Authors: Chao Fan*, University of Idaho
Topics: Spatial Analysis & Modeling, Remote Sensing, Climatology and Meteorology
Keywords: spatiotemporal modeling, urban heat island, remote sensing, spatial dependence, time series
Session Type: Paper
Start / End Time: 1:20 PM / 3:00 PM
Room: Lafayette, Marriott, River Tower Elevators, 41st Floor
Presentation File: No File Uploaded
Human-induced landscape transformation profoundly alters the surface energy balance that affects our living environment in many different levels. Of the most direct consequences is the significantly heightened temperature in urban areas due to the introduction of large quantity of dry and impervious materials in lieu of moist and pervious natural landscapes. Land cover impact on a city’s skin temperature has been well-documented through cross-sectional studies of numerous cities across the globe. While understanding the association at one point in time is essential, a systematic study of the longitudinal trend of the relationship provides a deep insight into a city’s climate pattern and permits analysis of such pattern in the context of an evolving urban landscape. This study uses sequential Landsat imagery coupled with spatial statistical approaches to explore the surface temperature variability with respective to the spatiotemporal urbanization pattern in the Phoenix metropolitan area over the time period from 1991 to 2010. Spatial indicators measuring local concentration of vegetation and built-up areas are generated from annual Landsat imagery. For a given year, Landsat scenes acquired at summertime are used to derive the land surface temperature. In quantifying the landscape-temperature association in a panel setting, we employ a spatial seemingly unrelated regression equations framework that explicitly accounts for both the spatial dependence in a cross section of spatial units and the serial dependence in the intra-equation disturbances. Results from this study will shed light on future urban planning strategies towards effective alleviation of the well-known heat island effect in the region.