Geographically-based Probabilistic Petroleum Potential Models: Complement or Competition for Deterministic Petroleum System Models?
Kirk G. Osadetz1, Zhuoheng Chen1, and Haiyu Gao2
1Geological Survey of Canada - Calgary, 3303 – 33rd St. N.W., Calgary Alberta T2L 2A7, Canada
2Schlumberger Information Solutions, Calgary Technology Centre, #600, 322 – 11 Avenue S.W., Calgary, Alberta T2R 0C5 Canada
One plans deliberately for exploratory success, making it is desirable to assess petroleum potential and evaluate risks at various scales. Often the most important decisions are made with the few pertinent data. Hence, assessment is important because decisions and action rely on sound estimates of undiscovered potential that characterize and rank exploration opportunities and their intrinsic and relative risks. Since the enunciation of the anticlinal and stratigraphic accumulation paradigms by Hunt and Leverson, respectively, there has been significant progress understanding the processes that contribute to petroleum source rock accumulation and petroleum generation, migration, entrapment and alteration. Implicit in the accumulation paradigms is a risk reduction strategy, based on discernable geological features favourable for accumulation and preservation.
Play and prospect characterization, petroleum systems analysis and accumulation-based probabilistic models are three general models types that provide key information which informs the exploration process. Each model type contributes specific value and insight to resource appraisal and risk assessment at a variety of scales. The first two model types are essentially deterministic. Both rely on specific observable material properties of the petroleum system and both confirm the accumulation paradigms that were, in general, successfully employed using structural and stratigraphic analysis. Recent developments suggest that additional geophysical characterizations can also contribute successfully to the application of the accumulation paradigms, essentially without the any prior petroleum system knowledge. Prospect-based models commonly provide accumulation size estimates that facilitate economic analysis.
Dow and Magoon, among others, defined petroleum system concepts that consider the petroleum generated, migrated and entrapped. Such models can be of unparalleled value because they force consideration and reconciliation of multiple data sources and types. When subject to appropriate sensitivity analysis they can indicate key factors controlling petroleum system function. Howerver, such methods typically requires either sufficient geological data or subjective inference to allow petroleum system characterization. Especially difficult is the specification of permeability in secondary migration pathways and the integrity and capacity of seals on traps. Computational sophistication may outstrip the quantity and quality of input data. The examination of “classic” petroleum system models indicates significant uncertainties in petroleum systems characterization and definition (i.e. see posters on thermal history and oil family definitions in Williston Basin). Petroleum system models may neither provide sufficient accumulation size data to permit economic analysis, nor a rigorous characterization of either model uncertainty or exploratory risk.
Play-based probabilistic assessments of accumulation characteristics or historical exploration data were developed specifically to predict undiscovered accumulation sizes for economic analysis of both opportunity. These models typically require a play-based accumulation size distribution, an estimate of the number of prospects, and both play and prospect level risks, although the risks are commonly informed by analogy or subjective inference. The most reliable of these models are those based on exploratory history, but the strength of such models is their ability to compare both conceptual and established plays. Major shortcomings include the tendency conservative estimates of potential and the inability to ‘map’ the assessment results back onto the geography of the play. Recent methodological developments have resulted in geographically-based probabilistic assessment models that link geological favourability to the resource potential. Historical analysis of such methods suggest that exploratory efficiency could be improved by ~20% if exploration decisions are informed by geographically-based assessments. Current geological favourability map construction methods are rudimentary and the potential to contribute to improved exploration efficiency is linked to a better understanding geological favourability. The integration of petroleum systems data and models into the geological favourability maps, while obvious, remains to be done.
The expansion of the petroleum supply and resource-base to include unconventional and continuous reservoirs and the new focus on natural gas presents an important challenge to all models. Key to these models is the development of geological proxies which are indicative of reservoir performance as a function of well cost. The subtle variations in accumulation characteristics that distinguish reserves from uneconomic resource remain to be defined. Classic prospect-based models can and have been effectively applied to unconventional resources, although these commonly identify regions of improved reservoir characteristics which approach or meet conventionally economic criteria for success within large rock bodies where resource is truly unconventional. Hybrid reservoirs where both free gas is in contact with either coalbeds or gas hydrates are illustrate this problem. Geographically-based probabilistic methods have features which show promise for being adapted to the evaluation of unconventional and continuous resources. While petroleum systems models may help to inform aspects of such resources their deterministic nature and complicated construction renders them less-well suited to the appraisal of both unconventional and continuous resources.
AAPG Search and Discovery Article #90091©2009 AAPG Hedberg Research Conference, May 3-7, 2009 - Napa, California, U.S.A.