Hydrologic response of watersheds to urbanization in the upper Great Lakes region

The Upper Great Lakes Region (UGLR) has experienced significant land cover/ land use change (LCLUC) since the last century, especially the expansion of urban land use at the expense of both forest and agriculture. Impervious surface area (ISA) has different surface characteristics from the natural land cover and has great influence on watershed hydrology. Water in highly urban landscapes flows over impervious surfaces or through drainage networks that can greatly increase the speed of flow to receiving water bodies. This can demonstrably reduce infiltration rates on the land surface and runoff response times, resulting in increases in flood frequency. Increased ISA also leads to different surface temperatures between urban and rural areas, resulting in urban heat islands (UHI). Subsequent changes in urban climate due to the increased land surface spatial heterogeneity can result in enhanced convergence during convective events. The overarching goals of this study are first to develop integrated modeling tools for urban hydrology studies at basin scales, and then to investigate the hydrologic response of watersheds to urbanization in the UGLR using these tools. A physically-based macroscale hydrologic model, the Variable Infiltration Capacity (VIC) model was modified to represent the effective impervious area (EIA) in urbanized basins using a bulk parameterization approach. An empirical model was established to estimate the EIA using urban landscape metrics based on land use maps only. The routing scheme of the VIC model was improved using a GIS-based routing model, which can preserve the spatially distributed hydrologic and geophysical characteristics in each VIC model grid cell. The performance of these tools was examined by comparing the model simulations with the observations through different hydrologic metrics. Results demonstrate the increase of streamflow flashiness (R-B index) and flood frequency, and the decrease of flow distribution (Tqmean) with increasing urban area. Above 3%–5% impervious area in a watershed urbanization starts to have a statistically significant influence on streamflow regime. In addition, the travel time of water from urban locations to the basin outlet plays an essential role in flood peaks. Flood peaks are likely to be smaller and come earlier if urban development occurs in the lower basin; this means the largest impacts of urbanization are not necessarily seen immediately downstream of the development. Utilizing the stochastic time series fractional autoregressive integrated moving average (FARIMA) model to further analyze the VIC model simulations, the long-term persistence (long memory) in hydrologic systems was found to decrease with urbanization in this entire region. The decrease of long memory is mainly due to the decreased low frequency power of soil moisture in the deep soil layer and the decreased amplitude of the soil moisture cycles. Finally, an integrated model coupling approach among land cover, atmospheric and hydrology models was used to assess the relative contribution of the increased urban area and the corresponding precipitation feedback on total runoff during thunderstorm events at HUC8 basin scale. For regions with heavy precipitation, urbanization can enhance the precipitation process with stronger convergence zones. Although there is a consistent and statistically significant increase in runoff from impervious areas, results for this study show no statistically significant correlation exists between the watershed urban percentage and the change of precipitation at the watershed scale. Overall, this study reinforces the dynamic linkage between urbanization and soil moisture which plays an important role in assessing the impact of urbanization on hydrology. The results in this study should be useful for environmental and urban planners in forming a sustainable urban development strategy