Abstract: Estimating Geochemical Anomaly Thresholds and Quantifying Anomaly Contrast for Near Surface Petroleum Exploration with Probability Plot Analysis

FONTANA, JOHN V.

Any geochemical survey for petroleum exploration will result in high and low values that, as a set of data, have unique frequency distributions. Observing the characteristic shape of a histogram is a simple way to examine the frequency distribution of a data set, and in clear-cut cases, can provide reliable anomaly thresholds. Background and anomalous data populations each have their own independent frequency distribution. Each population has a mean and standard deviation independent of the other population. These parameters can be modeled using probability plots, which is another way to display frequency distributions. This has been applied in mineral exploration over the past two decades and only recently to petroleum exploration methods.

After modeling the independent populations, the limits of the populations can be used to objectively determine anomaly thresholds; for example, the mean plus two standard deviations of the background population. This value represents all but the upper 2.5% of the background data, and identifies anomalous data above that value. More confidence is warranted to a geochemical anomaly threshold than other methods that use subjective and arbitrary methods to determine anomaly thresholds.

Other information gained from the probability plot models include contrast ratios of the background data mean and anomalous data mean. These values provide a method to compare anomaly-strength in different areas.

Actual examples of processed data area shown and caveats about the methods area also discussed.

AAPG Search and Discovery Article #90946©1997 AAPG Rocky Mountain Section Meeting, Denver, Colorado