Estimating long-term LULC changes in an agriculture-dominated basin using CORONA (1970) and LISS IV (2013-14) satellite images: a case study of Ramganga River, India

Abstract

The basin of Ramganga River, a major tributary of Ganga, has seen rapid industrialization, urbanization, and agricultural growth in modern times, especially during and after the 1970s, with consequent changes in Land Use and Land Cover (LULC) of the basin. Object-based classification of seldom-used CORONA images (for 1970) for historical LULC scenario and LISS IV (Resourcesat-2) images (for 2013-14) for the present scenario of the basin was performed with high overall accuracies of 93% and 87% and Kappa coefficients of 0.89 and 0.82 respectively. Object-based classification using the eCognition software tool combined with manual classification of the images revealed significant decrease in total forest cover (~ 485 km²) and riverbed (~ 498 km²) LULC classes along with correspondingly increased agriculture (~ 898 km²) and built-up (~ 313 km²) classes in the basin. Further, though Ramganga reservoir (spread over ~ 70 km²), operated since in 1974, has increased the water area by ~ 1295% in Afzalgarh sub-basin of the Ramganga Basin, the overall area of water class decreased by 1.64% in the basin, probably due to loss of ponds in agricultural lands. An upward shift of about 42 km in the confluence point of Ramganga and Gangan Rivers was also observed, which may be attributed to the course shift of Ramganga in a south-west direction.

Keywords: CORONA; Channel-shift; LISS IV; LULC; Object-based classification; Ramganga.