The Challenge to Integrate Sedimentology and Well Completion to Predict Production Performance in an Unconventional Reservoir

Abstract

Abstract This study presents the integration of sedimentology and well completion to explain variable production performances in wells from an unconventional siliciclastic reservoir. The significance of our study is that multidisciplinary approaches are among the most comprehensive solutions that can forecast areas with different performance and, thus, make consistent completion strategies before drilling and optimize the development of unconventional reservoir acreages. The main producing interval within the area of interest includes cm-thick, carbonate-rich layers that based on a total of 20 vertical wells are inferred to extend several km. The cumulative thickness of these layers and their frequency (number of layers/thickness of producing interval) vary across the area. This variation is compared to hydraulic fracturing and completion data that include breakdown pressure, pressure decline, hammer effect, ISIP gradient, fracture driven interactions, and completion strategies (i.e., proppant intensity, volume of fluids, cluster, stages, well spacing). All these data are then compared to the reported production from horizontal wells (> 150). Both the correlation of vertical wells and the distribution maps predict the existence of higher cumulative thickness and frequency of carbonate-rich layers towards the center of the study area. Also, the frequency of carbonate-rich layers reveals a positive relationship with the normalized production of nearby horizontal wells, which points to increased production as brittleness of the system increases. However, a variable degree of data dispersion is seen in this positive relationship, and it is attributed to different completion strategies over the years. One way, though, to explore the relationship between completion parameters, sedimentology, and well production is to use a correlation matrix that allows to differentiate the relative importance of each factor. For instance, we observe a strong and negative correlation between the frequency of carbonate-rich layers and the ISIP gradients, whereas the ISIP gradients and the well production exhibit a small, albeit positive, correlation. Completion and hydraulic fracture designs undoubtedly influence well production but there exists an inherent anisotropy, related to the presence of cm-thick, fragile, carbonate-rich layers between m-thick, plastic, clay-rich intervals, that affects the toughness of the system and, ultimately, impacts well production. This work presents solution seeking through integration of multidisciplinary approaches with application of data analytics to create predictive methodologies. The use of two different disciplines such as sedimentology and completion design, rarely integrated together for an unconventional play, is novelty by itself. Correlation matrices could play a fundamental role in the optimization of field development plans in general and bespoke hydraulic fracture design.