BIG DATA ANALYSIS FOR CALCULATING SUSTAINABLE YIELD AND DEFINING THE FEASIBLE NEGOTIATION SPACE FOR AN AQUIFER SYSTEM

This study presents a big data analysis of multiple recharge interpretations and pumping regimes in the context of stakeholder preferences and desired future conditions for the Barton Springs segment of the Edwards aquifer. A set of candidate solutions was generated using a simulation-optimization model and aquifer responses were tested using multiple scenarios for recharge and scientific uncertainty interpretations. Analyses and simulation runs were completed using High Performance Computing (HPC) extracted resources at the Texas Advanced Computing Center. Approximately, 40,000 simulation runs were completed to generate a set of aquifer responses to pumping and other variables of interest to stakeholders, such as springfow rates and water table levels.

Resulting analyses provide an example of how supercomputing can be used to accelerate evaluation of scientific uncertainty and geological knowledge in relation to policy and management decisions. Moreover, this study presents scientific knowledge on a graphic overlay alongside stakeholder preferences defining a feasible negotiation space to inform aquifer management and help delineate actionable available yields. As computational methods improve, scientists are able to expand the level and scale of experimental simulation and testing that is completed for case studies. Understanding the aquifer behavior helps policy makers avoid negative impact on the endangered species, environmental services and aids in maximizing the aquifer yield.