A SEQUENTIAL STRATEGY FOR HYDROGEOPHYSICAL INVERSION OF MULTIPLE GEOPHYSICAL DATASETS

Recently, geophysical methods have been widely used in hydrology to estimate the spatial distribution of some hydrological properties of the subsurface such as moisture content and porosity. Compared to traditional point measurements, geophysical methods provide a continuous image of the subsurface over scales from tens of centimeters meters to hundreds of meters and thus are much more efficient and effective for field hydrological studies. Despite many successful applications, it is, however, still challenging to interpret geophysical measurements as hydrological properties due to the non-uniqueness of geophysical inversion and uncertainty associated with petrophysical relationships of geological materials.

In this study, we propose a sequential strategy for hydrogeophysical inversion of seismic travel time and resistivity data. The first step of the method is to identify different zones of the subsurface based on the velocity image inverted from seismic travel time data. The second step is to determine the hydrological properties of the subsurface with a coupled concept, which incorporates petrophysical relationships directly in the inversion. In the forward and inversion, the adaptive irregular mesh is used and it allows us to include irregular topography and subsurface structures in the calculation. We designed several synthetic subsurface models with complex topography and structures, based on which, synthetic seismic refraction and resistivity tests are simulated. The “measured” geophysical data are processed using the traditional strategy and the proposed strategy. The traditional method also identifies different geological zones based on inverted velocity images but it directly inverts the apparent resistivity data to obtain the spatial distribution of resistivity, which is then translated into hydrological properties using known petrophysical relationships. The results show that the sequential strategy gives a better estimation of the hydrological properties within each geological zone. This probably due to that, by incorporating petrophysical relationships, the unknowns are better bounded in the inversion if comparing to the traditional method. The proposed method could be applied to hydrological studies that focus on the partitioning of precipitations in the critical zone.