Research On Watershed Soil Erosion And POC Flux Into The Sea Based On SWAT Model

Soil erosion seriously damages the natural ecological environment and restricts the sustainable development of the national economy.At the same time,soil erosion causes a large amount of organic carbon to be carried into the river,which in turn leads to a significant increase in river particulate organic carbon(POC)output flux.Thus,quantifying the soil erosion,determining the Critical Source Areas(CSAs)of soil erosion,and analyzing the impacts of meteorology,topography,land use types,landscape patterns on soil erosion and POC output flux are not only beneficial to soil and water conservation and management planning in the basin,but also important for river aquatic food web and global carbon cycle.The distributed hydrological model is an effective tool for simulating hydrology,erosion and sedimentation processes.The SWAT model,which adequately consider the factors of topographic,land use,soil and meteorological,can adequately simulate runoff,sediment,soil erosion and of the POC output flux.The rivers,in the south of the Qinling-Huaihe Line,have a large flux of runoff and play an important role in supplying freshwater to urban agglomerations in Chinese important economic centers,the Yangtze River Delta and Pearl River Delta,as well as in the formation and development of the delta.The sediment yield and POC flux of rivers in the region have attracted the attention of scholars at home and abroad.The temporal and spatial variation characteristics and influencing factors of sediment erosion and POC output in the four major river basins(the Yangtze River Basin,the Pearl River Basin,the Huaihe River Basin,and the southeast coastal rivers basin)in southern China are discussed,which are of great significance to regional ecological environment and terrestrial-ocean carbon cycle.The main contents and research results of this paper are as follows:(1)Based on DEM,land use,soil,meteorological data,etc.,the SWAT models for the four major river basins in the southern China were built respectively,to simulate the water discharge and sediment yield of each basin.Due to the limited data of hydrological monitoring stations,the Yangtze River Basin used monitoring data of water discharge and sediment yield from 2006-2007 to calibrate the model,whereas the other three basins(the Huaihe River Basin,the Pearl River Basin,and the southeastern coastal rivers basin)used monitoring data of water discharge and sediment yield from 2006 to 2011.For water discharge,during the calibration period,determination coefficient(R2)value of each basin was>0.80,the mean value of Nash coefficient(ENS)was≥0.72;in validation period,the average value of R2 was≥0.72,and the mean value of ENS was>0.62.For sediment simulation,in the calibration period,R2 value of each basin was≥ 0.72,ENS values were>0.63;in validation period,R2 values were≥0.72,ENS values were≥0.58.This result implies that the SWAT model can be used to effectively simulate the water discharge and sediment export in the four major river basins.(2)The temporal and spatial characteristics of sediment export and sediment erosion in each basin from 1988 to 2012(the Yangtze River basin was 1990-2010)were obtained,based on the calibrated models.In the study area,sediment export of the flood season(May-September)was higher than that of the dry season.The annual average sediment export into the sea of each basin decreased in the order of the Yangtze River Basin,the Pearl River Basin,the Huaihe River Basin,and the southeast coastal rivers basin,with values of 2.19×108 t/yr,8.56×107 t/yr,and 1.08×107 t/yr,3.70×106 t/yr,respectively.The intensity of soil erosion decreased in the order of the Pearl River Basin,the Yangtze River Basin,the Huaihe River Basin,and the southeast coastal rivers basin,the soil erosion modulus was 167.21 t/km2/yr,121.45 t/km2/yr,35.36 t/km2/yr,20.19 t/km2/yr,respectively.(3)The temporal and spatial characteristics of POC export in each basin were simulated using improved source program of sediment erosion module in SWAT.Similar to the sediment export,the POC export of each basin was mainly concentrated in the summer(May-September).The average annual POC export into the sea decreased in the order of the Yangtze River Basin,the Pearl River Basin,the Huaihe River Basin,and the southeast coastal rivers,with values of 2.13 × 106 t/yr,1.94× 106 t/yr,9.84×104 t/yr,4.70 x 103 t/yr,respectively.The spatial distribution of POC erosion in each basin was similar to that of sediment erosion.The POC erosion modulus of the basin decreased in the order of the Pearl River Basin,the Yangtze River Basin,the Huaihe River Basin,and the southeast coastal rivers basin,with values of 3.79 t/km2/yr and 1.18 t/km2/yr,0.32 t/km2/yr,0.03 t/km2/yr,respectively.The average annual POC yield of each basin decreased in the order of the Yangtze River Basin,the southeast coastal rivers basin,the Pearl River Basin,the Huaihe River Basin,with values of 1.7 t/ha/yr,1.487 t/ha/yr,1.081 t/ha/yr,0.059.t/ha/yr,respectively.(4)The Geographically Weighted Regression(GWR)model was used to quantitatively analyze the spatial response of sediment erosion to rainfall,slope,land use and other influence factors in each basin.And the spatial distribution of the contribution of the slope,rainfall and land use to sediment erosion in CSAs were calculated.The results showed that,in the study area,the average contribution of various factors to sediment erosion decreased in the order of rainfall,slope and land use.In the study area,the main land use types were forest,grassland and cultivated land,and the order of the corresponding sediment yield was:grassland>cultivated land>forest.In the Huaihe River Basin and the Yangtze River Basin,the sediment yield has the strongest spatial response to grassland.In the Pearl River Basin,the sediment yield has the strongest spatial response to cultivated land,and that in the southeast coastal rivers basin was forest.For the Critical Source Areas of soil erosion,the CSAs in the Huaihe River Basin were mainly distributed in the mountainous in the eastern,southern and southeastern areas of the basin.In the Yangtze River Basin,CSAs were mainly distributed in the Yalong River Basin,the Lancang River Basin and the Jialing River Basin,that are distributed in the junction of the first and second steps,as well as the Xiangjiang River Basin and the Gancang River Basin.In the Huaihe River Basin,CSAs were mainly distributed in the northern part of the basin,ie,Nanbeipan basin,Hongshui basin,Beijiang basin and Dongjiang basin.In the southeastern coastal rivers basin,CSAs were mainly distributed in the Minjiang Basin,the Minnan Basin,and the southwestern part of the Qiantang River Basin.Within the CSAs in all basins,the order of average contribution of different factors to sediment erosion was:rainfall>slope>land use.(5)The spatial responses of POC erosion to sediment yield and soil TOC content were quantitatively analyze by the GWR model.In the study area,there was a strong spatial correlation between the spatial distribution of POC load and sediment yield,and the relevant intensities were in the order of the Yangtze River Basin>Huaihe River Basin>Pearl River Basin>southeast coastal rivers basin.The spatial correlation between the spatial distribution of POC load and the soil TOC content was weak,and the correlation intensity decreased in the order of the Pearl River Basin,the Yangtze River Basin,the southeast coastal rivers basin,and the Huaihe River Basin.(6)The effects of changes in landscape pattern on soil erosion and POC erosion in the basin were analyzed.The results showed that,in our study area,the changes of Patch Cohesion Index(COHESION),Modified Simpson’ s Evenness Index(MSIEI)and Largest Patch Index(LPI)played a decisive role in soil erosion.The effect of these indicators on soil erosion was MSIEI>COHESION>LPI.Both COHESION and LPI were positively correlated with soil erosion modulus,whereas MSIEI was the opposite.POC erosion was closely related to changes in Patch Density Index(PD),Shannon’s Diversity Index(SHDI)and Fractal Dimension Index(FRAC),the correlation between SHDI and POC erosion was negative,and the correlation between the other indicators(PD,FRAC)were reversed.FRAC has the greastest impact on POC erosion,whereas the impact of SHDI was the weakest.Those indicated that soil erosion can be reduced by enriching the types of landscape patches,reducing the physical connectivity between patches,improving the types of landscape patches,and balancing the proportion of landscape patches.And POC erosion can be reduced by increasing the diversity of the landscape,reducing the geometry complexity of plaque types and the fragmentation of the landscape.