Detection of changes in hydrologic system memory associated with urbanization in the Great Lakes region

The Great Lakes region has experienced large land-cover / land-use change (LCLUC) in the last century, especially the growth of urban areas accompanied with increased human activities. Many hydrologic impacts of urbanization have been documented due to the reduction of the perviousness of urban landscapes, e.g. the reduced infiltration rate and runoff response time, the increased flashiness of daily streamflow and the magnitude of flood peaks. Such changes might be equivalently interpreted as a decrease in hydrologic system memory with urbanization, i.e. the current discharge in a flow time series has less correlation with its prior flows. Based on observed streamflow at several USGS gauge stations in the Great Lakes region, appropriate fractional autoregressive integrated moving average (FARIMA) models are introduced for daily and monthly streamflow respectively. The recently improved Variable Infiltration Capacity (VIC) model with urban representation is applied to simulate the water and energy response in many basins in the region with various degrees of urbanization. The VIC model is driven by different years’ land use/cover maps, including a 1992 base map and a 2030 map predicted by the Land Transformation Model (LTM). A no-urban scenario was also created by removing urban and rescaling other vegetations’ fraction based on the 1992 map. The simulated daily and monthly streamflow from some gauged basins is validated using the above fitted FARIMA models. The difference in the fitted FARIMA model parameters, e.g. AR(1) and Hurst exponent representing short and long memory respectively, between different land use/cover scenarios is mainly induced by differences in urban area. In order to further explore the relationship among urban growth, hydrologic system memory and other hydrologic change, metrics relating to the flow distribution, daily variation in streamflow and frequency of high-flow events are also calculated for selected basins.