From Bureau of Economic Geology, The University of Texas at Austin (www.beg.utexas.edu).
For more information, please contact the author.

2004 GSA Annual Meeting, Denver, Colorado, November 7–10, 2004

Descriptive Upscaling of Hydraulic Conductivity by Lithologic Mapping for Modeling Groundwater Flow in Heterogeneous Terrigenous-Clastic Aquifers

Alan R. Dutton* and Jean-Philippe Nicot

Abstract:

Descriptive upscaling of heterogeneous hydraulic conductivity for input to numerical models of groundwater flow can be conditioned with lithologic maps. Examples from the Carrizo-Wilcox aquifer in central Texas and the High Plains aquifer in the southern and central High Plains show laterally continuous, narrow, meandering, and bifurcating channels of high hydraulic conductivity. Accurate resolution of similarly dimensioned channels is limited in data-poor areas.

We hand-contoured hydraulic-conductivity data guided by lithologic maps and following defined rules, then digitized results for model input. Data quantity and quality can vary for both lithologic maps and hydraulic-conductivity data. Density of hydraulic-conductivity data is 0.02 measurements/km2 across >29,000 km2 for the Carrizo-Wilcox aquifer and 0.07 measurements/km2 across >110,000 km2 for the High Plains aquifer. Lithologic maps drawn from abundant geophysical logs have greater resolution than hydrologic data for the Carrizo-Wilcox aquifer and were used to upscale hydraulic conductivity. Mapping of hydraulic conductivity of the Ogallala Formation drew more on abundant specific-capacity data than on lithologic maps drawn using poor-quality sample logs from water-well drillers. Better model-calibration results in these aquifers were obtained using the descriptively upscaled hydraulic-conductivity distribution than using either a kriged or a uniformly distributed version of hydraulic-conductivity distributions.

One limitation of the approach is how much judgment and geoartistic license on the part of the hydrogeologist are needed in contouring data and mapping hydraulic conductivity. There can be multiple ways to honor hydrologic data that reflect the underlying geological fabric. Alternate depositional systems models might yield a different map fabric or different connectedness of hydraulic conductivity. Another limitation is that not all aquifer systems have the same availability of lithologic maps, well logs, and other data. Having sufficient hydrologic data, including specific-capacity data, to resolve fine-scale permeability features is crucial to upscaling on the basis of depositional systems.

*Earth and Environmental Sciences, Univ of Texas San Antonio, 6900 N. Loop 1604 W, San Antonio, TX 78249-0663, adutton@utsa.edu