STATISTICAL ANALYSIS TO DETERMINING MODALITY OF STREAM ORIENTATION DATA

Past research has statistically analyzed orientation data in many disciplines to determine randomness in data sets and evaluate similarity between different data sets. Different types of test can be conducted to determine if a Von-Mise or Gaussian distribution is the best fit for orientation data, yet few methods exists to properly analyze modality in data sets.

A GIS stream layer is used to obtain lineaments over the Arbuckle-Simpson aquifer in southern Oklahoma. Over 60,000 stream segments are obtained using GIS to study streams over the aquifer. Rose diagrams are created for 2 different discretizations over the study area. Field work is also conducted in the study area to correlate the GIS stream layer to fractures found in the field. A GIS layer of faults over the study area is also used to determine if a relationship exists. A new method is introduced to study the modality of length weighted orientation data. Results indicate that this modality analysis method is a helpful tool when characterizing lineaments. Unimodal, bimodal, and polymodal signals are found over the study area, which can help describe the hydraulic properties of the underlying fractured carbonate aquifer.