Time series assessment of the relationship between land surface temperature due to change in elevation: a case study from Hindukush-Himalayan Region (HKH)

Abstract

Environmental studies that include studies related to climate change, change in biodiversity, and hydrology consider land surface temperature (LST) and near-surface air temperature as significant variables to contribute. In this context, we investigated the time series assessment of the relationship between LST and elevation over the two decades (1995-2017) using remotely sensed (Landsat 5, Landsat 7, Landsat 8, MODIS) and ground weather stations' datasets. The study area that is Gilgit-Baltistan, northern province of Pakistan in Hindukush and Himalaya Region (HKH), was selected for quantitative analysis in this study as it has high significance in terms of climate change due to presence of very high and large number of mountains and difficult terrain heavily covered with snow and glaciers. To assess the LST, a linear regression model was developed to quantify the significance of key factors affecting it. Our research results show that there exists a strong negative linear relationship (with correlation coefficient of 0.61 on average) for all datasets (remotely sensed and in situ) between LST and elevation and it is consistent over the study period (1995-2017) for both seasons of winter and summer. Further, there is a difference of change in LST for different seasons (average winter and summer difference is 3.0 degrees C) and for different datasets (Landsat, MODIS, and weather stations' dataset). There is a greater change in LST for change of elevation in 1000 m vertically as compared to change horizontally. Our results indicate that the change trend in LST from 1995 to 2007 is downward with minor variation, however, this trend changes into upward trend from 2007 to 2017. The comparison of results of remotely sensed and weather stations' dataset shows that there is a consistency of change in temperature and LST due to change in elevation. However, there is a less difference in summers (0.76 degrees C) and large difference in winters, i.e., 3.94 degrees C.