URTeC: 1068 Geochemical Perspectives on Cuttings-Based Chemostratigraphy and Mineral Modeling in the Delaware Basin, Texas and New Mexico Harold D. Rowe* 1 , Pukar Mainali 1 , Michael Nieto 1 , John D. Grillo 1 , Harry B. Rowe, Jr. 2 ; 1. Premier Oilfield Group, Houston, TX 2. Data Analytics Consultant, Palm Harbor, FL. Copyright 2019, Unconventional Resources Technology Conference (URTeC) DOI 10.15530/urtec-2019-1068 This paper was prepared for presentation at the Unconventional Resources Technology Conference held in Denver, Colorado, USA, 22-24 July 2019. The URTeC Technical Program Committee accepted this presentation on the basis of information contained in an abstract submitted by the author(s). The contents of this paper have not been reviewed by URTeC and URTeC does not warrant the accuracy, reliability, or timeliness of any information herein. All information is the responsibility of, and, is subject to corrections by the author(s). Any person or entity that relies on any information obtained from this paper does so at their own risk. The information herein does not necessarily reflect any position of URTeC. Any reproduction, distribution, or storage of any part of this paper by anyone other than the author without the written consent of URTeC is prohibited. Abstract The interpretation of large geochemical data sets (10 3 to 10 5 sample points) and their derivatives are developed, checked for veracity and relevance, optimized, and integrated with associated data sets to address questions regarding lithological heterogeneity, depositional continuity, depositional conditions, diagenesis, and brittleness. A 110-well Delaware Basin XRF-based geochemical data set, largely consisting of 10-feet-resolution cuttings samples that span much of the Wolfcamp through Delaware Mountain Group, are used as an example. Data workflows employing a suite of unsupervised learning techniques (e.g., PCA, HCA) are evaluated to determine the strengths/limitations of each technique, and their collective/comparative utility. Elemental results are used as inputs to stoichiometry-constrained element-to-mineral (E-M) models that yield useful inferences. The strengths of an E-M model rest on the accuracy of the ED-XRF calibration, sample quality, analytical prowess of the ED-XRF analyst, the overall rigorousness of the XRD technique and analyst employed, and the specific approach of the E-M model. Further to this point, derivatives of the XRF-based modeled mineralogy, such as a mineral brittleness index (mBI) and derived chemofacies, are only as good as the analytical underpinnings of the inputs. Modern core- and cuttings-based stratigraphic studies frequently incorporate an inorganic geochemical component, often acquired with portable energy-dispersive x-ray fluorescence (ED-XRF). Despite the analytical limitations of the ED-XRF approach, the use of this technique yields large, quantitative data sets collected at the length-scale of inches (cores) to feet (cuttings). In their raw elemental form, these data sets provide additional correlation and lithological control at scales just above (core), to just below (cuttings) the scale of most downhole log suites. The significance of this approach is that it can be used to 1) refine rock signatures in well logs, and 2) resolve questions regarding stratigraphic succession and correlation. While the initial focus of the study is to reconstruct the spatial distribution of lithologies at the scale of cuttings sample collection (10 feet), the overarching goal of the project is to optimize the interpretation of log signatures through the addition of data generated from the rock. This approach has cross-disciplinary implications, including refinement of petrophysical, geomechanical, and regional geological models. The modeled mineralogy from the chemostratigraphy results has direct implications for modeling fluid-rock compatibility and the overall completions process, including a more strategic selection of stage lengths.