ABSTRACT
Everything should be made as simple as possible but not simpler.
A. Einstein
One of the most signicant challenges that hydrogeologic modelers continue to face is the “tyranny of scales” [1] in hydrologic systems, that is, the disparity of temporal and spatial scales at which mass, momentum, and energy transport is best understood (e.g., subpore to pore scale and seconds to days) and at which predictions are needed for practical applications (e.g., plume to aquifer scale and years to centuries) (see Figure 12.1). is is typical in remediation strategies of contaminated sites; management of water resources; petroleum, gas, and geothermal energy production; and geological CO2 sequestration, which require long time predictions over large spatial scales. Achieving a predictive understanding of hydrological systems response to anthropogenic stressors and environmental changes, and the associated risks [2], is a primary societal need since long-term strategies to accommodate ever-increasing energy demands, to control atmospheric CO2, and to understand nutrient cycles need rst to be evaluated and then implemented. Modeling approaches that incorporate process understanding at dierent temporal and spatial scales, here referred to as multiscale
models, are therefore necessary to improve our predictive capabilities of natural systems.
