USGS visual identity mark and link to main Web site

USGS Mendenhall Postdoctoral Research Fellowship Program

17-31. Data science and machine learning to inform energy assessments

The USGS conducts scientifically robust quantitative assessments of undiscovered, technically recoverable oil and gas resources in the United States and around the world.  These assessments are mandated by Congress and often receive significant attention on release (https://dailycaller.com/2017/04/14/govt-geologists-discover-the-uss-largest-natural-gas-deposit/).  They are relied upon by others (https://www.usgs.gov/news/forecasting-world-s-energy-resources) to understand the future availability of energy resources. USGS assessment geologists conduct these assessments using the best available data for a region; the quantity and quality of that data varies significantly across the nation and the world.

One of the key data sources that supports those assessments is IHS Markit’s Well Data and Production Data datasets (https://ihsmarkit.com/index.html).  The IHS Markit’s US Well Data and US Production Data databases contain close to 500 million records describing many facets of oil and gas exploration and development such as fields, reservoirs, well logs, formation tops etc. This massive data resource has the potential to provide much greater value for the USGS if approached with emerging data analytical approaches.  Research is needed into data science and machine learning techniques that can help us to identify untapped potential in these and related databases that could impact the geologic and quantitative aspects of USGS resource studies and assessments.

We seek applicants for this Mendenhall Research Opportunity who are interested in developing analytical approaches for intelligently sifting through large geologic datasets to identify key parameters and relationships that will provide an improved understanding of the processes and patterns that underpin USGS resource studies.  Possible research topics include:

  1. Improving estimated ultimate recovery (EUR) values.  EUR is an estimate of the total oil or gas that a given well will produce; it is one of the most critical values in a quantitative assessment of continuous oil and gas resources.  For new assessments in developing plays, it can be challenging to find wells with a sufficient duration of production history to develop well production curves using traditional approaches. Some specific research questions that could be explored include the relationship between n-year well production and 30-year EUR, the ability to predict a shorter-time span EUR (e.g., 5-year) based on actual production data of more mature wells, and whether we can train datasets and use them to predict EURs based on a limited amount of production history.
  2. Improving estimates of drainage areas. The total area that can be “drained” by each petroleum well is another key input to the quantitative assessment of oil and gas resources; these values are often determined simply by considering existing well density and spacing.  A better understanding of the geologic controls on well density and spacing, and hence drainage areas, will provide critical information to assessment geologists and significantly reduce uncertainty in the quantitative results.  Addressing this question is likely to require a combination of data from the IHS databases with additional spatially explicit data on the subsurface.
  3. Improving the definition of assessment unit boundaries.  Spatial boundaries of an energy resource are defined by assessment geologists based on elements of the petroleum system such as source rock distribution or reservoir quality. There is the potential to improve these definitions by making better use of the site-specific data within the IHS US production and well databases and related data.  For example, in addition to location and production data, well-specific data on well test results, hydraulic fracturing treatments, and other well treatment types are available for many wells. These spatially explicit and rich datgist; Rasets provide an ideal opportunity for exploratory spatial pattern recognition and prediction. Research is needed to understand how this data can help define the structural boundaries of subsurface energy resources.

Applicants are expected to have a strong background in data science and machine learning as well as some knowledge of petroleum geology.  Interested applicants are strongly encouraged to contact the advisors below early in the application process to discuss project ideas.

Additional Reading on Energy Resource Assessments

Overview: National Academies of Sciences, Engineering, and Medicine. 2018. Future Directions for the U.S. Geological Survey’s Energy Resources Program. Washington, DC: The National Academies Press. (see beginning of Chapter 3) https://doi.org/10.17226/25141.

Methods: Charpentier, R.R., and Cook, T.A., 2010, Improved USGS methodology for assessing continuous petroleum resources, version 2.0: U.S. Geological Survey Data Series 547, 22 p. and program. Revised November 2012.

Schmoker, J. W., & Klett, T. R., 2007. US Geological Survey assessment concepts for conventional petroleum accumulations: Chapter 24 in Petroleum systems and geologic assessment of oil and gas in the San Joaquin Basin Province, California (No. 1713-24). US Geological Survey.

Example: Whidden, K.J., Pitman, J.K., Pearson, O.N., Paxton, S.T., Kinney, S.A., Gianoutsos, N.J., Schenk, C.J., Leathers-Miller, H.M., Birdwell, J.E., Brownfield, M.E., Burke, L.A., Dubiel, R.F., French, K.L., Gaswirth, S.B., Haines, S.S., Le, P.A., Marra, K.R., Mercier, T.J., Tennyson, M.E., and Woodall, C.A., 2018, Assessment of undiscovered oil and gas resources in the Eagle Ford Group and associated Cenomanian–Turonian strata, U.S. Gulf Coast, Texas, 2018: U.S. Geological Survey Fact Sheet 2018–3033, 4 p., https://pubs.er.usgs.gov/publication/fs20183033.

Proposed Duty Station:  Denver, CO

Areas of PhD: Data science, machine learning, computer science, geology, geophysics, petroleum geology, or related fields (candidates holding a PhD in other disciplines that also have extensive knowledge and skills relevant to the Research Opportunity may be considered).

Qualifications:  Applications must meet one of the following qualifications:  Research Geologist, Research Geophysicist, Research Computer Scientist, Research Statistician, Operations Research Analyst. (This type of research is performed by those who have backgrounds for the occupations stated above. However, other titles may be applicable depending on the applicant's background, education, and research proposal. The final classification of the position will be made by the Human Resources specialist.)

Research Advisors:  Kate Whidden, (303) 236-7788, kwhidden@usgs.gov; Greg Gunther, (303) 236-5884, ggunther@usgs.gov; Seth Haines, (303) 236-5709, shaines@usgs.gov; Karen Jenni, (303) 236-5766, kjenni@usgs.gov; Leslie Hsu, (303) 202-4080, lhsu@usgs.gov.

Human Resources Office Contact: Elissa Gregory, ekgregory@usgs.gov, 303-236-9579


Go back to Summary of Opportunities

U.S. Department of the Interior, U.S. Geological Survey
URL: http://geology.usgs.gov/postdoc/opps/2019/17-31 Whidden.htm
Direct inquiries to Cara A. Campbell at ccampbell@usgs.gov
Maintained by: Mendenhall Research Fellowship Program Web Team
Last modified: 11:07:40 Fri 08 Feb 2019
Privacy statement | General disclaimer | Accessibility