Spatio-temporal variability in remotely sensed LST and its impacts by FVC in the Greater Khingan Mountains

Abstract

The present study explored both the temporal variation and spatial distribution of fractional vegetation cover (FVC) and land surface temperature (LST) in the Greater Khingan Mountains, a location distinguished by four types of surface cover formations and strong gradients in meteorological conditions. Furthermore, we assessed the relationships between FVC and LST in different time-space dimensions. We measured the spatio-temporal variability in LST through a harmonic analysis of time series (HANTS) for 8 days of LST time series product data from the Moderate Resolution Imaging Spectroradiometer (MODIS). Furthermore, the FVC was extracted through linear spectral mixture analysis (LSMA). The results show that the time series image data reconstructed via HANTS are effective for quantifying the morphological dynamics of the thermal environment and enabling sound calculations of environmental variables, such as vegetation abundance. We found significant differences in within-year and interannual variation among different eco-geographical regions. The within-year maximum value of vegetation coverage was observed in July, whereas the surface temperature was highest in June. This finding suggests that the decrease in warming can be mostly attributed to the increase in evapotranspiration associated with increased vegetation activity. In addition, a strong negative correlation was found between the FVC and LST throughout the study area only from April to September, whereas a triangular relationship was found in other months. This study investigated the time series variations in different eco-geographical regions for 2000 – 2015 and revealed that the most obvious LST and FVC variations occurred at the region II and junction between regions III and IV. An analysis of the average FVC and LST from April to September indicated that 86.9% of the entire study area showed a negative correlation, such that when the FVC increased by 10%, the LST showed a maximum average decrease of 2.76°C. We conclude that the increase in vegetation activity is the main cause of the reduction in the LST and that surface activities and latitude zonality are the main driving factors of the LST differentiation in the Greater Khingan Mountains. These findings will serve as a foundation for future studies seeking to better understand climate change processes and to estimate ecosystem responses to changing climatic conditions.