Purdue Climate Change Research Center

Urbanization Impacts on the Hydrometeorology of the Upper Great Lakes Region

Funded by NASA

This interdisciplinary project, beginning its third year of funding, addresses the impact of future urban growth on both meteorology and hydrology. Hydrologists have known for some time that through a reduction in infiltration and an increase in fast surface runoff, impervious area leads to a change in the character of river discharge. Similarly, recent research suggests that through increased heating and surface roughness, urban environments can increase convergence during convective events, impacting the magnitude and distribution of rainfall. For both of these, the role of the spatial arrangement, orientation and connectivity of urban development is unclear.

In the last year, land use projections have been completed for five year increments from 2010 to 2030 using the Land Transformation Model for the entire four state study area (Wisconsin, Michigan, Illinois and Indiana). Between 2010 and 2030, the largest rates of residential and commercial growth were forecast in Michigan. Agricultural and wetland areas decrease in all states, with some of the area replaced by forest re-growth in Wisconsin and Michigan.

The 2010 and 2030 land cover projections were used to specify the land surface characteristics for two applications of the Regional Atmospheric Modeling System (RAMS) model version 4.3 with the LEAF2 land surface scheme for a single convective storm event, on June 22, 2006. As shown in the figure below, the projected land cover change between 2030 and 2010 leads to heterogeneous changes in precipitation and temperature during this summer storm event. In general, areas of warming coincide with regions of precipitation decrease and cooling with precipitation increase. The spatial distribution in precipitation changes leads to differences in streamflow response, depending on the orientation of the contributing watershed relative to the urban area.

Streamflow has also been simulated for each scenario using the Variable Infiltration Capacity model with different land cover and weather inputs. For example, the Muskegon River in southern MI shows large changes in simulated peak streamflow due to projected precipitation increases, although the change in streamflow due to impervious area alone is small. In contrast, for the White River in Indianapolis, IN simulated changes in convective precipitation have negligible influence on the simulated hydrograph, but the impacts of increasing impervious area on infiltration are large.


  • Laura Bowling, Agronomy
  • Keith Cherkauer, ABE
  • Bryan Pijanowski, FNR
  • Dev Niyogi, Agronomy and EAS

Contact Information

Purdue University
203 S. Martin Jischke Drive
MANN 105
West Lafayette, IN 47907