Remote Sensing of Urban Groundwater Temperatures

Abstract:

In urban settlements temperatures are typically elevated compared to their rural surrounding. This so-called urban heat island (UHI) phenomenon exists in all diverse layers of modern cities such as surface and subsurface. While both are typically investigated individually, the coupling of these zones is not sufficiently understood. Using satellite derived land surface temperatures (LST) and interpolated groundwater temperature (GWT) measurements we compare the spatial properties of both heat islands in four German cities and analyze their dissimilarities in order to understand the interaction between the urban surface and subsurface. The best correlation of more than 80% is found in old, mature cities such as Cologne and Berlin. However, local groundwater hotspots under the city center and industrial areas are not revealed in satellite derived land surface temperatures. Overall groundwater temperatures are higher than land surface temperatures in 95% of the analyzed area due to the influence of below ground anthropogenic heat sources such as elevated basement temperatures. Thus, an estimation method is proposed that relates groundwater temperatures to mean annual land surface temperatures, building density and elevated basement temperatures. Using this method regional groundwater temperatures can be accurately estimated with a mean absolute error of 0.9 K.

Since land surface temperatures and building densities are available from remote sensing, this method has the potential for a large scale estimations of urban groundwater temperatures. Thus, it is feasible to detect subsurface urban heat islands (SUHI) on a global level. Furthermore, geothermal potentials and groundwater quality issues such as thermal stress to groundwater ecosystems can now be investigated using satellite derived data.