Groundwater Modeling in a Changing World: MODFLOW-and-More 2022 Special Issue

Abstract

The “MODFLOW-and-More” (MF&M) conference series arose from the vision of Chunmiao Zheng, Eileen Poeter, and Mary C. Hill. MF&M convened at the Colorado School of Mines on nine occasions commencing in 1998. Past attendees recall friendly (sometimes heated!) debates, lunch with luminaries—like Gregory J. Hobbs, Colorado State Supreme Court Justice, and James Eklund, lead negotiator of the Colorado River Drought Contingency Plan—dinners at Table Mountain Inn, and late-night pool at the Ace.

Following the challenges posed by COVID, the 10th rendition of MF&M convened in 2022 at Princeton University. The conference was a great success, sustaining the vibrance of nearly three decades by bringing private, public, and academic sector scientists together to share applications and bleeding-edge research and development of (mostly free and open-source) simulation codes and support software. While COVID challenges and the change in venue might have been anticipated to affect attendance, the 2022 crowd was right in line with previous meetings and attendees expressed genuine joy at rubbing shoulders again.

The “Changing World” theme of the 2022 conference could not have been more topical. Groundwater is now the subject of a series of articles in the New York Times. Such press coverage of groundwater has historically been occasional at best. What might our planet and its 8 billion population as of November 2022 be like without convenient access to local clean groundwater? We might discover the answer to this sooner and in more places in the world than we ever imagined.

This Special Issue emphasizes advances in computational methods and codes for evaluating groundwater data, availability, quality, at a range of scales; plus novel methods for communicating to decision makers and the broader community. To begin, Engdahl describes groundwater modeling using the innate capabilities of quantum computing. He notes that “…very little about how a QCP [quantum computing platform] … is the same as how our current BDCs [binary digital computers] work … Next generation GW models should not just mimic our current models but should instead address frustrations or deficiencies while building new capabilities, like better UQ for predictions.”

Next up, two groups of authors address certain challenges of model inputs. Brookfield et al. test four approaches to determining “Irrigation pumping – how much?”, while Ma et al. use artificial intelligence (AI) and machine learning (ML) methods to explore “Water levels – how deep?” at the CONUS scale using Random Forest techniques. Two challenges posed by groundwater quality concerns are then presented. Ozbek et al. evaluate data gaps in the delineation of a PFOA (Perfluorooctanoic acid) plume using PlumeTrackerTM; while Khambhammettu et al. present geologic facies modeling of properties important to contaminant temporal persistence using Traveling Pilot Points (TRIPS) and an iterative ensemble smoother (IES) to solve the resulting categorical inverse problem. Two articles then address challenges posed when defining hydrostratigraphy for CONUS-scale simulations: Swilley compares and contrasts hydraulic conductivity fields produced using geologically and analytically based techniques, while Tijerina-Kreuzer evaluates 80 different CONUS-scale subsurface products to inform a national hydrostratigraphy configuration.

The increased demands of practicing modelers push existing software capabilities and drive development. MODFLOW 6 advances are presented by Langevin and Hughes, who demonstrate that the architecture supports existing capabilities and enables expansion via the flexible discretization and package integration options it offers.

A technical commentary and software highlight address aspects of model-data interaction. Traylor presents an efficient data integration method using “localization” of observation importance to parameters, while McLane uses inverse modeling and analytic elements to identify potentially important new observations.

Attendees at the 2022 MF&M conference reaped the benefits of witnessing these presentations in person and engaging with the authors in the wonderful surroundings of Princeton. We hope to see readers at the next conference scheduled for June of 2024 at the Princeton campus!