Reports and Publications

Recent advances in Machine Learning (ML) have made strides in improving the prediction of river levels, and often outperform physics-based models that make a forecast based on the physics of how rivers respond to precipitation. However, expert human forecasters with field intelligence know how to modify and improve physics-based models to achieve high accuracy. Our research responds to key needs in the operational forecasting community: (a) by making direct comparisons between ML models and official forecasts that are based on both physics-based models and forecaster expertise, and (b) by combining the strengths of pure physics-based and ML approaches. We have developed a new “Knowledge-Guided” ML (KGML) model with its algorithm informed directly by hydrologic science, called the Factorized Hierarchical Neural Network, and demonstrate that it performs as well or better than NWS flood forecasts 2–7 days after a forecast is issued, and better than a leading ML alternative that does not incorporate physical science knowledge in its architecture. An expert human forecaster using a physics-based model is still more skilled than the state-of-the-art ML methods within the first day. If forecasters could use our KGML approach operationally, the skill of river forecasts has the potential to improve substantially.

Climate data and observations show that Minnesota is experiencing consistent changes in weather patterns. This report explores how projections of future weather trends may exacerbate conditions, including but not limited to drought, elevated temperatures and flooding for the design and evaluation of infrastructure and buildings constructed by the state of Minnesota and local governments. In addition, the report assesses the potential of future weather events to weaken existing systems creating the need for intervention to maintain and increase the amount and quality of food and wood production, reduce fire risk on forested land, maintain and enhance water quality, and maintain and enhance natural habitats. Because the relationships between infrastructure, future weather trends and the human-natural systems of agriculture, water, forests, and built environments are complicated, the research team developed a framework to analyze the Social, Ecological and Technological (SETs) relationships within each system, creating a common “language” to analyze potential interactions between multiple complex systems (Chapters 3 and 4). This approach is crucial for decision makers to be effective at mitigating costs and avoiding maladaptation or making things worse from some resilience strategies.

Climate change presents new and compounding challenges to natural resource management. With changing climate patterns, managers are confronted with difficult decisions on how to minimize climate effects on habitats, infrastructure, and wildlife populations. To support climate adaptation decision making, we first conceptualized an approach that integrates the principles of the resist–accept–direct framework, climate scenario planning, and decision analysis into a general framework to support adaptation planning. This framework was implemented and refined by working with three National Wildlife Refuge System refuges within the Midwest Region. The objectives of this report are to describe the climate adaptation decision framework and provide guidance for how to apply the framework to support transparent, consistent, and decision-focused adaptation planning. We include a workbook to support the application of each step of the framework as well as lessons learned from our experiences developing the framework. The climate adaptation decision framework has wide applicability to aid adaptation planning within natural resource management and underscores the important role of engaging interest groups in climate adaptation decisions.

Climate adaptation and the management of climate impacts require cross-sectoral and regional coordination and collaboration, but presently there is no thorough assessment of the adaptation network in the Midwest United States to evaluate how well it achieves such collaboration. We investigated the climate adaptation network across the Midwest to inform the strategic agenda for a climate adaptation boundary organization in Minnesota - the University of Minnesota Climate Adaptation Partnership (MCAP). We identified 150 organizations and more than 500 unique connections between them. About ten organizations with more than 25 connections each link the existing Midwest climate adaptation network, but most organizations have fewer than five connections. This asymmetry can affect the flow of resources such as information, technical assistance, and financial support. It can also hinder coordination and collaboration as called for by the Intergovernmental Panel on Climate Change (IPCC et al., 2019). The Midwest adaptation network is not well-balanced with respect to the adaptation cycle: many organizations focus on understanding or planning for climate change, with few organizations focused on problem identification, plan implementation, or monitoring. The gaps identified here suggest that MCAP and other regional adaptation organizations can (1) improve cross-sectoral and intraregional coordination and collaboration, and (2) fill gaps in the adaptation cycle, particularly implementation and monitoring. As more communities and jurisdictions move beyond climate planning toward adaptation implementation and management, and as an increasing number of state, federal and private sector funds become available to support implementation, climate service providers such as MCAP should evaluate their services and capacities and adapt alongside the communities they support.

The Minnesota Climate Resilience Indicators & Metrics Project, spanning two years from June 2022 to June 2024, represents a comprehensive effort to determine how to evaluate the state's resilience to the impacts of climate change. Led by the University of Minnesota Climate Adaptation Partnership and the Institute on the Environment in collaboration with the Minnesota Pollution Control Agency (MPCA), the project focused on developing climate resilience metrics aligned with the climate actions identified in the Resilient Communities chapter of the Minnesota Climate Action Framework (CAF).

Minnesota is the state with the strongest winter warming in the contiguous United States. We performed regional projections of the climate across Minnesota for the middle and end of the 21st century. We selected the results from eight recent global climate model projections to calculate climate data over an area of 10 km by 10 km with a regional climate model. Our results indicate that the future climate for the state of Minnesota is likely to be significantly different from what has been observed near the end of the 20th century. Over northern and central Minnesota, winters and summers are expected to be up to 6 and 4°C warmer, respectively, near the end of the 21st century. Spring precipitation may increase by more than 1 mm d⁻¹ over northern Minnesota. Over the central part of the state, winter snow depth is suggested to decrease by more than 12 cm. The number of days per year with snow depth of more than 2.54 cm (one inch) is expected to decrease by up to 55. These results are expected to influence regional decision-making related to agriculture, infrastructure, water resources, and other sectors.

MCAP partnered with the USDA's Midwest Climate Hub, the Great Lakes Integrated Sciences and Assessments, the Northern Forests Climate Hub, and the Northern Institute of Applied Climate Science to develop an agricultural vulnerability assessment for Minnesota. This report includes historical climate change information from 1979 to 2021, projected changes under future greenhouse gas emissions scenarios, impacts of these changes (both historical and projected) on agricultural operations, and considerations for adapting agricultural operations to these impacts. The climate metrics presented in these assessments include temperature, precipitation, humidity, and their extremes.

More frequent, effective climate conversations initiated by a diversity of trusted voices can help to increase climate concern and desire for action at the community level. However, in the United States, there is a disconnect between the level of concern individuals have about climate change and the extent to which individuals talk about the issue. To help bridge this gap, the University of Minnesota Climate Adaptation Partnership developed a training program aimed at inspiring and equipping local community members across Greater Minnesota with the skills and confidence to have effective conversations about climate change in their communities. This paper summarizes the programmatic activities we used to support our goals, and some reflections on the program's results. This pilot program provides a framework for future efforts that can be facilitated by Extension programs, community-based organizations, universities, and others to inspire and accelerate similar community-centered climate conversations.

This article presents a comparative case study of public valuation of glaciers in Alaska and Norway. The first case examines Alaska’s Mendenhall Glacier, which has been central in public debate over the US Forest Service’s proposed expansion of the Mendenhall Glacier Recreation Area. The second case centers on Norway’s Svartisen glacier, which garnered international attention when the startup company, Svaice, announced its intent to extract glacier ice cubes for cocktail coolers at high-end bars and restaurants. A rhetorical analysis of newspaper coverage relevant to each case reveals that in both debates, instrumental, relational, and intrinsic values are attributed to the respective glaciers, and that government, business, and community actors hold the most power in these conversations. However, nuances within articulations of instrumental value suggest that Norwegian actors strive to balance human and glacier needs, whereas Alaskan actors largely prioritize human needs by constituting the glacier as a utilitarian object.

Please contact Catherine Bruns for a copy of this publication.