Key Tools for Black Shale Evaluation: Geostatistics and Inorganic Geochemistry Applied to Vaca Muerta Formation, Neuquen Basin, Argentina

Nawratil, Alejandro¹; Gomez, Hugo²; Larriestra, Claudio³
¹CAPSA, Vicente Lopez, Argentina.
²CAPSA, Vicente Lopez, Argentina.
³Larriestra Geotecnologias, Ciudad Autonoma Buenos Aires, Argentina.

The assessment of hydrocarbons in Black Shales is a new challenge for oil and gas exploration. The detection of hydrocarbon generation potential in Black shales and the existence of potential reservoir zones within them, are key features that conventional well logs fail to adequately define. We present a method that allows the integration of chemical data (rapid hand held X-Ray Fluorescence analysis) of cutting and well logs using Sequential Gaussian Simulation.

This method allows creation of chemical element logs of V, Mo, S, Cr and Ni (high correlation with TOC) which allow assessing the potential hydrocarbon generation of source rocks. Moreover geochemical logs as Zr and Rb (negative correlation with TOC) allow to identify areas with better reservoir conditions in black shales, because they are related to clastic progradations occurred during Vaca Muerta sedimentation.

The methodology is based on selection of best correlations between the chemical elements and the set of wireline logs available in the well. Then Gaussian cosimulation is performed with Markov model type II, generating n realizations. The average value (E-value) is the expected log of the chemical element. The problem with this method is hard data (XRF analysis data) has depth location uncertainty, because the samples are drawn at approximate intervals (best case each 2 meters) and the sampling depth is not exact. This is the main source of uncertainty: the depth of which samples come from. Other additional uncertainty sources are sample contamination by wellbore collapse, drilling mud chemical composition, etc.. One solution is to assign each sample a random depth value within the range represented by sample and the samples subset is correlated with previously selected well log. This operation will select subsets of samples showing better correlation with selected well log. Later we proceed to the execution of 100 realizations for each subset. After running all realizations for all subsets, we calculate the mean, variance and the probability for any given cutoff. This methodology was applied successfully in the evaluation of source rock thickness and it helped to understand the probable reservoirs of Vaca Muerta Fm.

AAPG Search and Discovery Article #90155©2012 AAPG International Conference & Exhibition, Singapore, 16-19 September 2012