Haynesville Shale: Predicting Long-Term Production and Residual Analysis To Identify Well Interference and Fracture Hits

Abstract

Summary In this study we analyze the production data from 2,755 horizontal wells in the Haynesville Shale. Correlations were generated to predict 4-, 5-, 6-, and 7-year cumulative production from initial, 6-, 12-, and 24-month production data. These correlations can help in field-development planning and economic analysis. The residuals (Predicted – Actual Cumulative Production) from these correlations were also analyzed, and this technique can be used to identify wells affected by interference, refracturing, fracture hits, and other factors. The cumulative-production estimates from the developed correlations were also compared with the corresponding estimates from the decline-curve-analysis (DCA) equations (seven different DCA methods were used). The accuracy of prediction made on the basis of correlations developed in this study is close to that for various standard DCA methods published and used in the industry. The developed correlations are faster to use and easier to implement for a large number of wells. Another of our objectives was to develop P10, P50, and P90 type curves for the Haynesville Shale using the available production data. A subset of the total wells (i.e., 150 wells evenly distributed throughout the study area) was used for predicting type curves. These type curves were generated and compared using different DCA models. The different DCA methods predicted an uncertainty of 5 to 27% for the P10, P50, and P90 production profiles.