Problem scale has a significant effect on the way that hydrologic phenomena are observed and explained. Recent hydrologic research has clearly demonstrated that laboratory results may not adequately reflect field scale behavior, that models developed at one scale may not be valid at other scales, and that small scale variability may have large scale consequences. This research project is designed to help hydrologists obtain a clearer understanding of scale effects and their relevance to practical hydrologic investigations. It examines the complex relationship between natural variability and the models and data collection programs that from the basis for scientific hydrology. The first phase of the project reviews stochastic methods for describing natural variability over a range of time and space scales. The second phase examines new techniques for deriving scale-dependent estimates of hydrologic quantities. The final phase of research considers how field data can be used to improve the accuracy of these estimates. The stochastic approach adopted in the project provides a way to quantify important scale-dependent tradeoffs between model resolution, sampling effort, and estimation accuracy. This approach is illustrated with applications to groundwater flow and transport.
