KEYNOTE SPEAKERS
Borland Hydrology Award
Dr. Hoshin Vijai Gupta – Regents Professor of Hydrology and Atmospheric Sciences, University of Arizona
Keynote Lecture: March 21, 2023 1pm – CSU Lory Student Center, Grand Ballroom
Towards Physical-Conceptual Modeling of Mass, Energy and Information Flows Using Machine Learning Technology
Abstract: The success of any Machine Learning strategy depends on the conceptual and algorithmic Representation that is selected for Encoding and Processing Information. Further, the chosen encoding/representation completely determines the questions that can be asked, analyses that can be performed, and the answers that can be obtained. Ultimately, the effectiveness and efficiency of any ML strategy depends on Information Theoretic choices related to what Information we chose to encode (and store), the form in which we choose to encode that Information, and the method by which that encoded Information is processed. This raises interesting questions regarding (1) the kinds of Informational Encoding that are possible and useful when addressing a particular problem, (2) when and how Physics-based Encoding can synergistically interact with Data-based Information Processing strategies to achieve outcomes that are both Interpretable and Non-Lossy (ie., that achieve maximal possible performance), and (3) the value of multi-representational approaches to support/enable scientific discovery. My view is that by rooting the development of Machine Learning/Artificial Intelligence and Physics-Based Modeling in the fundamental perspectives and language of Information Theory, we can hope to achieve the most rapid progress in the Domain Sciences. While my thoughts may perhaps be speculative, I dont think I am alone in thinking this way, as evidenced by ML literature related to Information Bottleneck theory, and also to the fundamentals of Computational Science.
Biography
Hoshin Vijai Gupta is Regents Professor of Hydrology and Atmospheric Sciences at The University of Arizona. He received his BS in Civil Engineering from IIT Bombay, and MS and PhD degrees in Systems Engineering from Case Western Reserve University. His broad interest is in how “Learning” happens through the development and use of “Models”, and more specifically in how to combine Physics-Based Knowledge with Machine Learning (via Information Theory) for developing Earth & Environmental Systems Models that can progressively learn from interactions with the environment.
In 2017 and 2018, Hoshin was ranked in the top 1% on the Clarivate “Highly Cited Researchers List” for Environment/Ecology. He is a Fellow of the American Geophysical Union and the American Meteorological Society, recipient of AMS’s RE Horton Lecture Award (2017) and EGU’s Dalton Medal (2014), and has served as an Editor of Water Resources Research (2009-2013).
Hoshin teaches an introductory class on “The Bare Minimum” one needs to know about the physics-based approach to Environmental Systems modeling, and an advanced-elective class on “How We Learn from Data” that integrates relevant concepts from Statistics, Information-theory, Machine-Learning, Deep-Learning, and Physics-Based model development.
AGU Hydrology Days Award
Dr. Mary C. Hill – Professor of Geology, University of Kansas
Keynote Lecture: March 21, 2023 4:30pm – CSU Lory Student Center, Grand Ballroom
A Strategy that Includes National Models to Evaluate Water Availability for Arid Agricultural Areas Being Impacted by Climate Change: the Case of FEWtures in the Central Arkansas River basin (CARB)
Abstract: The NSF FEWtures project seeks to evaluate potential ways to lower carbon production and enhance local economies and communities in arid agricultural regions using local renewable energy supplies. To serve this purpose, the study focuses on the Central Arkansas River basin (CARB) in the central USA, and explores the economics and stakeholder adoption potential of two local renewable energy powered enterprises: (a) locally produced green ammonia for use as fertilizer and energy storage, and (b) treatment of water that has been historically unusable due to salinity or other water quality issues.
The system depends on water resources that have been dramatically depleted by irrigation, a characteristic it unfortunately shares with many other systems worldwide. This talk presents a strategy to quantify water availability under climate change developed by the FEWtures Water Supply and Treatment Teams, composed of Patience Bosompemaa, Sam Zipper, Andrea Brookfield, and Edward Peltier. Deep groundwater resources that are not highly interactive with surface water are evaluated using a water-balance method based on historical annual precipitation and head changes. Surface water resources and closely connected groundwater are evaluated using national models. Two national models are available: the USGS National Hydrologic Model (NHM) and the National Water Model (NWM). Adding to water supplies through treatment of saline groundwater, saline groundwater produced with oil and gas extraction, and water from feedlots is also considered. The strategy and approaches for addressing difficulties are discussed in this talk.
Biography
Mary C. Hill is a professor of Geology at the University of Kansas, Fellow of the American Geophysical Union, and member of the National Academy of Engineering. Professor Hill graduated from Hope College in Holland Michigan in 1976 with AB degrees in Business Administration and Geology, and spent a year at Michigan State University in Civil Engineering. She received her PhD in Civil Engineering – Water Resources from Princeton University in 1987 and was a Hydrologist with the USGS from 1981-2014 in New Jersey and then Colorado, achieving the level of GS-16, the highest level a scientist can reach in the US government. At the USGS she used the very popular model MODFLOW to focus on computer modeling of groundwater and groundwater-surface-water hydrologic systems, integration of models and data, and quantification of prediction uncertainty. Since 2014 at KU, Prof. Hill has focused on the integration of science and policy, with an emphasis on creating an environmentally and economically sustainable future for people on earth. To this end, she has worked to add associated human systems such as agriculture and renewable energy development to models and decision support systems. She has worked with very talented students to link data sets from these fields to create visualization software and applications (DiscoverFramework, DiscoverWater and DiscoverHABs) and agriculture-energy-water decision support (FEWCalc). She focuses largely on the food, energy, water (FEW) nexus. Prof. Hill is PI of the $2.5M NSF FEWtures project that combines staff from KU and KSU to focus on the technical, economic, and adoption feasibility of two opportunities of using local renewable energy to support local rural economies and communities and reduce carbon emissions. The two opportunities are local treatment and use of contaminated water, including produced water from oil and gas development, and local production of green ammonia. Preliminary results suggest water treatment is likely to be economically feasible in limited circumstances, while local production of green ammonia appears to hold promise in establishing a prosperous, sustainable future for rural areas.
Borland Hydraulics Award
Dr. Fred L. Ogden – Chief Scientist (ST), National Oceanic and Atmospheric Administration, National Water Center Office of Water Prediction
Keynote Lecture: March 22, 2023 1pm – CSU Lory Student Center, Grand Ballroom
Transformative Infiltration Modeling using The Soil Moisture Velocity Equation
Abstract: Hydrologic predictability suffers from a lack of a comprehensive theory of stormflow generation. This is particularly true in situations where soil hydraulics dominate partitioning of rainfall into runoff and soil moisture. For this reason most hydrologic models use simplified methods to calculate infiltration. To model infiltration processes, rigorous hydrologic models often numerically solve some form of the Richardson/Richards Equation (RRE); the partial differential equation that describes the variation of water content over time at a point in unsaturated porous media in response to rainfall, plant water uptake, or groundwater table dynamics. Solving the RRE is one of the most challenging problems in hydrologic prediction because of the highly nonlinear dependence of hydraulic conductivity and capillarity on water content, plus numerical solver convergence challenges associated with steep gradients in water content or discontinuous media properties such as soil layering and tillage/compaction. One common method used to reduce computational effort and improve RRE solution robustness involves the use of a coarse spatial discretization of the soil. However, coarse discritizations can violate the Representative Elementary Volume (REV) assumption and can smooth heterogeneities, leading to inaccurate solutions. This presentation reviews a Lagrangian reinterpretation of the RRE called the Soil Moisture Velocity Equation (SMVE) which was first published in 2017 by Fred’s research team. The one-dimensional vertical solution of the SMVE employs a finite water content discretization of the soil, which is advantageous because it avoids the need to discretize the soil in space. The SMVE advection-like term can be solved as an ordinary differential equation using standard numerical methods. The SMVE solution simulates infiltration in layered soils as well as the effects of groundwater on infiltration. It does this with guaranteed mass conservation, and without the numerical reliability issues that can impede RRE solutions. Compared to an appropriately applied RRE solution, the SMVE solution exhibits less than five percent difference in calculated infiltration over multi-month simulations. This presentation derives the SMVE from the RRE, and gives examples showing the advantages of this transformative new infiltration modeling tool.
Biography
Raised in a farm and ranch family near Lamar, Colorado, Fred earned B.S. (‘87), M.S. (‘89), and Ph.D. (‘92) degrees in Civil Engineering from Colorado State University. After a two year post-doctoral research position at the Univ. of Iowa, Iowa Institute for Hydraulic Research (IIHR), in 1994 he accepted a tenure-track Civil Engineering faculty position at the Univ. of Connecticut in Storrs. At UConn he received the US Army Research Office (ARO) Young Investigator Award to study hydrologic applications of weather radar observations. In 2006 he moved his family to Laramie, where he held an endowed chair position at the Univ. of Wyoming until 2017. During his time in academia, Fred developed several hydrologic and hydraulic simulation models with students and collaborators that remain in use by the US Army Corps of Engineers and other federal agencies. Fred also has extensive field and laboratory experience in physical hydrology and hydraulics. From 2003-2017 he collaborated with multi-disciplinary teams studying hydrologic ecosystem services in the Panama Canal Watershed through ARO and NSF funding. Working with students he instrumented 14 catchments with different land covers, and made hydrologic, meteorologic, geochemical, isotopic, and geophysical measurements to evaluate numerical models for flood and drought predictions. Dr. Ogden spent 2015-16 on sabbatical leave with NOAA-NWS at the National Water Center in Tuscaloosa, Alabama helping to plan upgrades to the National flood & drought warning systems. He returned to Tuscaloosa to continue that work with NOAA-NWS in 2017, as Chief Scientist. He currently serves as co-chair of the Hydrology and Watershed Systems subcommittee of the US Global Change Research Program, and as a member of the NOAA Council of Fellows. Fred has received numerous awards from ASCE including the Best Paper award from the J. Irrigation & Drainage Engineering with co-author and CSU Professor Emeritus Professor James Ruff of CSU, the Collingwood Prize with co-author the late Tatsuaki Nakato of IIHR for a paper describing sediment control at riverside intake structures using submerged vanes, and the Arid Lands Hydraulic Engineering Award.