The goal of this project is to advance water sustainability research capabilities by providing the Sustainable Water through Integrated Modeling (SWIM) human-technology framework that fosters participatory modeling of future water scenarios through dynamic cyberinfrastructure (SWIM 2.0). SWIM 2.0 automates the ingestion of data into models (i.e., data-to-model integration) and the integration of decoupled models (model-to-model integration), supporting the generation and interpretation of model scenarios and results.

The SWIM project addressed this goal through convergent research from three synergistic subprojects: SWIM-SEM, SWIM-PM, and SWIM-IT. SWIM-PM focused on enabling participatory analysis of the socio-economic-environmental water system through developing and delivering workshops for stakeholders for future water scenarios. Research on data- and model-based reasoning and developing sustainability science competencies was performed. "Canned scenarios" were created to answer specific questions on water's future as a prelude to guiding stakeholder users through developing and running their own scenarios. SWIM-IT focused on developing SWIM 2.0, which includes a Web-based interface, software tools, and services to support water research and the use of water-related models. SWIM 2.0 is available at https://purl.org/swim in English and Spanish and was developed considering stakeholders' input. SWIM 2.0 provides infrastructure for integrating water models created in different modeling software (e.g., SciLab) or languages (e.g., Python), automatically composing scientific workflows for model-to-model integration, and connecting to national infrastructure (i.e., tools and services offered by the NSF-funded Consortium of Universities for the Advancement of Hydrologic Science - CUAHSI) to foster the reuse of SWIM's data and models. Software products developed in this project are available at public repositories to foster reusability. The Web-based interface of SWIM 2.0 enables the use of scenarios to answer questions about water's future, provides recommendations of model outputs based on the user's roles (e.g., farmer, water manager), data visualizations, and scientific narrative elements (i.e., context-enhanced textual representation of model outputs). SWIM-SEM focused on using formally-described semantics to create knowledge graphs that enhance the execution and understanding of data and models generated by SWIM 2.0. A recommender system was integrated into SWIM 2.0 to suggest model-output variables. SWIM 2.0 vocabularies and knowledge graphs were created to facilitate the annotation of data, provenance capture, and enable automated workflow composition. Scientific narrative elements were generated to provide context on model outputs and facilitate the interpretation of model results. 

The results of this project were disseminated to communities of interest, including researchers and stakeholders, in regional, national, and international venues. The research team presented at events such as the Permanent Forum of Binational Waters - Science talk series, the CUAHSI Cyberseminar series, and the American Geophysical Union (AGU) meetings. Peer-reviewed papers were published at relevant conferences (e.g., IEEE International Conference on eScience) and journals (e.g., Journal of Water Resources Planning and Management).

The research activities in the SWIM project provided opportunities for training a diverse group of students from The University of Texas at El Paso, a Hispanic-serving institution, to acquire the knowledge and develop skills required to collaborate in interdisciplinary research and be capable of working effectively with data to address societal-relevant challenges. This project contributed to our capacity for engaging in socio-environmental-technical activities by improving our understanding of integrating knowledge, theories, methods, data, languages, and diverse perspectives for complex challenges such as water sustainability.

Last Modified: 04/30/2023
Modified by: Natalia Villanueva Rosales