April 2008, Volume 47, No. 4 1 PAPER: 2006-045 Geological Controls on the Origin of Heavy Oil and Oil Sands and Their Impacts on In Situ Recovery H.P. HUANG China University of Geosciences and University of Calgary B. BENNETT, T. OLDENBURG, J. ADAMS, S.R. LARTER University of Calgary Peer reviewed PaPer (“review and Publication Process” can be found on our web site) Introduction B iodegradation significantly alters the composition of petro- leum by the preferential removal of hydrocarbons (especially light end fractions), thus, adversely impacting oil physical properties and, consequently reducing oil producibility. Viscosity and den- sity are key properties for the evaluation, simulation and devel- opment of petroleum reservoirs. To develop and manage heavy oil fields cost effectively, it is essential to understand the varia- tion in petroleum fluid properties; especially viscosity throughout each reservoir within a field. A variety of studies demonstrated how oil properties in biodegraded oil accumulations can be pre- dicted from core and cutting extracts prior to well testing using geochemical parameters (1-5) . McCaffrey et al. (2) analyzed sidewall cores to identify geochemical parameters that are sensitive to sec- to identify geochemical parameters that are sensitive to sec- identify geochemical parameters that are sensitive to sec- y geochemical parameters that are sensitive to sec- geochemical parameters that are sensitive to sec- ondary charge. They then developed transforms that related those geochemical parameters to oil quality. Smalley et al. (3) used a sim- ilar approach to predict oil viscosity in a biodegraded heavy oil ac- cumulation. Guthrie et al. (4) developed a predictive model of oil quality based on a sample set of produced oils from Venezuela for Abstract Biodegradation of crude oil in subsurface petroleum reservoirs is an important alteration process affecting most of the world’s oil deposits. The process preferentially removes light compo- nents from conventional oil to form heavy oil and oil sand, which are more difficult to produce and are more costly to refine. Al- though reservoir temperature is a key control on biodegradation, large variations in oil properties have been documented in accu- mulations from similar depths within a play area. Data from the Liaohe Basin, NE China and other basins in China and elsewhere, indicate that biodegradation is most active in a narrow zone at or near the base of the oil column in contact with the water leg. The availability of nutrients from mineral dis- solution within the water leg is also thought to have a significant impact upon the degree of biodegradation. Thus, the level of bio- degradation increases with water leg thickness. Charge history and in-reservoir mixing of continuously charged oil with residual biodegraded oil also have a significant impact on oil physical properties. The conceptual biodegradation model proposed combines geo- chemical and geological factors to provide a coherent approach to estimate the impact of degradation on petroleum and to help reliably predict biodegradation risk at the prospect level. Our geochemical approach can be used to locate sweet spots (areas of less degraded oil), optimize the placement of new wells and completion intervals and help with production allocation from long production wells. predicting viscosity, API gravity and sulphur content in oil-stained sidewall cores where these properties cannot be measured directly. Koopmans et al. (5) analyzed oils from a single oilfield in the Li- oils from a single oilfield in the Li- oils from a single oilfield in the Li- aohe Basin, NE China. They found the large variations in viscosity across the field can be explained by mixing, to various extents, of heavily biodegraded oils with less degraded oils. They established a simple binary mixing model, which may assist in predicting the viscosity of reservoir oils before production. Although all these studies empirically track bulk oil proper- ties within a reservoir using geochemical parameters from oils or densely sampled core material, it has been well documented that empirical and theoretical relationships of viscosity to oil compo- sition for whole oils cannot be universa1ly applied (6) . Something about the process of biodegradation is universal and controlled by a few key factors, which can be used to predict petroleum compo- sitions and, thus, fluid properties (7) . Clearly, to predict viscosity variations within a reservoir requires an understanding of the fac- tors that influence compositional variations due to biodegradation under the dynamic conditions associated with subsurface systems. In this paper, we describe how reservoir geochemical studies of core samples from production wells, coupled with geological cri- teria, can be used to make predictions of downhole viscosities. The model proposed here can make such predictions more pre- cise and direct, since the level of biodegradation and the presence of a mixed charge are not random events. This conceptual model involves all of the main processes involved in the biodegradation of an oil column and opens the possibility of model-driven pre- diction of oil properties and production sweet spots across biode- graded reservoirs. Sample Background and Analytical Method Sample Background The oil sand sample suite is from the Liaohe Basin, NE China. A detailed description of the petroleum system and the effects of biodegradation on the petroleum accumulation could be found in Koopmans et al. (5) and Huang et al. (8-10) . Briefly, the Liaohe Basin is one of the Tertiary rift basins located in northeast China, which is mainly influenced by the motions of the Pacific Plate. It is one of the most important Cenozoic lacustrine sedimentary basins in China. The Shahejie Formation (Es) is widely distributed and con- tains the most important source rock and reservoir units in the entire basin. Based on lithology and fossil assemblages, it was sub- divided into four members (Es4, Es3, Es2 and Es1, from oldest to youngest). The reservoirs in the Liaohe Basin show strong het- erogeneities with relatively high porosity and permeability due to their poorly sorted and shallow burial depths. The present