Study On Stormwater Management Mode Based On Land Use Pattern Optimization

Stormwater best management practices (BMPs) were developed to achieve three objectives guiding the design of stormwater management plans:flood prevention, ecological restoration, and sustainability. However, the current approaches in designing BMPs have not addressed all stormwater management concerns. The composition and configuration of the landscape, or land-use patterns, strongly influence hydrological connectivity and control catchment responses. Watershed land-use management should be designed to manage stormwater under conditions of spatial heterogeneity at the watershed scale. However, although there are an increasing number of studies on the relationship between changesinland-use types and hydrological processes, few have focused on the relationship between other land-use pattern characteristics (such as pattern fragmentation, connectivity and coherence) and hydrological processes.This study presented a model to analyze the effects of land-use pattern changes on watershed hydrology and to examine the effects of such patterns at multiple scales in minimizing stormwater runoff. The Yong-Ding watershed in western Beijing, China, serves as a case study for this research. We developed a physically based, fully distributed-parameter hydrological model (GSSHA) to derive the impact of various land use patterns on the hydrology of the watershed. In addition, we conducted stepwise regression analysis to estimate the peak flow volume and total volume based on variables representing different types of landscape metrics and land-use patterns. Further, we identified sensitive metrics to be controlled and the most suitable land-use patterns for future development with the help of multivariate regressions. Sensitivity analysis was also conducted to assess the effects on the relationship between land-use patterns and the hydrology related to changing storm intensities.The results reveal that land-use pattern management is suitable to manage stormwater under different sizes of designed storm events and is more effective for small storm events than for large storm events. For the177mm storm event, we estimated that for every increase of1in the LPI of forestland use, the runoff characteristics of a storm event decreased by more by3.903m3/s for the peak flow volume and by165158m3for the total volume compared with the values for the139.5mm storm. This study adds an important dimension to our understanding of the impact of land use patterns on watershed hydrology by introducing multiple parameters of land use types and patches and by showing that the spatial scale of analysis is critical in determining how such patterns relate to stormwater runoff. Additionaly, based on the relationsip between land use pattern and hydrological output of the watershed, this study demonstrates a method of estimating optimal land use pattern for reducing the negative impact of urbanization on watershed stormwater systems. The method utilizes a regression model to estimate how the watershed hydrology responds to the land use pattern change. This model is next integrated with a land use allocation model under the land use pattern optimization modeling system (LPOP model), in order to find the land-use pattern that minimizes the peak flow or total volume at the watershed outlet. The CLUE-S model was used to simulate empirical land-use patterns under different development intensities. The impacts of optimization were detected by comparing the land use pattern characteristics and watershed hydrology of optimal scenarios versus empirical ones, with the same development intensity.The results of land-use pattern analysis suggest that, in order to reduce the peak flow, the forestland patches should be more fragmented, with complex shapes by dissecting the largest one to the smaller ones and merging some small patches with larger ones. The urban land patches should be more fragmented and the grassland patches should have a simple shape. In order to reduce the runoff volume, the forestland patches should be more fragmented with complex shapes, and the agricultural land patches should be more spatially discrete. The results of the hydrological evaluation suggest that, compared to land-use location control, land-use composition and configuration control is also a powerful method for minimizing the negative hydrological impact of urbanization. In the case of minimizing peak flow under the optimized land-use pattern, the peak flow of the land-use patterns for the entire watershed in2030decreased from1993peak flow by3.17-5.49%, and reduced by12.17-23.94%as compared to peak flows resulting from different, empirical scenarios in corresponding urbanization intensity in2030.Then, land-use demand for optimization was detected by comparing the optimized watershed hydrology under different development intensities and types. The results of the land-use demand analysis suggest that,30%of the watershed area should be maintained as forestland to reduce the negative hydrological impact of urbanization; the urban land and the agricultural land in the watershed should represent less than50%and55%, respectively.