Conclusions from the Examination of Surface Geochemical Data from around the World
HARTMAN, BLAYNE, Transglobal Exploration & Geoscience, Leucadia, CA
A powerful approach for interpreting onshore and offshore surface geochemical data is to construct cross-plots of key hydrocarbon parameters and to examine the plots in terms of a mixing model. Hydrocarbon additions, such as seepage, appear on the plots as mixing trends away from natural background populations. The characteristics of the hydrocarbon source for each trend can be deduced by the direction of each trend. This approach enables identification of hydrocarbon inputs without statistical manipulation and enables differentiation of hydrocarbons from several different sources in the same area. Applying this approach to data collected around the world (at least eight of which will be shown) demonstrates the following: (1) cross-plots of surface geochemical data from petroliferous reas generally contain several well defined mixing trends indicating multiple hydrocarbon inputs; (2) areas devoid of petroleum always show low values and a lack of trends, unless anthropogenic inputs exist; (3) natural background hydrocarbons generally exist at low quantities, but often have characteristic ratios similar to petroleum sourced hydrocarbons. Thus, interpretation methods highly dependent on ratios may lead to erroneous results; (4) natural background hydrocarbons in sediments and soils are more variable in magnitude and characteristics than in water. Thus, identification of low-level natural seepage in sediments and soils is more difficult than in waters; (5) the larger the data base (number of analysis locations), the better mixing trends can be defined and characterized. ater column programs (sniffers) generate much more data over a much larger territory, which increases the odds for detecting hydrocarbon seepage; and (6) adsorbed gas data, generated by subjecting soils and sediments to acid desorption, should also show variations in background and mixing trends. However, the limited amount of published adsorbed data does not seem to show the expected patterns. This situation poses a question worth investigating further.
AAPG Search and Discovery Article #91004 © 1991 AAPG Annual Convention Dallas, Texas, April 7-10, 1991 (2009)