4 Hydrogeology and Environmental Geoscience

Wednesday, 3 April 2002
1:00 PM - 5:00 PM

CHARACTERIZATION OF HETEROGENEITY AT MULTIPLE SCALES USING INDICATOR GEOSTATISTICS: WARD AND MCHENRY COUNTIES, NORTH DAKOTA

Paper number: 4-2
Program: Technical Sessions
Time: 1:20 PM – 1:40 PM

Authors

Abstract

The Sundre and New Rockford aquifers are two Pleistocene buried-valley aquifers occupying the same bedrock valley in north-central North Dakota. The indicator geostatistics approach was used to characterize and model proportion, geometry, and juxtapositioning of mud and diamicton facies (m), and sand and gravel facies (s). Facies m were coded as 1 (low permeability), and facies s as 0 (high permeability) at 0.5m depth intervals. The bias due to clustering of wells was reduced by declustering in the calculation of the volume proportion of the facies. Three zones (lower, middle, and upper) were found within the buried valley, each having a different proportion of facies m (Pm). The lower zone, which is the aquifer zone, is generally 35 m thick and Pm is approximately 21%. Facies m within this zone have a mean thickness of approximately 10.3 m, and facies s have a mean thickness of approximately 23.3 m. The middle zone is generally 50 m thick and Pm is approximately 56%. Facies m within this zone have a mean thickness of approximately 5.0 m, while facies s have a mean thickness of approximately 4.9 m. The upper zone is generally 50 m thick and Pm is approximately 92%. Facies m have a mean thickness of approximately 10.8 m, and facies s have a mean thickness of approximately 1.7 m.

Transition probability models were created from these results. The models were chosen to be exponential within the lower and middle zones, and spherical for the upper zone based on the coefficient of variation in facies length. Lateral correlation is currently being evaluated. Once all of the characteristics are determined, they can be utilized to simulate the 3-D distribution of facies m and s among the three zones. Such simulations can then be utilized to model ground-water flow and transport.