Simulation of land use dynamics and impact on land surface temperature using satellite data

Abstract

Increasing of global average surface temperature naturally leading to major problems as global warming, which has typically attracted the interest of multinational organizations, civil society groups and scientists in the world. The main aim of this research is to properly investigate the impact of urban development via land-use dynamics on the future micro-climate of Beijing by predicting land surface temperature (LST) distribution and simulating land-use dynamics. Land use/land cover (LULC) and LST were traditionally produced using multispectral and thermal bands. Land use dynamics were calculated using a statistical model. LST of each LULC was mapped to assess temporal changes of LST responding to LULC changes in the study area for 20 years over three periods (1997–2005, 2005–2017 and 1997–2017). Prediction of LST was employed using single and multiple linear regression models until 2027. The results show that the average LST increased from 22.06 to 35.85 °C, from 20.98 to 37.04 °C, and from 17.00 to 38.29 °C in 1997, 2005 and 2017, respectively. This can be explained by the population growth and change in LULC. In addition, the comprehensive index of land use degree showed significant changes in land use of the area caused by several human activities. The annual rate of the urban area showed a remarkable increase as a direct result of urban sprawl. Urbanization may impact negatively on vegetation, causing it to decrease progressively to 2017. Furthermore, the results of the prediction models show that the Urban Index and Modification of Normalized Difference Water Index are highly influential to the forecasting of LST. The study has also demonstrated that remote sensing and GIS techniques are useful tools for assessing spatial and temporal changes in LULC dynamic and its impacts on LST.