A New Paradigm for Groundwater Modeling
Shu-Guang Li and Qun Liu
Part of the Studies in Computational Intelligence book series (SCI, volume 79), 2008
Chapter from book Quantitative Information Fusion for Hydrological Sciences (pp.19-41)
A 1987 landmark National Science Foundation report on scientific computing and visualization (McCormick et al., 1987) envisioned the future of scientific computing to be real-time interactive with the modelers being dynamically-engaged and in full control throughout the computational process. The report stressed: scientists not only want to solve equations or analyze data that results from computing, they also want to interpret what is happening to the data during computing. Researchers want to steer calculations in real-time; they want to be able to change assumptions, conceptual framework, resolution, or representation, and immediately see the integrated effects, the ultimate implications, and the complex interrelationships presented intelligently in a meaningful context. They want to be an equal partner with the computer, interact on-line with their data, and drive in real-time the scientific discovery process. While this would certainly be the preferred modus operandi and is finally becoming computationally feasible even for many 3D dynamic problems on a personal computer, it is not the current standard of groundwater modeling. Although these thoughts were first reported nearly twenty years ago, they express an idea that is current and more relevant than ever before as the computing power continues to grow exponentially.