| As a key national infrastructure,railways play an important role in the construction of modern infrastructure systems.In the context of continuous development of economic construction,railway tunnel projects have been gradually extended to karst mountain areas,but as the mountain railway tunnel projects will traverse complex and variable karst geological formations in the process of implementation,and the tunnel site area is rich in underground dark river system,resulting in its construction process will produce great risk of sudden surge of water.The surface runoff formed by the water surging from the tunnel will cause some disturbance and damage to the landscape type and ecological environment along the route,causing a very serious negative impact on the ecological environment,so there is an urgent need to assess the ecological risk of the landscape generated by the water surging from the tunnel and the ecological network construction.Therefore,based on the theory of landscape ecology,this paper evaluates the landscape ecological risk of the area along the water surges in the mountain tunnel,and explores the degree of influence of the driving factors related to the water surges on the landscape ecological risk,and constructs an ecological network on this basis,which is important for improving the ecological pattern of the surrounding environment along the water surges in the mountain tunnel,improving the regional ecological environment risk and carrying out regional environmental management.This study takes the land use data of the area where the case mountain railway tunnel is located in 2000,2010 and 2020 as the basis,compares the parameters of the fitted model with the variation function of different buffer zones,selects the 1000 m buffer zone along the case tunnel gushing water as the study area,and analyses the spatial and temporal changes of its land use and landscape pattern;selects the landscape fragility index,landscape disturbance index and landscape The landscape vulnerability index,landscape disturbance index and landscape loss index were used to construct a landscape ecological risk index model to explore the dynamic changes of landscape ecological risk in the study area;the drivers of landscape ecological risk changes associated with water surges were selected and analysed using a geographic probe to clarify the extent to which different drivers explain the evolution of landscape ecological risk in the study area;the ecological safety pattern of the study area was constructed based on the minimum cumulative resistance model,and a rationalisation of the ecological safety pattern was proposed based on this model.The paper is based on a minimum cumulative resistance model,which is used as a basis for making rational recommendations.The main findings of this paper are as follows:(1)From the perspective of land use change characteristics: the main land use types in the area along the case tunnel surging water flow path are man-made surface and arable land,with the area of arable land decreasing from 360.36hm2 to 266.49hm2 from 2000 to 2020,and the area of man-made surface increasing from 90hm2 to 181.44hm2;the land use type with the greatest change in the single dynamic attitude is man-made surface,with The rate of increase reached 10.16%.In terms of landscape pattern change characteristics: the edge density,number of patches and patch density of arable land all show an increasing trend,and the landscape type with the greatest patch area change is man-made surface;the southwestern part of the study area is always a patch density low value area,the landscape shape index has a large increase in 2010,and shows a "beltlike" connectivity(2)From the dynamics of ecological risk in the landscape(2)From the perspective of dynamic changes in landscape ecological risk: the ecological risk of the landscape along the case tunnel surging water flow path showed an increasing trend between 2000 and 2020,the area of low ecological risk areas decreased,while the area of the rest of the ecological risk areas increased,the proportion of areas with an increasing ecological risk level was larger,medium ecological risk areas were scattered,higher and high ecological risk areas were mainly The global Moran’s I values for ecological risk in the study area are 0.080,0.074 and 0.057 in 2000,2010 and 2020 respectively,indicating a positive spatial correlation.The distribution of outliers is scattered.The distance from the tunnel entrance has the greatest influence on the dynamic change of landscape ecological risk,with an explanatory power of 18.301%,while the least influential driver is slope,with an explanatory power of 0.0609%,and the results of the interaction detector show that the interaction of any two drivers has different degrees of The results of the ecological detector show that the distance from the tunnel entrance and all indicator factors have a more significant difference on the spatial distribution of landscape ecological risk in the study area,and the results of the risk area detector show that the landscape ecological risk of all types of corresponding areas in the DEM and distance from the road are significantly different.(3)From the ecological network construction: the case tunnel gushing water runoff along the regional ecological source is selected from the highest value of ecosystem services land use type,namely arable land,the value of 1,462,307.91 million yuan;elevation,land use type and slope for weighted superposition to obtain a comprehensive resistance surface,on this basis the use of the minimum cumulative resistance model to establish the minimum cumulative resistance surface,the results show that The closer the area to the tunnel gushing outlet,the greater the integrated resistance value;the ecological corridor in the study area avoids the man-made surface and the tunnel outlet,while seven ecological nodes are identified,most of which fall on cultivated land.Finally,the ecological network is constructed and targeted suggestions are made to improve the surrounding ecological environment so that the landscape pattern can be optimised. |