Quantitative Information Fusion for Hydrological Sciences

Volume 79 of the series Studies in Computational Intelligence pp 69-103

Trajectory-Based Methods for Modeling and Characterization

  • D. W. VascoAffiliated withLawrence Berkeley National Laboratory, University of California

* Final gross prices may vary according to local VAT.

Get Access

Given the increasing data flow associated with characterizing and modeling natural hydrological systems there is a need for efficient data analysis tools. New experimental techniques such as multi-level samplers and crosswell pressure tomography produce extensive sets of hydrologic observations. Furthermore, integrating geophysical and remote sensing data can produce a massive array of measurements. Relating these data to flow properties in the subsurface and using them for characterization can be a computational challenge. In this chapter I discuss trajectory-based methods for integrating transient head, tracer, multi-phase flow, and associated geophysical data. The techniques, which are similar to ray methods in geophysics, are motivated by two methodologies outlined in this chapter. The first methodology, implemented in the frequency-domain, is along the lines of methods associated with elastic and electromagnetic wave propagation [26, 28, 30]. The second approach, the method of multiple scales is somewhat more general, applicable to nonlinear and diffusive propagation, and has been applied to gas dynamics and shock propagation [2]. The techniques described have connections with other methods used in the modeling of fluid flow. For example asymptotic approaches have been used to describe solute transport in the limit of long times, giving for example traveling wave solutions [19, 22, 46, 47].