In 2015, GLISA formally started the development of an Ensemble of future climate projections for the Great Lakes region. This project is motivated by the need for high-quality climate projections for use in climate change adaptation work. Previous evaluation of a subset of models for the region revealed strong inconsistencies between observed and simulated physical processes of lake-land-atmosphere interactions—the U.S. Great Lakes are known for their impact on local and regional weather and climate, however, the processes responsible for producing lake-effects and lake-induced modifications of weather are often poorly represented or missing from climate models.
To address the need for high-quality climate projections for the Great Lakes region, the Ensemble work will:
1. Develop an evaluation framework, specifically tailored to the Great Lakes region, to provide a regional perspective on the quality of information coming from the models.
2. Apply the evaluation framework to several climate model data sets—including regional modeling efforts—to provide expert guidance regarding the limitations, shortcomings, and appropriate uses of the data.
3. Integrate projections from the models that “pass” our evaluation framework into GLISA’s existing products (i.e., regional climatologies) to provide narratives and visual representations of future climate change information to stakeholders.
What is an ensemble?
"Ensembles" are collections of data from multiple climate model simulations. Typically, an ensemble consists of data from several different climate models to show the range of variability. Given that no single model perfectly simulates the dynamics governing any scale of climate (global, regional, or local), the reliance on several model simulations allows users to better characterize the range of possible future climate outcomes. To that end, ensembles provide measures of uncertainty related to the model-based information.
Why is GLISA developing a Great Lakes Ensemble?
Many climate data sets exist (including downscaled climate data) that do not provide credible information for the Great Lakes region. Many models poorly represent the Great Lakes, so their information about future climate is not based on known interactions between the lakes and the atmosphere. GLISA is evaluating the models to establish a set of those that best represent important components of Great Lakes regional climate. In addition to model evaluation, data processessing (i.e., downscaling) methods will also be assessed for their impact on the quality of the data.
How is GLISA developing the Great Lakes Ensemble?
There is no standard methodology for developing ensembles, but using simple model selection procedures with detailed documentation is one way to build a credible ensemble.1 GLISA starts by considering widely accepted climate model data sets (global and regional) within the climate community. Then, GLISA eliminates models that do not meet the following criteria:
1. The model provides data that is continuous in space and time
a. The Great Lakes are simulated in some fashion
b. A signal of the lakes (i.e., effect of the lakes on regional air temperatures and precipitation) is apparent
3. Any downscaling method applied to the model does not assume climate stationarity
Active Project Work
Additional resources and information related to the Ensemble work can be found on its project page at the Climate Workspace. This site is where project members are actively collaborating and developing materials to support the Ensemble.
- 1. a. b. Overland, J. E., Wang M., Bond N. A., Walsh J. E., Kattsov V. M., & Chapman W. L. (2011). Considerations in the Selection of Global Climate Models for Regional Climate Projections: The Arctic as a Case Study*. Journal of Climate. 24(6), 1583 - 1597.
- 2. McSweeney, C. F., Jones R. G., & Booth B. B. B. (2012). Selecting Ensemble Members to Provide Regional Climate Change Information. Journal of Climate. 25(20), 7100 - 7121.
- 3. Stainforth, D. A., Downing T. E., Washington R., Lopez A., & New M. (2007). Issues in the interpretation of climate model ensembles to inform decisions. Philosophical Transactions Of The Royal Society A-Mathematical Physical And Engineering Sciences. 365, 2163-2177.