Correlation modelling between land surface temperatures and urban carbon emissions using multi-source remote sensing data: A case study

Global warming has imposed substantial global negative impacts on different sectors of human societies, such as extreme weather events. In this sense, it is imperative to ascertain whether the rise in global temperature will accelerate carbon emissions simultaneously. The land surface temperature (LST) serves as a common indicator to represent the spatial temperature and urban areas account for a majority portion of global emission. To fill this gap, selecting the central urban area of Fuzhou City as a study case, this paper aims to examine the potential correlation between LST and overall carbon emissions, based on the Land use and cover change data and nighttime lighting remote sensing data in three years (2012, 2016, and 2020). A spatially explicit distribution model of carbon source is presented in this paper. Based on Remote sensing data and Land use and cover change data, this model used inverse distance weighting spatial interpolation to calculate urban carbon emissions and retrieve LST. Moreover, the potential statistical correlation between Land surface temperature (LST) and urban carbon emissions is explored by both polynomial and spline regressions. In summary, in general, a positive statistical correlation between LST and carbon emissions is revealed in this case. Specifically, urban carbon emissions generally statistically increase when LST varies between 26 °C and 33 °C, remaining stable for the rest of the temperature ranges (i.e., below 26 °C or above 33 °C). This indicates a potential marginal effect, which requires verifications in future research.