Investigation of land cover changes in relation with anthropogenic presence and its impact on climate change of Mansehra District, Pakistan

Abstract

Mapping land use land cover (LULC) change by using Remote Sensing data (RS) and Geographic Information System (GIS) has gain a great concern in current strategies for managing natural resources. Both natural and anthropogenic activities are modifying the land cover which needs to be mapped and quantified. The objective of this study was to map the LULC in Mansehra district during 2000-2016 in relation with population, and finding its relationship with the climatic variables like temperature and precipitation. Moreover, climate variables were also incorporated to find their impact on snow cover. This study utilized satellite Landsat (TM, ETM+ and OLI) data for mapping land cover changes and MODIS snow data (MODI 0A2) for finding the change in snow cover area. Two different sources of climate data were used i.e. observed data from Pakistan Meteorological Department (PMD) and Water and Power Development Authority (WAPDA) and gridded atmospheric reanalysis data (ERA-Interim) for finding trend of climate variables and their comparison from these two different sources. Primarily, LULC maps of year 2000, 2005, 2010 and 2016 were prepared using supervised classification approach and the temporal change from 2000- 2016 was quantified by post-classification comparison. LULC change showed significant increase in built-up (+327.5%), cultivated area (+49.18%) and decrease in forest cover (- 13.4%). Population also showed increase of 16% from 1998 to 2017 which also corresponds to increase in built-up. Mann-Kendall trend was used to find the trend of climatic variables and snow. Temperature showed positive (increasing trend) while precipitation showed negative (decreasing trend) at all stations both annually and seasonally. In comparison with the PMD stations (low-elevation stations), ERA-Interim underestimated the temperature values (positive bias) while the data obtained from WAPDA stations (high elevation stations). it overestimated the average temperature (negative bias). In case of precipitation, negative bias is shown which clearly indicates that ERA-Interim overestimates the precipitation values in comparison with the ground observation. Bias occurs between ground station and ERA-Interim values but overall same pattern of trend line is traced in case of temperature. However, in case of precipitation, trend pattern somewhat differs for different years. A strong Correlation was observed between ground-station and ERA-Interim temperature values in comparison with precipitation data (weak correlation) which shows that ERA-lnterim temperature product can be used as a reliable data for capturing temperature trends. Snow cover didn ·t show any significant trend.