Transient thermal effects below complex topographies

The topographical perturbation of steady-state subsurface temperature fields has been an important issue in geothermal interpretations throughout the past century. This paper reports a numerical study, which considers the possible influence of terrain topography on transient temperature signals. Typical morphological situations over wide areas in central Europe affect most likely that same depth range which also contains the temperature signals resulting from the most interesting ground surface temperature changes (i.e. during the last 200 years). The evaluation of the interaction is performed on a synthetic sinusoidal topography with varying wavelength and amplitude. Vertical profiles (i.e. temperature logs) were extracted from these numerical forward 2-D calculations. Thereby, the error could be estimated by comparing ground surface temperature time-histories inverted from temperature logs both with and without topographic correction. The results show that a topographic correction of temperature data is absolutely necessary to achieve a consistent inversion result. Even rather flat topographies with 20-km wavelengths and 100-m amplitudes may introduce topographical effects which confuse the inversion process. On the other hand, the palaeoclimatically induced temperature signal persists even in rough topographies and will show correct inversion results when data are adequately treated. Only extreme situations cause a lateral interference of these transient signals with depth. The results from such 2-D synthetic models have been confirmed by an analysis of a real situation. The example chosen is the area surrounding the German Continental Deep Drilling (KTB) project. The area is situated in a moderately undulating surface topography with maximum altitude variations of the order of 250 m. The additional 3-D simulation demonstrates that a strong topography-dependent variation of the transient temperature signal can occur even at greater depths. The introduction of corrections for topography influence, reduces apparent differences in profiles from different locations in the surroundings of the KTB site to a maximum of 0.2 K. (C) 1999 Elsevier Science B.V. All rights reserved.