Automatic filtering of ERT monitoring data in mountain permafrost
ABSTRACT Continuous monitoring of Electrical Resistivity Tomography (ERT) surveys can be a powerful tool for all kind of long‐term applications in the field of hydrogeophysics and cold‐region geophysics due to its high sensitivity to changes in water and ice content of the near subsurface. However,...
| Published in: | Near Surface Geophysics |
|---|---|
| Main Authors: | , , , |
| Format: | Article in Journal/Newspaper |
| Language: | English |
| Published: |
Wiley
2013
|
| Subjects: | |
| Online Access: | http://dx.doi.org/10.3997/1873-0604.2013003 https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.3997%2F1873-0604.2013003 https://onlinelibrary.wiley.com/doi/pdf/10.3997/1873-0604.2013003 |
| Summary: | ABSTRACT Continuous monitoring of Electrical Resistivity Tomography (ERT) surveys can be a powerful tool for all kind of long‐term applications in the field of hydrogeophysics and cold‐region geophysics due to its high sensitivity to changes in water and ice content of the near subsurface. However, the large amount of data often calls for autonomous data processing schemes. In this study, a new filter algorithm designed to automatically detect and delete measurement errors from multiple ERT monitoring data is presented. Three successive filter steps were developed in order to eliminate technical errors, overall high‐value outliers and relative outliers within single data levels. The filter procedure is site‐independent and was tested on four different mountain permafrost sites in the Swiss Alps, representing various landforms (talus slope, rock plateau, rock glacier, bedrock). The filter performance is assessed by analysing the effect of the filter procedure on the mean apparent resistivity and on the resulting data misfit of the inversion and both, after the entire filter procedure as well as after each individual filter step. The new filter procedure is expected to yield rapid and high‐quality filtering for monitoring applications. In our study, the procedure is developed to support early detection of electrical resistivity changes associated with freezing and thawing events in permafrost conditions. The filter is applied to 128 ERT data sets from permafrost monitoring stations in Switzerland, including a four year long (2005–2008) ERT monitoring data set from the high‐mountain permafrost monitoring station Stockhorn, which serves as an illustrating example. |
|---|