Simulations And Analysis Of Landscape Pattern Change Impacts On Catchment Runoff At Different Temporal Scales Based On SWAT

As one of the most important drivers leading to hydrological changes, landscape pattern change may influence interception, evapotranspiration, soil infiltration, and eventually lead to runoff change. Relative to longer time-scale runoff, the shorter time-scale runoff is more unstable and complicated relationship with underlying surface. It is meaningful to understand underlying surface change impact on hydrology processes through the analysis of runoff response to landscape change at multiple time scales. Jinjiang catchment is located in the south-eastern coastal region of China with significantly land use changes. What’s more, precipitation is unevenly distributed inner year and there were much flood, drought disasters in the past time. Therefore, the research conducted a comprehensive investigation into runoff responses on the daily scale as well as annual and monthly time scales using SWAT in Jinjiang catchment, and compared the impacts of different time scales quantitatively, which can provide more reliable information for water resource allocation, reducing flood and drought disasters.The watershed controlled by Shilong gauge was selected as the study area. Runoffs at three hydrological stations along Jinjiang river were simulated by the SWAT model for nine years (2002-2010), under the same meteorological inputs but two different landscape scenarios (1985and2006). Moreover, the Pearson correlation was applied to quantify the contribution of the changes of landscape indices on the variation in flood and low flow. The results indicated that:1) Parameters sensitivity and parameter value of SWAT model have some regional difference because of regional heterogeneities of climate, soil and vegetation.2) SWAT model can simulated yearly, monthly and daily runoff process accurately, respectively.3) From1985to2002, the most significant characteristics were the massive decrease of cropland, grass and forest, and distinct increase of orchard and urban.4) Different time scales indicated different runoffs response under landscape change, the response of annual runoff was slight, and a remarkable response was found on daily runoff, such as peak or drought flow. Changes of monthly runoff between the two scenarios showed a difference among seasons:decreased in drying season but increased in remaining months, especially in the spring.5) The characteristics of precipitaion influence the runoff response, dry year was more remarkable than wet year and the wetter the catchment soil was prior to the flood event the smaller the change ratio was.6) There are significant correlations between the variation of flood flow and landscape indices. It’s negatively related with forest area, LPI and CONTAG, and positively related with SHDI and SHEI. However, there was not significant correlation between drought and landscape indices.