Analysis of the relationship between settlement density & land surface temperature using multi-sensor RS data
Berger, Christian; Baumgart, Caroline; Voltersen, Michael; Walde, Irene; Hese, Soeren; Schmullius, Christiane
University of Jena, GERMANY
This study investigates the relationship between settlement density (SD) and land surface temperature (LST) for the City of Cologne, Germany. To this end, a four-stage methodology is presented that makes use of multi-spectral Ikonos imagery, an object height model derived from light detection and ranging (LiDAR) data, as well as thermal information provided by the Landsat 7 satellite mission. In a first step, a geographic object-based image analysis (GEOBIA) approach is employed to infer six basic urban land cover (LC) classes from the Ikonos and LiDAR data. In a second step, and based on the extracted LC and object height information, existing and new measures of SD are computed that take into account horizontal and/or vertical characteristics of a city. In a third step, multi-temporal LST maps are produced from the Landsat Enhanced Thematic Mapper (ETM) data. In the fourth and final step, the values of the SD measures are compared to those obtained from the LST maps to analyse their mutual correlation. The results of this study are described and discussed with regard to their dependency on the predominant type of urban land use (LU) and the acquisition date (season) of the Landsat ETM data.