Loading...
Please wait, while we are loading the content...
Similar Documents
Improved characterization through joint hydrogeophysical inversion: Examples of three different approaches
| Content Provider | Semantic Scholar |
|---|---|
| Author | Linde, Niklas Chen, Jinsong Kowalsky, Michael B. Finsterle, Stefan Rubin, Yoram Hubbard, Susan S. |
| Copyright Year | 2004 |
| Abstract | Improved characterization through joint hydrogeophysical inversion: Examples of three different approaches Niklas L i n d e n , Jinsong Chen*, Michael Kowalskyl, Stefan Finsterle^, Yoram Rubin^ and Susan Hubbard ^ 1 Earth Science Division, Lawrence Berkeley National Laboratory, MS 90-1116, 1 Cyclotron Rd., Berkeley, CA 94720, USA. ^Department of Civil and Environmental Engineering, University of California, Berkeley, CA 94720, USA. ^Department of Earth Sciences/Geophysics, Uppsala University, Villav. 16, 752 36 Uppsala, Sweden. With the increasing application of geophysical methods to hydrogeological problems, approaches for obtaining quantitative estimates of hydrogeological parameters using geophysical data are in great demand. A common approach to hydrogeological parameter estimation using geophysical and hydrogeological data is to first invert the geophysical data using a geophysical inversion procedure, and subsequently use the resulting estimates together with available hydrogeological information to estimate a hydrogeological parameter field. This approach does not allow us to constrain the geophysical inversion by hydrogeological data and prior information, and thus decreases our ability to make valid estimates of the hydrogeological parameter field. Furthermore, it is difficult to quantify the uncertainty in the corresponding estimates and to validate the assumptions made. We are developing alternative approaches that allow for the joint inversion of all available hydrological and geophysical data. In this presentation, we consider three studies and draw various conclusions, such as on the potential benefits of estimating the petrophysical relationships within the inversion framework and of constraining our geophysical estimates on geophysical, as well as hydrogeological data. In the first approach, we use information obtained from radar tomographic velocity zones to invert tracer test data. We invert the hydrogeological field using non-stationary, unknown empirical petrophysical relationships and only use the information in the tomogram that helps us to fit the tracer test data. Synthetic studies are used to assess the effects of non-random errors in the intrinsic petrophysical relationships, and how geophysical data acquisition errors (e.g., unknown borehole deviations, unknown zero- times) affect our estimates. We conclude that we can estimate the spatial variability of the hydrogeological field given a strong intrinsic petrophysical relationships (p>0.8) and very carefully collected geophysical data. The results also illustrate the limitations of a sequential deterministic inversion approach. In the second approach, crosshole ground-penetrating radar (GPR) travel times and hydrological measurements, collected before and during transient flow experiments, are used jointly to estimate flow parameter distributions in the vadose zone (and their uncertainty). This approach employs concepts from the Pilot Point method in a maximum a posteriori framework and requires the joint simulation of variably saturated flow and GPR travel times. In a synthetic 2D example, the permeability distribution is estimated, and additional parameters of the relative permeability and capillary pressure functions are |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://cloudfront.escholarship.org/dist/prd/content/qt9jn9v917/qt9jn9v917.pdf?t=li5vuy |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
