Detecting and Analyzing Urban Heat Island Patterns in Beijing, China

The study reports a research aiming to detect the spatial pattern of urban heat island (UHI) phenomena in Beijing, China, and to determine the relationship between the UHI pattern and a set of landscape and demographical variables. The study area has been one of the fastest growing metropolises in China during the past two decades. Land surface radiance temperature (SRT) values were extracted from a Landsat TM image by 2001. Built-up area and road system were got from 1:10,000 scale topographic maps. A NDVI index map was extracted from the same image. Multiple correlation and regression analyses were used to examine the relationship between UHI pattern and land use/cove parameters (i.e. NDVI, built-up density and road density) and demographical variables (i.e. population density, non-agricultural population density and family households density) at the township level. It is found that dry river bed area across city was hottest, but multiple heat islands emerged over several satellite cities within the metropolis. The statistical analysis indicates that the UHI intensities had a negative relationship with NDVI, but positively correlated with built-up ratio, non-agricultural population density and road density. Interestingly, the stepwise regression analysis indicates that population density did not show any meaningful statistical relationship with the UHI intensity. * Corresponding author. [email protected]