix using Google Earth data to rectify modeled (classified) data, which showed an overall accuracy of 82.8%-92%, and the Kappa coefficient estimated at 0.80-0.90 shows the satisfactory results of the LULC classification. The study suggests the need to increase water storage potential with the appropriate water conservation techniques to enhance the spate irrigation system in the hilly tracts for sustainable developments, which mitigates drought impact and reduces migration rate by providing more opportunities through agricultural activities in the study area.