ABSTRACT
A 59 m-thick section of a freshwater oil shale interbedded with marlstone of Lower Carboniferous (Tournaisian) age from the Big Marsh area in Antigonish Basin, Nova Scotia, Canada, was examined using reflected light microscopy, Rock-Eval pyrolysis, X-ray diffractometry analysis, inductively coupled plasma-mass spectrometry for elemental analysis, and prompt γ for boron concentration. The oil shale was deposited in a lacustrine environment based on geology, sedimentology, variation in organic matter, and boron content (28-54 ppm). Organic petrology classified the oil shale into three broadly distinct types. Type A oil shale is a coastal facies shale containing terrestrially derived macerals, such as vitrinite and inertinite, sporinite, with some lamalginite, and amorphous bituminous matrix. Type B oil shale was deposited in a shallow-water facies and contains mostly lamalginite and some vitrinite and sporinite. Type C oil shale is a relatively deep-water facies, associated with open-water Torbanite-type oil shale and contains mostly Botryococcus colonial telalginite. The oil shale is thermally mature (T max is 441-443 °C). Total organic carbon (TOC) varies from 5.8 to 7.3 wt %, and the hydrogen index is between 507 and 557 mg HC/g TOC. The rate of sedimentation as determined by the Th/U ratio indicates possibility of three sedimentation periods: an irregular but mostly slow rate of sedimentation from the base of the section up to 68 m, followed by a regular and slow rate between 68 and 53 m, and a regular and fast rate between 53 m and the top of the section. The higher Th/U ratio during deposition of the shallow-water facies was due to the input of allochthonous U. The redox conditions, as reflected in the variation of Cr to Mo, U, and Ni + V, indicate that the oil shale was deposited under suboxic-dysoxic conditions. The high organic productivity by phytoplankton and bacteria is characterized by a low Cr and high V/Cr ratio and suboxic conditions. In contrast, the well-oxygenated and uniform, warm-temperature upper water level supports a dysoxic environment. Variation of Sr/Ca vs Mn/Ca ratios indicates that most samples have low values, a characteristic of colder water and high terrigenous influx. The post-Archean Australian shale (PAAS)-normalized rare earth elements (REEs) follow three trends. Type A oil shale has the highest concentration of total REEs (648 ppm) and light REEs (LREEs, 605 ppm) as compared with type C (269 and 233 ppm), which are less than half of type A. Type B oil shale has the lowest total REEs (184 ppm) and LREEs (152 ppm). The concentration of heavy REEs decreased from 43 ppm in type A oil shale to 36 ppm in type C oil shale. Comparison of PAAS-normalized REEs for the three oil shale types indicates a reduction of the negative Eu anomaly with depth, which is possibly related to sedimentary sorting as a result of accumulation of fine sediments in the deeper water zone of the lake. The concentration of most elements of environmental concern is similar to and/or lower than the world shale. However, there are instances of higher concentrations of hazardous elements (e.g., As, Cd, Mo, and Se).
ABSTRACT
In this paper, we present a customized method for estimating sonic shear velocity (Vs) from compressional velocity (Vp) logs in the Montney Formation, in wells lacking dipole sonic data. Following a multi-scenario analysis that comprised of assessing empirical Vs estimation relations [including lithology, porosity (Ø), and volume of clay (Vsh)-based Vs estimation techniques], bivariate statistics, and machine learning, we found that the Greenberg & Castagna (1992) shale lithology constants yield Vs log estimates that best match the measured Montney Formation Vs in our study area, with a regional correlation coefficient of 0.8. We have therefore customized the Vs estimation method in our study to use the Greenberg & Castagna (1992) shale lithology constants. Our working method: â¢Improves the efficacy of Vs log estimation from Vp logs in the study areaâ¢Demonstrates the importance of calibrating empirical relations for Vs estimation to a specific formation, andâ¢Provides a more accurate complementary Vs log dataset for subsequent regional reservoir characterization studies.
