Landscape Pattern Optimization Of Danjiangkou Reservoir Basin Based On Water Purification

In recent years,the study of factors affecting basin water quality,such as non-point source pollution,land use/cover type,landscape spatial pattern,has been one of the focuses of basin hydrology and ecology.Danjiangkou Reservoir,as the water source area of China’s South to North Water Transfer Project,undertakes the important task of alleviating the water resource shortage in North China,and is an important water source protection area with national strategic significance.However,the land use around Danjiangkou Reservoir for residents’living,industrial and agricultural production are complex,the risk of point source pollution and non-point source pollution coexist,and it is difficult to control,which greatly threatens the water ecological environment security of Danjiangkou Reservoir Basin,and the water quality control of the reservoir basin is extremely urgent.Based on the analysis of land use and landscape pattern in Danjiangkou reservoir area,combined with the measurement of water quality indicators in Zhangjiashan basin catchment area and Danjiangkou reservoir area basin water quality monitoring station,this study carried out the research on the relationship between"source-sink"landscape pattern index and spatial pattern of basin landscape morphology and water quality;The principal component analysis and multiple linear regression analysis of partial least squares regression analysis were used to discuss the main influencing factors of water quality in the basin.Based on the analysis of landscape connectivity and the minimum cumulative resistance model,the ecological network of water quality purification in the basin is constructed,which is composed of ecological source,ecological corridor and ecological pinch points,in order to provide technical support for the landscape configuration optimization and protection measures of water quality purification in the Danjiangkou reservoir area.The main conclusions of the study are as follows:（1）The correlation analysis between the"source-sink"landscape pattern indexes such as the Flow Accumulation Index（FAI）,Location-Weighted Landscape contrast Index（LWLI）and Landscape Capacity Distribution Index（LCDI）of Zhangjiashan small basin and the Water Quality Index（WQI）of the small basin showed that the correlation between the landscape convergence accumulation index and the water quality index was not significant;The landscape spatial load ratio index was significantly correlated with the water quality index（r=-0.716,P<0.05）;There was a significant correlation between landscape nutrient retention index and water quality index（r=-0.847,P<0.01）.It shows that the basic characteristic parameters of landscape spatial load ratio index and landscape nutrient interception index,such as roughness coefficient,runoff coefficient,nutrient input,flow length and patch number,are the influencing factors of water quality in small basin.（2）The regression analysis and trend analysis of"source sink"landscape pattern and water quality index in Zhangjiashan small basin showed that the Location-Weighted Landscape contrast Index（LWLI）and Landscape Capacity Distribution Index（LCDI）had a strong negative regression relationship with Water Quality Index（WQI）;With the increase of the"source"landscape area of the basin,the landscape nutrient retention index and landscape spatial load ratio index showed an upward trend,while the Water Quality Index（WQI）showed a downward trend.It shows that increasing the area of"source"landscape is not conducive to the improvement of water quality in the basin,and the"source-sink"landscape pattern index can be used to predict the water quality of small basin.（3）The study on the relationship between the spatial pattern of landscape morphology and water quality in Danjiangkou reservoir basin shows that the index of spatial pattern of landscape morphology is also the main influencing factor of water quality in the basin.Among them,the spatial structure of basin landscape morphology that has a negative regression relationship with the pollutant content of basin surface water includes CORE（Regression coefficient,Rc=-0.22）and Bridge（Rc=-0.33）,indicating that the more ecological source patches and corridors in the basin,the better the basin water quality;The spatial structure of landscape morphology with a positive regression relationship with the pollutant content of surface water in the basin includes Islet（Rc=0.16）,Perforation（Rc=0.37）,Edge（Rc=0.16）,Loop（Rc=0.10）and Branch（Rc=0.40）.It shows that the higher the degree of landscape fragmentation and the lower the connectivity in the basin,the worse the water quality in the flow area.（4）The partial least squares regression model for basin water quality prediction was constructed by using land use type and"source-sink"landscape spatial pattern indicators.The comparative analysis of the performance of the two models shows that the basin water quality prediction based on the basin landscape pattern index can greatly improve the interpretation and prediction function of the partial least squares regression model for water pollution degree.Among them,the model’s interpretation rate（R²）of permanganate index（CODmn）,the main indicator of water pollution in the basin,is 99.8%,and the prediction rate（Q²）is 95.5%;The interpretation rate of BOD5 reached 99.1%,and the prediction rate was 37.9%;The interpretation rate of total phosphorus（TP）was 99.5%,and the prediction rate was 91.0%;The interpretation rate of total nitrogen（TN）was 79.0%,and the prediction rate was34.8%;The interpretation rate of ammonia nitrogen（NH₄⁺-N）was 98.9%,and the prediction rate was 49.3%.It can be seen that the land use type and landscape pattern index of the basin affect the water quality of the basin;When using the model to analyze and predict the water quality indicators of the basin,not only the land use types of the basin should be considered,but also the landscape pattern indicators should be introduced.（5）The partial least squares regression analysis of the comprehensive influencing factors of water quality in the basin show that among the many basic characteristic indicators of the basin,the main influencing factors of water quality with high Variable importance for the projection（VIP>1）are the Interspersion and Juxtaposition Index（IJI）,Patch Cohesion Index（COHE）,Aggregation Index（AI）,Patch Density（PD）,and Edge Density（ED）;There is a significant negative regression relationship（Rc<0）with the permanganate index（CODmn）,five-day biochemical oxygen demand（BOD5）,total phosphorus（TP）,total nitrogen（TN）,and ammonia nitrogen（NH₄⁺-N）of the basin water quality pollution index.The influencing factors include terrain humidity index,terrain undulation,slope,and forest land volume.It shows that fragmentation degree,dispersion and aggregation characteristics,landscape connectivity,landform characteristics and ecological source area of basin landscape patches are the main influencing factors of basin water quality.（6）The connectivity index of landscape patches was used to identify ecological sources in the Danjiangkou Reservoir area,which have improved connectivity and need to be protected urgently;The priority of ecological corridor construction and protection is determined;Based on the analysis of basin ecological resistance with the minimum cumulative resistance model,the ecological corridor between various ecological sources was constructed;The ecological pinch points on the ecological corridor were identified.In the optimized ecological network of each sub-basin（sub-basin A to J）in the reservoir basin,the average value ofαindex describing the frequency of loop occurrence is 0.41,the average value ofβindex describing the complexity of connection is 1.80,and the average value ofγindex describing connectivity is 0.61.Among them,the sub-basin I withα,βandγindexes reaching the superior level accounts for 3.8%of the area;The sub-basin E,which accounts for 1.5%of the total area,has reached a good level;The sub-basins B,F,G and H all reach the medium level,accounting for 19.6%of the total area;The sub-basins A and D reach the general level,accounting for 34.6%of the total area;The sub-basins C and J are at a poor level,accounting for 40.5%of the total area.The basin area with sub-basinα,βandγindexes reaching the average or above level is 24.1×10⁴hm²,accounting for 59.5%of the total area,indicating that the optimization of basin landscape pattern can greatly improve the landscape connectivity of each sub-basin,and form a more complete landscape spatial pattern of basin water purification.