| The research will test the hypothesis that integrated models incorporating land surface, soil moisture, groundwater and stream flow processes forecast ungauged basin response. Every catchment/watershed/basin is under-gauged or completely ungauged for general applications in the fields of hydrology, ecology, and geochemistry. Gauges, as related to surface water, are essential for water resources assessment as they allow the estimation of water budgets, support estimation of erosion rates, and aid in calibrating/validating hydrological models for water resources management in general. This research constructs an integrated, multi-process model for estimating long-term stream flow in an ephemeral, ungauged semi-arid basin, the Rio Salado in central New Mexico. The basin drains 3,575 square kilometers and includes seasonal channel flow from summer thunderstorms. The basin is presently without a stream gauge, although stream flow records near the outlet to the Rio Grande are available from late 1947 to mid 1984. The focus of this research is to develop a strategy for supporting the physical model with an appropriate space-time data for geometry, physical parameters, and forcing for the model using available a priori data. The development of the hydrologic conceptual model is critical for constructing the model and verifying its plausibility of the a priori simulation. The conceptual model defines a list of rules that the simulation must follow, such as geometry, architecture, channel flow regimes, and water budgets. As part of this research, a geodatabase was developed to support the integrated model using an appropriate data model. It includes terrain, soil and hydrogeologic properties, historical time series of leaf area index, precipitation, potential evapotranspiration, etc. The study has implemented MODHMS (developed by HydroGeologic Inc.), a physically-based and spatially-distributed model with variably saturated subsurface flow, overland flow, and channel flow. A priori data is found to sufficiently populate the conceptual model for this modeling study, even though significant knowledge gaps and data uncertainties are known to exist. Additional qualitative/quantitative data were obtained by the aid of remotely sensed images. ASTER day and night temperature and reflectance products were used to identify ephemeral and perennial stream reaches. The concept is based on apparent thermal inertia (ATI=1-albedo/temp day- temp night). Remote sensing improved the hydrogeologic conceptual model and it further suggests that groundwater from ephemeral reaches in the lower Rio Salado drains to the south beneath the watershed divide, and discharges to the Rio Grande at San Acacia. The last phase of the study used MODHMS to simulate 50-year scenarios of climate and land cover change on the water resources of the Rio Salado, as well as recovery from drought 2000--2007 also compared to ASTER. The results show climatic differences from the headwater to the outlet, as well and hydrologic response from soil parameter changes, as likely results from land cover change. |
