A subset of Dakota wells having a common set of wire-line logs (Spontaneous Potential (SP), Gamma Ray (GR), and Deep Resistivity (ILD)) was generated for spatial analysis of engineering and geologic factors. These common wire-line logs were used to extrapolate special core data and other engineering, geological, and production data from data-rich areas (those with many wells) to areas with low data density. Many of these features were subsequently mapped as GIS layers.

In particular the very complicated relationships between the three wire- line logs and several core properties were computed using neural network analyses: Porosity, Water Saturation, and permeability relationships were computed. A strong relationship was found for porosity (Fig. 1), a good relationship for Water Saturation and a potentially useable relationship for permeability.

As well as providing useful map data at more than 898 wells in the Dakato play, the study also provides a more detailed look at vertical variations within each predicted well. One potential use of this data is to geostatistically interpolate pseudo core data to provide a core porosity volume for the play (Fig 2); thus more accurate estimates of initial and remaining reserves could be made.