Stage-level hydraulic fracture width and height quantification through stochastic inversion of distributed fiber-optic strain sensing data

The measurement of quasi-static rock deformation, specifically strain, utilizing distributed fiber-optic sensing is an innovative application. Low-frequency distributed acoustic sensing (LF-DAS) has been successfully applied in monitoring hydraulic fracturing treatments. In this study, we develop a stochastic inversion algorithm to characterize hydraulic fracture dimensions and quantify the associated uncertainties using cross-well distributed fiber-optic strain measurements. Strain modeling is accomplished by employing a 3D displacement discontinuity method based geomechanical model, and an efficient Markov chain Monte Carlo (MCMC)-based optimization method is utilized to generate the posterior distribution of the model parameters. The accuracy of the inversion algorithm is validated using a synthetic case. Then a field case study is conducted to demonstrate the practical value of the algorithm in obtaining insights on the geometry and propagation characteristics of hydraulic fractures.