As concerns about water availability increase due to a changing climate and extreme weather, there is a need for computational test beds to help explore water management challenges and elucidate interdependences within coupled human-natural systems. These test beds can range from process-based simulations that take advantage of domain-informed approaches for hydrologic cycling to more data-driven models that leverage the underlying statistical properties of the data. However, development of data-driven models can be inhibited by lack of access to model source code or data. For example, the Water Rights Analysis Package (WRAP) is used by the state of Texas to assess basin-level water availability. WRAP is closed source software, and the available data is insufficient for exploring diverse hydrologic conditions. To address these issues, we have used the Colorado River Basin in Texas to: 1) simulate 1,000 historical realizations using Hidden Markov Models; 2) route these realizations through WRAP to obtain water user-level outputs; and 3) emulate WRAP's water allocation processes by developing an open-source capability using a long short-term memory (LSTM) neural network model. Although the current LSTM is trained using limited inputs, results indicate that the model can forecast water shortages with high accuracy (<10% error in shortage ratios) across space and time. The LSTM model is currently being used to assess the basin’s sectoral water resilience under increasing drought conditions. Our findings underscore the importance of integrating socio-hydrological considerations and systems thinking for the adoption of these methods across multiple sectors. This approach is pivotal for advancing watershed modeling techniques as well as supporting robust and resilient water management strategies in the face of climate variability.