Data-Driven Water Systems Solutions for Communities at World Water Week
The West Hub hosted the Data-Driven Water Systems Solutions for Communities session at World Water Week on Monday, August 21, from -10am PT/10-11am MT (18:00-19:00 in Sweden). The event explored the intersections of AI-centric data, analysis, and water research and management as it relates to community-driven actions for equitable water solutions for all. Co-hosts included Centro de Inteligencia Artificial; the eScience Institute, University of Washington; FAIR in ML, AI Readiness and Reproducibility (FARR) Research Coordination Network; GO FAIR US; the New Mexico Water Resources Research Institute; and the San Diego Supercomputer Center, UC San Diego; UC Berkeley; and the University of Bergen.
​
Whether you attended live or viewed the recording, we look forward to learning more about your work and determining ways to make meaningful progress together.
This session promoted an understanding of the challenges and opportunities associated with implementing multi-disciplinary data and AI-driven capabilities to further equitable water solutions. It explored data and AI-driven advancements that could enable people to take action against climate change-exacerbated water scarcity, which to date remains widely inaccessible. These capacity gaps contribute to the disconnect between people’s needs and the application and advancement of data and AI-driven water science. This session’s presentations and exchange were rooted in the understanding that revolutionizing this connection will help realize the societally impactful potential of science by catalyzing participatory approaches to data and AI-driven analysis, thus empowering communities around the world to take informed action in the face of climate change and realize a resilient water future. Seventy participants joined the live event. You can view their responses from the Mentimeter interactive introduction here.
Session Recording
WEST HUB PANEL
Bio:
Ashley Atkins is the Executive Director of the National Science Foundation-funded West Big Data Innovation Hub, which is led by UC Berkeley, UC San Diego, and the University of Washington. Previously, she was a Research Scientist at the New Mexico Water Resources Research Institute. Her research has focused on data-driven computational modeling of the interconnections between hydrologic and human systems. She is a Co-PI on the West Hub and on the Transboundary Groundwater Resilience Network, an NSF-funded AccelNet project that aims to connect and leverage the strengths of water, social, data, and systems science to set the foundation for a more resilient water future. Ashley is also an alumna of Fulbright Bulgaria and Fulbright Greece. She has an M.S. in Water Science and Management from New Mexico State University and a B.A. from Davidson College, where she was a William Holt Terry Scholar.
Promoting the Implementation of Multi-Disciplinary Data and AI-driven Technology at a Water-Stressed Irrigation District in Northern México
Centro de Inteligencia Artificial
Alfredo Granados-Olivas, Víctor H. Esquivel-Ceballos, Rafael Corral-Díaz, Eduardo Castillo-Luna; José Mireles Jr., Manuel Resendez, Joam M. Rincon
Bio:
Alfredo has been a professor at the Universidad Autónoma de Ciudad Juárez, at the Department of Civil and Environmental Engineering since 1987. He graduated from NMSU in 2000 (AgronomyHydropedology [Soil-Water interactions]). He has an MS in Groundwater Hydrology from UACJ in Mexico (1987) and a BS in Ag Engineering from ESAHE in Mexico (1984). Presently, he is a member of the National System of Researchers in Mexico from 2005 till present (2025) (SNI-Level 1). Dr. Granados was recently appointed as the vice president of the Mexican Association of Hydrology for the state of Chihuahua. In 2016 he was awarded by the National Association of Schools and Faculties of Engineering of Mexico (ANFEI) for distinguishing himself as one of the academics who are being an actor in training engineers with academic excellence. His professional interest is in the sustainability of water resources in arid ecosystems to promote equity and sustainable development in rural communities of Mexico. President and owner of the Center for Technology Transfer on AI located in Ascension, Chihuahua, Mexico, and leader for the Precision Ag Project at the AI Center of Chihuahua, Mexico.
Abstract:
In arid northwestern México, water is a limited resource for irrigation and pecan orchards are continuously growing while demanding vast amounts. Water consumption of pecans is high while using traditional irrigation systems (i.e., flood irrigation) and the state of Chihuahua is the biggest producer in México with above 100 thousand tons per year (2020).
However, irrigation efficiency is below 45% at most farms with minimum or non-existent adaptation of AI-driven technology, impacting aquifer systems and losing productivity. In this project, we are promoting local implementation of multi-disciplinary data while monitoring real-time ET, soil moisture, and spectral response to soil temperature using remote sensing, user-friendly app development, and regenerative agricultural practices.
Results show that local farmers are interested in learning and implementing these technologies to reduce water stress at their orchards while increasing their productivity. The local implementation of these data-driven technologies while educating and training farmers on how to use these tools could reduce the capacity gaps that contribute to the disconnection between people’s needs and both the application and advancement of data and AI-driven water science.
Data Governance for Water Governance: Reflecting on the Useability and Accessibility of NASA’s GRACE Data Product Suite
eScience Institute, University of Washington
Akshay Mehra, Sameer H. Shah, Yuanning Huang, Kimberly Kreiss, Maia Powell, Aanchal Setia, Dharma Dailey, Vaughn Iverson
Abstract:
Readily available and open-source remotely-sensed data present several accessibility challenges. These challenges are attributable, in part, to the specific needs and audiences such products directly cater to. Processing and using such data to answer interdisciplinary socio-environmental problems requires a deep understanding of what products are available as well as how their properties can contribute to the specific, problem-oriented research questions at-hand. Here, we present a workflow capable of downloading, interpreting, and visualizing NASA’s Gravity Recovery And Climate Experiment (GRACE) data to estimate groundwater change dynamics. We elaborate on the development of this workflow, which was built as part of the 2023 Data Science for Social Good program at eScience at the University of Washington. Ultimately, our presentation highlights the complexities and non-straightforwardness of the data product itself, ultimately providing a set of reflective lessons on useability for which scientists must be aware of in data product design.
A Worldwide Network of Transboundary Groundwater Research: Construction and Preliminary Analysis
San Diego Supercomputer Center, UC San Diego
Ilya Zaslavsky
Bio:
Ilya Zaslavsky is a Co-PI on the NSF-funded Transboundary Groundwater Resiliency Research (TGRR) Network of Networks project. He is Director of Spatial Information Systems Laboratory at the San Diego Supercomputer Center, University of California San Diego. His research focuses on distributed information management systems—in particular, on spatial and temporal data integration, geographic information systems, and spatial data analysis. Ilya received his Ph.D. from the University of Washington (1995) for research on statistical analysis and reasoning models for geographic data. Previously, he received a Ph.D. equivalent from the Russian Academy of Sciences, Institute of Geography, for his work on urban simulation modeling and metropolitan evolution (1990).
Abstract:
The talk will demonstrate the construction and analysis of the network of transboundary groundwater research based on co-authorship information, undertaken within the NSF-funded Transboundary Groundwater Resilience Project. The results are being used to better understand the global research community and strengthen the networking activities of the project. The software to create and analyze such domain networks is freely available online.
Harnessing Systems Thinking and Participatory Modeling for Water Systems Research & Solutions
University of Bergen
Jefferson K. Rajah
New Mexico Water Resources Research Institute
Christine Tang
Bio:
Jefferson K. Rajah is a PhD Research Fellow at the System Dynamics Group, Department of Geography, University of Bergen. He holds a M.Phil in System Dynamics from the University of Bergen and a B.A.(Hons.) in Global Studies and Political Science from the National University of Singapore. Jefferson has a background in the social sciences and is a computational system dynamics modeller by training. His research interests centre on human behaviour and social system dynamics within larger ecological, technological, and health systems, as well as participatory modelling and stakeholder engagement. He has facilitated several participatory modelling workshops, where stakeholders have collaboratively constructed qualitative causal maps on topics related to sustainable development goals, public health, and groundwater resilience. Christine Tang CT is currently working part-time as a Research Scientist for the New Mexico Water Resources Research Institute (NM WRRI) and is an Interdisciplinary PhD Student in System Dynamics at Worcester Polytechnic Institute (WPI). CT focuses on dynamic computer simulation modeling methods and uses other Operations Research methods as needed. Before joining NM WRRI, CT built healthcare models. Working on the Transboundary Groundwater Resilience (TGR) Network-of-Networks project has been an enlightening experience.
Abstract:
Water stored below earth’s surface, known as groundwater, serves as the main source of freshwater for over two billion people around the world. Climate change has exacerbated global depletion trends for these critical resources. Navigating challenges associated with groundwater depletion becomes increasingly complex when these resources, known as transboundary aquifers, are shared between multiple countries. Participatory modeling approaches maintain the unique potential to bring together researchers, stakeholders, and decision makers with the collective expertise necessary to understand the complex human, hydrologic, and climate interconnections that affect transboundary groundwater systems. This presentation explores the potential of systems thinking and participatory modeling to advance use-inspired research of groundwater systems, with findings from a pilot group model building workshop to capture the mental models regarding transboundary aquifers from a group of cross-disciplinary researchers within the Transboundary Groundwater Resilience Network of Networks. It will also briefly discuss the complementariness of participatory modeling with emerging capabilities of AI-centric data collection and analysis.
FARR: FAIR Principles and Reproducibility
San Diego Supercomputer Center, UC San Diego
Julie Christopher
Bio:
Julie Christopher is a Technical Project Manager at the San Diego Supercomputer Center at UC San Diego. Current projects include the FARR RCN, GO FAIR US, West Big Data Innovation Hub, and National Science Data Fabric. She has experience supporting various projects regarding FAIR principles, data, and community building.
Abstract:
In this talk, I will introduce the FARR: FAIR in ML, AI Readiness, and Reproducibility RCN, provide an overview of the FAIR Principles, and highlight the research inconsistencies with ML reproducibility. The goals of this project are to build communities to promote better practices for AI, improve efficiency and reproducibility, and stimulate and enhance new research.
Bio:
Karen Stocks is the Director of the Geological Data Center at Scripps Institution of Oceanography, where she specializes in the documentation, discovery, access, integration, and curation of oceanographic data. She currently serves as the Director of the CCHDO global database of hydrographic measurements, the Director of Information Services for the International Ocean Discovery Program’s Science Support Office, and the Scripps lead for Rolling Deck to Repository, a multi-institution collaboration managing data from the US Academic Research Fleet. She holds a PhD in Biological Oceanography from Rutgers University, and her expertise includes information systems for vessel-based sensors, scientific ocean drilling, biodiversity and biogeography, metagenomics, and ocean observing systems.
Abstract:
Artificial Intelligence and Machine Learning show great promise for addressing complex scientific and management challenges around understanding and managing water resources. However, progress can only be made if data are ready for AI. The FAIR - Findable, Accessible, Interoperable and Reusable - data principles offer a framework for evaluating and achieving AI-readiness. While there have been successes in this area, there also remain obstacles to overcome.
