In those places where there are no data, or where data are absent, the value '-999' is assigned to the grid cell to denote null values in the modeled surface. In the case of the Digital Elevation Model, which was imported from other sources, the edges of the input grid are all null data (-999) to constrain the geometry of the grid to reflect the NTS map boundary geometry. This geometry was propagated throughout all of the grids surfaces by the assignment of '-999' to ensure their edges conform to the DEM geometry.
Stratigraphic pick data used to generate the surfaces were derived from a number of sources, including new picks made during the course of the study. Because of varied degrees of certainty regarding the quality of those data, all data were ranked in terms of quality, with the author's picks ranked highest. This ranking was applied to screen data only in areas where more than one stratigraphic pick was made at a given location. In those cases, the highest ranked value was incorporated into the calculations. It's important to note that all data were used in process, and that in the absence of any other data, lower quality data were used along with data of higher quality.
Digital data sets in the model have a varied degree of precision depending on the nature of the surface modeled. For example, the surface DEM carries the greatest amount of precision as is it modeled from a 76m-grid-resolution source file, and it is exposed and can be easily validated. Precision of the other surfaces varies in accordance to data density and knowledge regarding the processes of formation or deposition. Therefore the modeled surfaces reflect not only data density and data quality but also knowledge regarding the processes, which is depicted through the addition of synthetic data such as digital hand drawn contours. Consequently, grids of surfaces with the least amount of knowledge or data were validated and constrained by surfaces that carry a higher degree of confidence. Surfaces of geological units were tested against the two major unconformities in the study area - the topography of the bedrock surface, which itself, was tested against the topography of the present-day land surface. This internal validation and testing was done more rigorously for the bedrock topography and lower-drift stratigraphic units. This process produces grids of surfaces that have geospatial integrity (that is, lower surfaces do not cross surfaces of higher units), and enables the construction of bedrock subcrop map that is entirely constrained by the geometry of the bedrock surface. Testing of geospatial integrity of the bedrock geology surfaces that lie beneath the bedrock surface was not as rigorous, and in some cases surfaces of lower bedrock units may cross higher units. These occurrences, while present, are rare.
A series of digital grids were generated for surfaces of geological units ranging from the top of the McMurray Formation to the present-day lands surface. The grids were generated from hard data consisting of stratigraphic picks made from geophysical logs and lithologs, as well as from 'synthetic' data consisting of digital hand-drawn contours lines, particularly for the bedrock topography and the drift aquifer units resting on the bedrock surface. Initial digital modeling was done using MCadContour software on a Apple Macintosh computer, which were later exporting to PC supported software platforms such as Viewlog (ESR 2002-03). All modeled surfaces were done on a 250m grid-cell size, using Inverse Distance Weighted interpolation method.
The bedrock geology units were modeled from stratigraphic picks derived from Alberta Energy and Utilities Board databases (AGSWDB). The bedrock topography and drift aquifer units were modeled from stratigraphic picks made on resistivity and gamma logs derived mainly from oil and gas exploration. Hand-digitized contours of the bedrock surface were used as synthetic data to enhance the resolution of the modeled surface. All bedrock geology surfaces were tested against the bedrock topography (an unconformable surface) to create a bedrock subcrop map. Gridded surfaces are then stacked in proper stratigraphic order to generate a 2.5D model of the geological setting.