Based On The Scale Effect In SWAT Hydrological Processes

Soil and water resources issues attracted widespread attention in the context of global climate change, the study of landscape pattern and ecological processes is the focus of ecological research and hotspot issues, the size effect on the hydrological processes is the main research direction. Based on ArcSWAT module under ArcGIS platform, the paper based on the upper reaches of the Hanjiang River Basin (Ankang Hydrological Station area) as the study area, and built the upper reaches of the Hanjiang River hydrological process model using digital elevation model, meteorological, soil and land/cover data. First, the analysis of different catchment threshold hydrological feature extraction and its impact on hydrological processes to determine the optimum model catchment threshold. Secondly, to express the impact of changing grain size on the DEM and land use data to further explore the sensitivity to the effect of the multi-scale DEM and LUCC data granularity change on hydrological processes and reveal its own rules, provide the basis for the scale conversion of hydrological processes. Finally, paper build the upper reaches of Hanjiang high-precision hydrological process model, verify the applicability of the model parameter sensitivity analysis and calibration of the model by SWAT-CUP software, and temporal and spatial variation of rainfall-runoff relationship in the region for nearly40years the analysis reveals the importance of rainfall on hydrological processes. The main contents and conclusions are as follows:(1)Sub watershed division and analysis of hydrological processes. The watershed threshold smaller, more to increase the number of sub-basins divided hydrological response unit fragmentation, low efficiency of the model calculations, the outlet runoff and sediment concentration was not significant.40000hm2catchment of the threshold to ensure river basic characteristics and the accuracy of hydrological process simulation, is the simulation optimal threshold of the upper reaches of the Hanjiang River Basin.(2) The terrain scale analysis results. Three paired watershed area watershed topographic features as the sampling unit area is increasing, the three basin topographic features changed in the same elevation and maximum elevation difference tends to reduce the minimum tends to increase basin the average elevation and standard deviation were presented slowly increasing and decreasing trends. Maximum slope, average slope and slope standard deviation showed a decreasing trend. Topographical features on270m resolution DEM displacement minimum comparing30mDEM, is terrain characteristic scale.(3) Hydrological processes analysis on topographic scale. With the increase of DEM sampling grid area, the small watershed sediment yield sampling grid average slope significant downward trend from30m to1020m level, can be controlled. The upper reaches of the Hanjiang River sediment yield and runoff overall downward trend showing fluctuations grid in the30m-270m level, the regularity of changing grain size on hydrological processes, we can use a mathematical model to generalize, hydrological processes research DEM resolution can be pushed270m.(4) Land use and land cover pattern analysis of different resolutions. Sampled in the resolution of30m-300m level, plaque size smaller amplitude showed a slight fluctuation. Aggregation were significantly lower, Shannon diversity index showed minor fluctuations, the area weighted fractal dimension at a resolution of300m when the area weighted fractal dimension close to the30m,120m due to the crest. When the sampling grid is greater than300m level, resample raster pixel area of the larger polygons boundaries and area differences, the greater area of change is uncertain. Area weighted fractal dimension declining, stable aggregation area, Shannon diversity index fluctuations amplified significantly uneven land use types.(5) Different resolution land use and land cover change hydrological processes analysis results. Sampling grid30m-300m level, changes in land use pattern of the upper reaches of the Hanjiang River is relatively weak, is not sensitive to the impact of hydrological processes. Sampling grid is greater than300m, changes in land use pattern changes relatively obvious, but its mode of change on hydrological process more complex. On the whole, the land-use changes in data granularity weak influence on the hydrological process simulation of SWAT model, the SWAT model suitable for large temporal and spatial scales of land use change analysis.(6) The papers build the upper reaches of the Hanjiang River SWAT model using precision data, select1999-2004model rate regularly, GLUE model SWAT model parameter sensitivity analysis and correction, the simulation results (R2=0.85, ENS=0.78RSR=0.46). Extraction2005-2010water-caliber stream data and the well-being of station measured data validation, correlation coefficient R2=0.87the ENS and RSR were0.95and0.22, are in the range of values in a high-quality solution, high precision mold, The simulation results are reliable mode.(7) Rainfall is the power source of the entire hydrological processes, its impact on different scales. The spatial distribution of average annual rainfall in the basin is680-1412mm, basically showing the trend of the south than in the north; average annual runoff depth of the sub-basin level of distribution and space distribution of average annual rainfall, and the same level of extreme value area, runoff depth up to668mm and a minimum of133mm. The interannual variability of the average annual runoff depth of each sub-basin intense rainfall, the region runoff impact factor is not just rainfall led, as well as other factors interfere with the changes.The thesis Features as follow:Scale study, the grain size range is too large and small range from30m for the data source, using the nearest neighbor resample grid cell sampling methods for60m,90m,120m,150m,180m,210m,240m,270m,300m,360m,420m,480m,540m,600m,660m,720m,780m,840m,900m,960m and1020m22scale DEM and LUCC data landscape pattern analysis, and simulation of hydrological processes andinput data sensitivity analysis, to make up for the lack of landscape ecology and the total size change, and provide the basis for the scale conversion of hydrological processes.Landscape pattern often only study the pattern of the number and configuration changes not the pattern of deep-seated changes in the ecological significance. This article looks at the landscape pattern and ecological processes. Hydrological Processes Modeling Based on the ArcGIS platform ArcSWAT module, terrain and land-use two input data for hydrological processes.