Historical and future land-cover changes in the Upper Ganges basin of India

Abstract

The green revolution represents one of the greatest environmental changes in India over the last century. The Upper Ganges (UG) basin is experiencing rapid rates of change of land cover and irrigation practices. In this study, we investigated the historical rate of change and created future scenario projections by means of 30 m-resolution multitemporal Landsat 5 Thematic Mapper and Landsat 7 Enhanced Thematic Mapper Plus data of the UG basin. Post-classification change analysis methods were applied to Landsat images in order to detect and quantify land-cover changes in the UG basin. Subsequently, Markov chain analysis was applied to project future scenarios of land-cover change. Fifteen different scenarios were generated based on historic land-cover change. These scenarios diverged in terms of future projections, highlighting the dynamic nature of the changes. This study has shown that between the years 1984 and 2010 the main land-cover change trends are conversion from shrubs to forest (+4.7%), urbanization (+5.8%), agricultural expansion (+1.3%), and loss of barren land (-9.5%). The land-cover change patterns in the UG basin were mapped and quantified, showing the capability of Landsat data in providing accurate land-cover maps. These results, in combination with those derived from the Markov model, provide the necessary evidence base to support regional land-use planning and develop future-proof water resource management strategies.

Assessment of Ecosystem Services Value Based on Land Use and Land Cover Changes in the Transboundary Karnali River Basin, Central Himalayas

“…Rapid economic and development activities are major factors for LUCC [69][70][71][72]. Urban area expansion of the Kathmandu region and agricultural expansion were the major reasons for LUCC in the Koshi basin [28].…”

Application of landscape metrics and a Markov chain model to assess land cover changes within a forested watershed, Taiwan

“…Three basic assumptions have to be made before applying the Markov chain analysis: (1) land cover distribution is statistically dependent only on the distribution of the previous period, (2) this dependence is a first‐order Markov dependence, and (3) the system is time stationary (Muller and Middleton, ; Tsarouchi et al ., ). A Markov chain, defined as a stochastic process, shows that the value of a process at time t , X t , depends only on its value at time t − 1, X t − 1 , and not on the sequence of values X t − 2 , X t − 3 , …, X 0 that the process passed through before arriving at X t − 1 (Weng, ).…”

Land-use change may exacerbate climate change impacts on water resources in the Ganges basin

“…They were subsequently classified using a Maximum Likelihood classifier method with pixel training data sets, resulting into a land cover map of eight different classes. For more information see Tsarouchi et al (2014).…”

“…Land-cover map of the UG basin, for year 2010, as developed by Tsarouchi et al (2014). Landsat 7 ETM+ images, for October 2010 were acquired from the US Geological Survey Global Visualization Viewer.…”

“…are not large and the main trends of change identified are forest growth, urbanisation, and on the other hand loss of bare soil, grasslands and shrubs. For a more detailed description of the method used to generate the future land-use scenarios see Tsarouchi et al (2014). …”