T39 Three-Dimensional Geological Mapping for Groundwater Applications

Sunday, 22 October 2006
1:30 PM - 5:30 PM

PROTOTYPE THREE-DIMENSIONAL GEOLOGIC MODEL OF THE MARLBOROUGH QUADRANGLE, MASSACHUSETTS: 3-D MODELING ON A SHOESTRING

Paper number: T39-4
Program: Technical Sessions
Time: 2:20 PM – 2:35 PM

Authors

Abstract

Borehole data was integrated with surficial and bedrock geologic mapping to produce a true 3-D model of the geology within the Marlborough quadrangle, Massachusetts. The 3-D model was then used to parameterize a steady-state groundwater flow model of a sub-region of the quadrangle. Using a surficial materials map of the quadrangle, layers representing each of six major stratigraphic units were isolated in Adobe Illustrator, saved as six tiff images, and then converted in RockWorks2004 to six grid files, one for each stratigraphic unit. Nodes within each grid file were assigned thicknesses using 190 boreholes logs as a guide, effectively creating an isopach for each unit. Subtracting the uppermost isopach from a 25-meter DEM created a base grid for that unit. Repeatedly subtracting each isopach from the overlying grid produced a series of base grids that RockWorks2004 could model as a block diagram with the area and volume of each unit. After assigning hydraulic conductivities to each unit, the grid files were imported into MODFLOW-2000 to model steady-state groundwater flow within a sub-region of the quadrangle. Boundary conditions for the model were assigned by extending the model to meet true hydrologic boundaries. Prior to any model calibration, simulated and observed hydraulic heads were compared for select boreholes and resulted in a good agreement between the two. The first-order agreement between the hydraulic heads suggests that the 3-D geologic model is a useful tool for groundwater modelers looking to expedite the modeling and calibration process.

Overall, a reasonable 3-D geologic model can be assembled on a shoestring budget but with some limitations. First, a good surficial materials map must be available at the start or forget it. Our 3-D map is based more on a working knowledge of morphosequences than borehole-to-borehole correlation. Second, inexpensive modeling applications are limited. We had difficulty handling the complex geology and highly variable topography, e.g., addressing the intertonguing of lake bottom fines with coarser deposits. Finally, “on-the-fly”, real time editing of model cross sections is a must. Users must have the ability to edit the geology with either borehole data, when available, or professional judgment, particularly in areas of complex geology with discontinuous units.