Machine Learning Guided Optimization (MLGO) of Water Management Options with a Physical Groundwater Model for the Mid-County Basin Optimization Study

The Santa Cruz Mid-County Basin Groundwater Sustainability Agency conducted a Regional Water Optimization Study (Study) to support the selection of water supply projects and management actions within the critically overdrafted Santa Cruz Mid-County Groundwater Basin (Basin), for long-term operations and shared regional benefits, including sustainable groundwater management and regional water supply needs. To support this objective, the Study utilized an integrated groundwater surface-water flow (GSFLOW) model of the Basin and surrounding areas to simulate water supply projects and management actions into the future (water year 2023 to 2075), covering the sustainability planning horizon under California’s Sustainable Groundwater Management Act (SGMA). Water supply projects and management actions considered under this study include aquifer storage and recovery (ASR), indirect potable reuse (IPR) , and interagency transfers. Many different implementations are possible for each type of project s, such that the number of potential project configurations is practically endless. To facilitate optimization, we developed a novel workflow utilizing machine learning algorithms to conduct optimization by autonomously designing, preprocessing, post-processing, and evaluating physical model scenarios. We term this workflow Machine Learning Guided Optimization (MLGO). MLGO successfully produced novel simulations and produced better performing project configurations over time.