Distributed Hydrologic Models: Can we reduce complexity and simulate multiple processes in a watershed?
Abstract
Coupled physically-based distributed hydrologic, sediment, and water quality models attempt to represent a number of linked physical processes occurring at various space and time scales. Although the basic equations of mass, momentum, and energy may be solved, the solutions still require parameters related to the actual physical system. The number of parameters depends on the number of processes and their mathematical description. Surface measurements are often not possible at the scale of a fundamental computational unit, which cover an areal region 100's to 1000's of square meters in size. Subsurface measurements are possible at the fine scales required for vertical discretization, but are limited by the effort and cost of exploration and the use of the information is hampered by subsurface heterogeneity. Some researchers propose simplified modeling approaches with reduced complexity and fewer parameters. Unfortunately, this path limits the amount of information that can be verified regarding flow path and process- level validation. We believe that model parameter estimation is a necessary requirement, and that simpler model formulations are not the only way forward. Distributed models offer the only viable approach for validating fluxes and storages. The use of data types other than those used in the typical calibration cycle is imperative to verify the flow paths and residence times are being correctly simulated. These data can include distributed soil moisture, groundwater level, solute concentration, light stable isotopes, and high-resolution lidar data for micro-topographical measurements. We will provide several examples where the use of other data sources add confidence to model performance and reduce parameter uncertainty.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2006
- Bibcode:
- 2006AGUFM.H41F0459D
- Keywords:
-
- 1805 Computational hydrology;
- 1836 Hydrological cycles and budgets (1218;
- 1655);
- 1838 Infiltration;
- 1849 Numerical approximations and analysis;
- 1873 Uncertainty assessment (3275)