1,705
edits
Changes
Jump to navigation
Jump to search
Statistical analysis in oil correlation (view source)
Revision as of 16:34, 29 January 2014
, 16:34, 29 January 2014Initial import
{{publication
| image = exploring-for-oil-and-gas-traps.png
| width = 120px
| series = Treatise in Petroleum Geology
| title = Exploring for Oil and Gas Traps
| part = Critical elements of the petroleum system
| chapter = Oil–oil and oil–source rock correlations
| frompg = 8-1
| topg = 8-71
| author = Douglas W. Waples, Joseph A. Curiale
| link = http://archives.datapages.com/data/specpubs/beaumont/ch08/ch08.htm
| pdf =
| store = http://store.aapg.org/detail.aspx?id=545
| isbn = 0-89181-602-X
}}
==Statistical methods==
In studies involving large numbers of samples, many different analyses, and a huge variety of ratios and quantities to be evaluated, it is easy to become bogged down in an overabundance of data. To speed up the process of evaluating correlation data and to make the process less subjective, statistical methods have been used (e.g., .<ref name=ch08r51>Sofer, Z., Zumberge, J., E., Lay, V., 1986, Stable carbon isotopes and biomarkers as tools in understanding genetic relationship, [[maturation]], biodegradation and [[migration]] of crude oils in the northern Peruvian Oriente (Maranon) basin: Organic Geochemistry, v. 10, p. 377–389., 10., 1016/0146-6380(86)90037-9</ref><ref name=ch08r54>Telnæs, N., Dahl, B., 1986, Oil-oil correlation using multivariate techniques: Organic Geochemistry, vol. 10, p. 425–432., 10., 1016/0146-6380(86)90042-2</ref><ref name=ch08r8>Curiale, J., A., 1987, Distribution of transition metals in North Alaskan oils, in Filby, R., H., Branthaver, J., F., eds., Metal complexes in fossil fuels: American Chemical Society Symposium Series 344, p. 135–145.</ref> These techniques include both cluster analysis and multivariate statistical analysis (principal components analysis).
==Application of cluster analysis==
The figure below shows the application of cluster analysis in using data on trace-element concentrations to correlate eleven crude oils from northern Alaska. Although the concentrations of elements in an oil may vary due to alteration episodes (such as biodegradation), the relative distribution of these elements often remains constant and source distinctive. The concentrations of copper, iron, manganese, nickel, and vanadium in each oil were used to construct this cluster. The Bray-Curtis distance shows the degree of dissimilarity from oil to oil. Here eight oils of type A and three oils of type B are indicated. Other geochemical evidence supports this subdivision.
[[file:oiloil-and-oilsource-rock-correlations_fig8-1.png|thumb|{{figure number|8-1}}From Curiale<ref name=ch08r8 />); reprinted with permission of the American Chemical Society.]]
==Integrate with geology and geochemistry==
Unfortunately, many of the applications of such statistical methods fail to emphasize either the chemistry or the geology of the system. Future applications should attempt to integrate the purely statistical analysis with geochemical and geological knowledge to achieve the most useful and realistic interpretation possible.
==See also==
* [[Introduction to correlations]]
* [[Basic principles]]
* [[Limitations of correlations]]
* [[Practical suggestions]]
==References==
{{reflist}}
==External links==
{{search}}
* [http://archives.datapages.com/data/specpubs/beaumont/ch08/ch08.htm Original content in Datapages]
* [http://store.aapg.org/detail.aspx?id=545 Find the book in the AAPG Store]
[[Category:Critical elements of the petroleum system]]
[[Category:Oil–oil and oil–source rock correlations]]
| image = exploring-for-oil-and-gas-traps.png
| width = 120px
| series = Treatise in Petroleum Geology
| title = Exploring for Oil and Gas Traps
| part = Critical elements of the petroleum system
| chapter = Oil–oil and oil–source rock correlations
| frompg = 8-1
| topg = 8-71
| author = Douglas W. Waples, Joseph A. Curiale
| link = http://archives.datapages.com/data/specpubs/beaumont/ch08/ch08.htm
| pdf =
| store = http://store.aapg.org/detail.aspx?id=545
| isbn = 0-89181-602-X
}}
==Statistical methods==
In studies involving large numbers of samples, many different analyses, and a huge variety of ratios and quantities to be evaluated, it is easy to become bogged down in an overabundance of data. To speed up the process of evaluating correlation data and to make the process less subjective, statistical methods have been used (e.g., .<ref name=ch08r51>Sofer, Z., Zumberge, J., E., Lay, V., 1986, Stable carbon isotopes and biomarkers as tools in understanding genetic relationship, [[maturation]], biodegradation and [[migration]] of crude oils in the northern Peruvian Oriente (Maranon) basin: Organic Geochemistry, v. 10, p. 377–389., 10., 1016/0146-6380(86)90037-9</ref><ref name=ch08r54>Telnæs, N., Dahl, B., 1986, Oil-oil correlation using multivariate techniques: Organic Geochemistry, vol. 10, p. 425–432., 10., 1016/0146-6380(86)90042-2</ref><ref name=ch08r8>Curiale, J., A., 1987, Distribution of transition metals in North Alaskan oils, in Filby, R., H., Branthaver, J., F., eds., Metal complexes in fossil fuels: American Chemical Society Symposium Series 344, p. 135–145.</ref> These techniques include both cluster analysis and multivariate statistical analysis (principal components analysis).
==Application of cluster analysis==
The figure below shows the application of cluster analysis in using data on trace-element concentrations to correlate eleven crude oils from northern Alaska. Although the concentrations of elements in an oil may vary due to alteration episodes (such as biodegradation), the relative distribution of these elements often remains constant and source distinctive. The concentrations of copper, iron, manganese, nickel, and vanadium in each oil were used to construct this cluster. The Bray-Curtis distance shows the degree of dissimilarity from oil to oil. Here eight oils of type A and three oils of type B are indicated. Other geochemical evidence supports this subdivision.
[[file:oiloil-and-oilsource-rock-correlations_fig8-1.png|thumb|{{figure number|8-1}}From Curiale<ref name=ch08r8 />); reprinted with permission of the American Chemical Society.]]
==Integrate with geology and geochemistry==
Unfortunately, many of the applications of such statistical methods fail to emphasize either the chemistry or the geology of the system. Future applications should attempt to integrate the purely statistical analysis with geochemical and geological knowledge to achieve the most useful and realistic interpretation possible.
==See also==
* [[Introduction to correlations]]
* [[Basic principles]]
* [[Limitations of correlations]]
* [[Practical suggestions]]
==References==
{{reflist}}
==External links==
{{search}}
* [http://archives.datapages.com/data/specpubs/beaumont/ch08/ch08.htm Original content in Datapages]
* [http://store.aapg.org/detail.aspx?id=545 Find the book in the AAPG Store]
[[Category:Critical elements of the petroleum system]]
[[Category:Oil–oil and oil–source rock correlations]]