Defining the width and average fracture frequency of fracture intensification domains using a partial least squares analysis, linear piecewise regression, and the Akaike information criterion: implications for the evolution of fracture-dominated fault damage zones
MetadataShow full item record
This dissertation quantifies fracture frequency variations using new methodologies that incorporate the partial least squares analytical method for multivariate data, linear piecewise regressions of cumulative fracture frequency, and the Akaike information criterion. The predictive equations and novel fracture frequency analyses presented in the two manuscripts provide valuable constraints when developing models of fluid flow pathways by accurately quantifying the distribution of fractures within a given geologic setting. The statistical analyses presented in this dissertation focus on predicting geometric variations of fracture intensification domains (FIDs). FIDs are identified in the field as volumes of elevated fracture frequency as compared to a relatively lower background frequency level. The data set utilized in these analyses consists of 3678 fractures and 17 minor normal faults at 34 field sites in the Ordovician Utica Group and Schenectady Formation in the Mohawk Valley of eastern New York State.