Climate Change and Climate Extremes

  • How will the Great Lakes regional climate and lake-atmosphere-land interaction evolve under the warming climate?
  • How will the changing regional climate impact coastal microclimates and extreme events? 
  • What are effective strategies for disseminating our in-house future climate projections for the Great Lakes region in formats that enhance their applicability and usability for stakeholders?
  • How can the application of machine learning and artificial intelligence enhance the accuracy and reliability of seasonal water level predictions for the Great Lakes?

To answer these questions for the Western Upper Peninsula, we leverage our team’s expertise in regional climate modeling and projection, demonstrated through advanced regional climate modeling systems like the Great Lakes Atmosphere-Regional Model (GLARM) and the Climate Risk and Resilience Portal (ClimRR), each uniquely tailored to future climate projections. 

Developed at Michigan Tech by Dr. Pengfei Xue and team, GLARM is a regional climate modeling system that two-way couples the the International Center for Theoretical Physics (ICTP) Regional Climate Model version 4 (RegCM4) with a three-dimensional (3D) representation of the Great lakes using a hydrodynamic lake and ice model based on the Finite Volume Coastal Ocean Model (FVCOM). RegCM4’s spatial domain covers the Great Lakes Basin with a horizontal 18-km spaced grids, while FVCOM provides finer detail within the lakes, with horizontal resolutions ranging from ~1–2 km near the coast to ~2–4 km offshore. By integrating a 3D representation of the Great Lakes, GLARM effectively captures detailed lake-atmosphere interactions which are crucial for accurately modeling the Great Lakes regional climate. Lake-effect phenomena such as suppression of temperature variability and enhanced precipitation are well captured in GLARM, making it a valuable tool for projecting the regional climate under climate change. 

Climate projections from regional climate modeling systems are derived by dynamically downscaling climate projections from coarser-resolution global climate models (GCMs). Additionally, climate projections from GCMs are based on various future scenarios, each scenario assuming different trajectories of greenhouse gas emissions and socio-economic factors. One such group of scenarios is the Representative Concentration Pathways (RCPs) which consist of pathways created for the Intergovernmental Panel on Climate Change (IPCC) assessments. Each RCP pathway assumes different levels of future greenhouse gas concentrations and socio-economic variables. Among the numerous RCP pathways, the most commonly referenced include RCP 4.5 (representing a moderate future) and RCP 8.5 (representing a worst-case scenario future).

Using GLARM, we produce future climate projections by dynamically downscaling three different GCMs for the entire 21st century under the RCP 4.5 and 8.5 scenarios. Projections for daily 3D lake temperature, ice cover, and 2-meter air temperature from GLARM are currently available for download (as NetCDF files) from the MTU Digital Commons repository under the dataset name, GLARM-Proj1. Detailed documentation of GLARM and its climate projections can be found in these 2017 and 2022 papers.
In contrast to GLARM, ClimRR is developed by Argonne National Laboratory and is a web-based portal that allows the public to access future climate projections derived from their regional climate modeling system which uses the Weather Research and Forecasting (WRF) model. The major difference between GLARM and ClimRR is the fact that GLARM has a two-way coupled 3D representation of the Great Lakes while ClimRR does not. Thus, the pronounced effects of the Great Lakes on the regional climate are not accurately captured in ClimRR projections. Nevertheless with its 12-km horizontal spatial resolution and its interactive web portal where people can easily access and visualize future projections for over 60 climate variables (such as air temperature, precipitation) and indices (such as fire weather index, heat index), ClimRR is a valuable climate modeling tool in the C-CHARM initiative. ClimRR provides projections from three GCM downscaling under RCP 4.5 and 8.5 for the historical, mid-21st century and end-21st century.

Outcomes

Publications (forthcoming)

  1. Kayastha, M. B., Xue, P., Liu, T., Titze, D., Havens, T. C., & Huang, C. (2025). Enhancing Coastal Climate Resilience with Advanced Projections: High-Resolution 3D Modeling of the Great Lakes Thermal Structure for the 21st Century. 105th AMS Annual Meeting, New Orleans, Louisiana, January 12-16 2025
  2. Woolway, R. I., Kayastha, M. B., Tong, Y., Feng, L., Shi, H., & Xue, P. (2024). Subsurface heatwaves in lakes. Nature Climate Change (Under Review)
  3. Xue, P., Huang, C., Zhong, Y., Notaro, M., Kayastha, M. B., Zhou, X., Zhao, C., Peters-Lidard, C., Cruz, C., & Kemp, E. (2024). Enhancing Winter Climate Simulations of the Great Lakes: Insights from a New Coupled Lake-Ice-Atmosphere (CLIAv1) Model on the Importance of Integrating 3D Hydrodynamics with a Regional Climate Model. Geoscientific Model Development (Under Review)
  4. Zhu, L., Meadows, G. A., Kayastha, M. B., & Xue, P. (2024) Sediment Transport and Budget Influenced by Harbor Jetties in Storm Events. Journal of Great Lakes Research (Under Review)
  5. Xue, P., Huang, C., & Kayastha, M. B. (2024) High-resolution 3D projections of the Great Lakes thermal structure for the 21st century. Nature Scientific Data (Under Review)