A DEUTERIUM-CALIBRATED GROUNDWATER-FLOW MODEL OF A REGIONAL CARBONATE ALLUVIAL SYSTEM

The White River Flow System (WRFS), a regional carbonate-alluvial groundwater system in southeastern Nevada, U.S.A., contains large amounts of water in storage, especially in the underlying carbonate reservoir. As the population of Nevada grows, it may become necessary to tap the resources of this and other regional carbonate systems. Because of the depth to the carbonate reservoir and, until now, lack of motivation to collect detailed hydrogeological data on it, the state of knowledge of flow in the carbonate system is poor. However, a simple mixing-cell flow model of the WRFS can be constructed and calibrated with the spatial distribution of the stable isotope deuterium. This type of model subdivides the system into carbonate and alluvial cells and routes water and deuterium through the entire cell network. It provides estimates of recharge rates, groundwater ages and volumes of water in storage. Transience in recharge rates and their deuterium signatures are unaccounted for by the model. The lack of constraints on the system mandates the calibration of three different flow scenarios, each of which differs slightly from the other. Despite these differences, some consistent quantitative results are obtained. Foremost among these are: (1) the carbonate aquifer may contain as much as 752 km3 of water in storage; (2) recharge from the Sheep Range to Coyote Spring Valley is at least 90% greater than previously believed; (3) Lower Meadow Valley is part of the WRFS and contributes underflow to Upper Moapa Valley; (4) underflow with an average value of 0.163 m3 s-1 flows westward out of the system along the Pahranagat Shear Zone; (5) recharge to the alluvial system is greater than that to the carbonate system; (6) groundwater mean ages range from 1600 to 34 000 years, with the oldest waters exceeding 100 000 years old. The results also demonstrate that deuterium can be used to calibrate simple flow models and provide groundwater ages. Despite the uncertainties and lack of constraints in mixing-cell models, they provide first approximations to information which, until now, has been difficult, if not impossible, to obtain. These models are especially useful for analyzing sparse-data systems, testing different flow hypotheses with minimal effort, providing ranges in parameter estimates, guiding future data collection and serving as precursors for the development of more sophisticated models. © 1990.