�鶹��Ƶ

Civil Engineering PhD Student Madeline Pernat received a 2024 National Science Foundation Graduate Research Fellowship (GFRP), a prestigious award that recognizes and supports outstanding students in a wide variety of science-related disciplines, for her research on large-scale water management in the Colorado River Basin. Pernat is advised by Associate Professor Joseph Kasprzyk and Research Professor Edie Zagona, the director of the Center for Advanced Decision Support for Water and Environmental Systems (CADSWES).

The GRFP recognizes outstanding graduate students from across the country in science, technology, engineering and mathematics (STEM) fields, paving the way for their continued work exploring some of the most complex and pressing issues of our time. This year the NSF awarded 27 �鶹��Ƶ students, including 18 from the College of Engineering and Applied Science, with the graduate research fellowship.

Fellows receive a three-year annual stipend of $37,000 and full coverage of tuition, fees and insurance, along with opportunities for international research and professional development that span five years. 

What does it mean to you to have received the NSF GRFP?
Firstly, I am honored to receive the NSF Graduate Research Fellowship. I am passionate about the research that I do, and I like to think that my enthusiasm for my work resonated with the reviewers. Secondly, I am grateful to everyone who has helped me reach this point in my academic journey, from my family, partner and undergraduate professors at Montana State University, to my current professors and advisors here at CU Boulder. I really see this award as a testament to the support of all these individuals – without them, I don’t believe I would be where I am today.

Tell me about your research.
My research focuses on large-scale water management in the Colorado River Basin. The Colorado River is a vital source of water for the American Southwest, fueling its productive cities and agricultural regions. The utilization of the Colorado River’s waters is made possible by engineered infrastructure designed to store and deliver the river’s waters effectively – particularly through Lake Powell and Lake Mead, the largest reservoirs in the basin. These reservoirs function as “savings accounts,” storing water during times of surplus and allowing for the release and use of that water during drier periods.  However, since the onset of the Southwest's "megadrought" in 2000, these reservoirs have consistently experienced higher rates of withdrawals than replenishment, leading to significant drops in reservoir levels. As of now, Lake Powell and Lake Mead are only about one-third full.

The current policies governing the operations of Lakes Powell and Mead are set to expire at the end of 2026, necessitating the development of new "post-2026" policies. This process is being undertaken by the Bureau of Reclamation, amidst challenging conditions of ongoing drought and low reservoir levels. My research, currently funded by the Bureau of Reclamation, takes place against the backdrop of these negotiations. The development of these new policies involves significant challenges, to name a few: a) many people and stakeholders depend on the Colorado River, often with competing interests that are difficult to balance, and b) the future, especially in terms of hydrology, is uncertain – we don’t know how much water will be available in the post-2026 era.

My research addresses these challenges through a twofold approach. The first part utilizes multiobjective optimization to generate new potential operating policies. This approach is used due to the presence of many conflicting objectives within the basin, such as ensuring water deliveries, producing hydropower, and meeting environmental flow requirements. Given these conflicting objectives, there is no single "optimal" policy that can satisfy all objectives simultaneously. Instead, multiobjective optimization is used to search for a set of policies, where each policy strikes a unique balance among the objectives. The generated policies cover a wide range in terms of performance characteristics, enabling various stakeholders to identify policies that cater to their specific interests while also understanding their tradeoffs and potential shortcomings. 

Secondly, I am developing a novel visual framework to enhance the decision-making process. This framework will enable stakeholders and decision-makers to visualize the array of potential future scenarios we could experience post-2026, while also illustrating how various policies would perform across these potential scenarios. Utilizing a visual “map” (as shown in the bottom figure), users can navigate through these potential future scenarios and assess the performance of different policies at a simple glance.

Why is this research important to you?
My research in water management is important because it bridges many important domains: engineering, data science, science communication, politics, economics and social justice. Although my academic background is primarily in engineering, I believe that in order to maximize my impact as a researcher, I must have a comprehensive understanding of these other diverse areas. I view the Colorado River Basin, with its pressing water management issues, as an exceptional case study for understanding this interconnectedness.

My interest in this research began in my undergraduate years after reading “Where the Water Goes” by David Owen, which discusses water management in the Colorado River Basin. Reading this book just seemed to connect the dots for me. It showed me a field where having a strong technical background in engineering is important for understanding the system and its challenges, while also emphasizing the need to understand and appreciate the system's broader societal and environmental dimensions.

Tell me a little more about your work with the Kasprzyk Research Group and CADSWES.
The Kasprzyk Research Group primarily focuses on multiobjective decision problems and the optimization and analysis of environmental systems and CADSWES specializes in water resources systems modeling. Over the years, Professors Kasprzyk and Zagona have not only developed a strong connection between their research interests but have also established collaborative ties with the Bureau of Reclamation. Their collaborative efforts and the work of the many students they have co-advised, have significantly advanced water resources management, especially in the Colorado River Basin. I feel incredibly fortunate to be a part of this team and to learn from some of the best in the field.

Image caption:  Madeline Pernat developed a novel visual framework to enhance the decision-making process. Utilizing a visual “map," users can navigate through potential future scenarios and assess the performance of different policies at a glance. The figure on the left displays a map of potential future scenarios, where each hexagon, referred to as a neuron, represents a possible scenario in terms of hydrology and water demand. The figure on the right presents the same map but illustrates how a specific policy performs across each neuron/scenario.