Practitioners participate in a group exercise around a table during a training session on using climate projections.
Practitioners from various disciplines participate in a group exercise during a training session on using climate projections. Representatives from the U.S. Department of the Interior’s regional Climate Adaptation Science Centers train planners and decision-makers to work with downscaled climate projections by solving a real-world management challenge. Credit: Jessica Blackband

Many planners and decision-makers—some of whom have already experienced clear impacts of climate change firsthand—hope to adapt to or mitigate future impacts by using climate model projections to improve outcomes for their organizations or jurisdictions. However, these practitioners may be frustrated to find there is not one “best” climate projection available for their particular application. How does a nonscientist practitioner choose among—or combine—various climate model projections and determine how much confidence to place in the results?

Another problem practitioners face is that these climate model projections typically must be downscaled from coarse-scale global models to higher spatial resolutions if they are to provide meaningful, spatially detailed local information on future temperatures, precipitation, or other variables. For example, water managers may use downscaled precipitation projections to plan for future changes in drought or heavy rainfall events that affect water quantity and quality in their municipality or state. How can practitioners identify which downscaled climate projections are appropriate for their application?

Scientifically justifiable interpretations and applications of climate projections can aid planning decisions and ultimately result in more resilient communities, businesses, and peoples. On the other hand, misapplication could be costly at best and maladaptive at worst [e.g., Nissan et al., 2019].

In 2017, our team developed and offered a role-playing activity to better equip practitioners in their efforts to include downscaled climate model projections in their adaptation planning efforts. We have now implemented this activity multiple times with various participants and have found that participant groups are more comfortable using climate projections and working with climate scientists as a result.

A Daunting Task

Our collective experiences at the U.S. Department of the Interior’s regional Climate Adaptation Science Centers (CASCs) have shown us the frustration that practitioners face when multiple sources of uncertainty [e.g., Wootten et al., 2017] confront them with the need to use many projections. For example, current best practices at the South Central CASC encourage decision-makers to use multiple emissions scenarios, global climate models, initial conditions (i.e., for natural variability), and downscaling techniques in creating an “ensemble” of projections that represent a realistic range of uncertainties in our future climate.

Climate projections have multiple sources of uncertainty, so policies and plans should be designed flexibly to accommodate multiple possible outcomes.

This task is daunting for practitioners, who typically lack the necessary time and climate science staff to create such ensembles. Plus, each application depends on organization-specific factors (e.g., risk tolerance, time horizon, geographic region). Therefore, the combined expertise of climate scientists and practitioners is desirable to identify, interpret, and apply a useful and scientifically justifiable subset of future projections. How do we encourage these collaborative relationships?

In our hands-on training for practitioners, our CASC team aimed to build participants’ confidence in working with downscaled climate projections by solving a real-world management challenge.

Training Structure

Our team first conducted the hands-on activity at the 2017 National Adaptation Forum in Saint Paul, Minn. After a brief, nontechnical presentation that introduced the scientific uncertainties in climate projections, we simulated a real-world water management challenge in breakout groups of six to 10 people. Participants varied in disciplines, backgrounds, and professions; their specialties included energy policy, hydrology, air quality, conservation biology, environmental planning, and health care. Some participants had no prior experience using climate projections, while others had applied projections in several projects.

At each table, CASC facilitators played the roles of climate scientist and science translator (i.e., someone who both understands the relevant science and is capable of clearly communicating it to nonspecialists). Participants each played the role of district water managers who had to recommend whether their state governor should sign a 50-year contract to sell water from a regional aquifer to a growing, water-hungry metropolis in a neighboring state.

Our activity sheets detailed the challenge, including physical (e.g., watershed attributes) and societal (e.g., water needs) information, along with a suite of climate plots. These plots included annual and springtime precipitation from historical climatology (1981–2010) and future regionally downscaled projections (early-, mid-, late-20th century) from three global climate models, three downscaling techniques, and two future emissions scenarios (e.g., Figure 1). We omitted state boundaries on the plots to expand the applicability of the activity to multiple regions.

One training activity plotted 1981–2010 average annual precipitation and the projected change in precipitation by 2070–2089.
Fig. 1. Example plots from one training activity represent (a) 1981–2010 average annual precipitation across a selected area as well as the average projected change in annual precipitation by 2070–2089 relative to the earlier period using (b) a lower emissions scenario (Representative Concentration Pathway (RCP) 2.6) and (c) a higher emissions scenario (RCP 8.5). In Figures 1b and 1c, projections are based on three different global climate models (GCM) and three different downscaling techniques (DS) and are representative of the sources of uncertainty in such climate projections. Dots represent the location of the city providing water (yellow) and requesting water (blue). The yellow rectangles represent the regional aquifer. Credit: Adrienne Wootten

Participants were not expected to gain a complete understanding of climate projections. Rather, they saw a glimpse of how to use projections in their own planning efforts. We also demonstrated the benefits of building relationships among practitioners, science translators, and climate researchers, encouraging collaboration among these groups.

What We All Learned

During the activities, we found that a participant’s prior experience using climate model projections was the most important factor in having successful group discussions. If small groups comprised both experienced and inexperienced users, our facilitators struggled to keep every participant engaged. Group members with similar levels of experience with projections spoke with one another; they disengaged from discussions when information was either too basic or too advanced for them.

Science translators at the trainings had to use strong facilitation skills to direct conversations and strike appropriate balances between discussions of climate factors and other related and tangential topics. Some participants, for example, focused significant attention on nonclimatic details of the management challenge (e.g., how water would be moved or how elections or water prices might affect the agreement), distracting others from the primary goal of applying the climate projections. However, participants with more climate projection experience benefited from the added realism of discussing these nonclimatic factors.

After the activity, we asked participants what they had learned. The most common responses were that climate projections have multiple sources of uncertainty, so policies and plans should be designed flexibly to accommodate multiple possible outcomes, and that climate science translators (e.g., at the CASCs) can assist practitioners in using projections.

Many practitioners were unaware that these science translators are available as a resource; thus, the climate science community must continue building bridges to practitioners. Overall, participants stated that they better understood how to interpret the ensemble of climate projections, would use them for their own planning, and would seek climate science translators to help them. These outcomes are similar to those of Rumore et al. [2016], who have organized community engagement activities related to climate information, adding more evidence that role-playing activities can improve the scientific literacy and collaborative capacity of participants.

Moving Outward and Forward

Our activity has been adapted for multiple venues. In 2018, we conducted a training at the Inter-Tribal Emergency Management Coalition Summit to assist tribes in Oklahoma with all-hazards preparedness planning. In 2017, 2018, and 2019, we conducted the role-playing activity for undergraduate students in climate science internship programs and climate change courses at the University of Oklahoma. In addition, members of Natural Resources Canada and Environment and Climate Change Canada (ECCC) held a training at the Eighth Annual National Roundtable on Disaster Risk Reduction and Canada’s Climate Change Adaptation Platform Plenary—developed after an attendee of our 2017 National Adaptation Forum session expressed interest in adapting the activity for their constituents. Since then, modified versions of the activity have been held in Madison, Wis., at the 2019 National Adaptation Forum and for Canadian audiences through the Canadian Centre for Climate Services at ECCC.

In response to participant feedback, we will expand our water management–centered activity to other management challenges, such as emergency management or public health. Eventually, we will design introductory and advanced versions of our trainings to reduce the difficulty in facilitating groups with different experience levels. We also intend to transfer our team’s knowledge and materials to others who desire to partner with practitioners by developing a “train-the-trainer” short course.

With the continuing development, sharing, and implementation of such hands-on trainings, we hope to foster improved understanding of climate projections among as many local and regional planners as possible, so they can confidently apply this knowledge to improve the flexibility and resilience of their communities in the face of a changing climate.

For more information, contact


The Department of the Interior’s South Central CASC funded this activity under Cooperative Agreement GA12AC00002 from the U.S. Geological Survey. The contents herein are solely the responsibility of the authors and do not necessarily represent the views of the CASCs or USGS. We thank Ryan Bisel for his helpful guidance in communication and organizational culture. We appreciate the work of science translators across the CASC network. And to our participants, we say, thank you! We could not have been successful without you.


Nissan, H., et al. (2019), On the use and misuse of climate change projections in international development, WIREs Clim. Change, 10, e579,

Rumore, D., T. Schenk, and L. Susskind (2016), Role-play simulations for climate change adaptation education and engagement, Nat. Clim. Change, 6, 745–750,

Wootten, A., et al. (2017), Characterizing sources of uncertainty from global climate models and downscaling techniques, J. Appl. Meteorol. Climatol., 56, 3,245–3,262,


Author Information

Derek H. Rosendahl (, Renee A. McPherson, and Adrienne Wootten, South Central Climate Adaptation Science Center, University of Oklahoma, Norman; Esther Mullens, Department of Geography, University of Florida, Gainesville; Jessica Blackband, George Washington University, Washington, D.C.; and Alex Bryan, Williams College, Northampton, Mass.


Rosendahl, D. H.,McPherson, R. A.,Wootten, A.,Mullens, E.,Blackband, J., and Bryan, A. (2019), Making sense of local climate projections, Eos, 100, Published on 14 November 2019.

Text © 2019. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.