Coupling Relationship Of Marsh Wetlands Patterns-Characteristics-Functions In The Qilian Mountains Of Qinghai,China

Alpine marsh wetland is a unique ecosystem type formed by long-term adaptation to the alpine climate environment.It plays a vital role in regulating river water sources,maintaining biodiversity on the plateau,and safeguarding alpine livestock production.The marsh wetland ecosystem is a composite whole whose landscape,structure,function,and various ecological processes are coupled and reflect its ecosystem’s wholeness.Currently,most studies on marsh wetlands focus on the response to environmental changes at a single level,such as area,landscape pattern,characteristics,or functions,but lack comprehensive studies on the changes among various levels of marsh wetlands.Therefore,it is urgent to comprehensively explore the spatial and temporal patterns of ecosystem patterns,characteristics and functions of alpine marsh wetlands,the driving factors,and the coupling relationship between the levels,which is of great significance to the ecological protection,scientific management,and effective restoration of alpine marsh wetlands.Based on the research framework of patterns-characteristics-functions of the alpine marsh wetlands ecosystem,this study focused on the alpine marsh wetland in Qilian Mountain of Qinghai（QMQ）,constructed a series of environmental indicators including climate,vegetation,soil,topography,and human activity factors,simulated the spatial distribution pattern of alpine marsh wetland in the study area,and clarified the main driving factors affecting the formation and distribution of the marsh wetland.Then,combined with the field sample survey,this study explored the relationship of marsh wetlands on hydrological-vegetative-soil characteristics under different degradation degrees and utilized the model to quantitatively assess the pattern of spatial-temporal change in the ecosystem service function of marsh wetlands as well as influencing factors.Finally,the coupling relationship between ecological patterns,characteristics,and functions of alpine marsh wetlands was revealed by the regression modeling,and adaptive suggestions and management strategies were proposed to provide a scientific foundation and theoretical guidance for the protection,restoration,and sustainable development of marsh wetland resources in the region.The main results of the study are as follows:（1）Spatial distribution patterns of alpine marsh wetlands and the drivers of their formative distribution in QMQThe overall accuracy of the spatial distribution of marsh wetlands in QMQ simulated by the marsh-producing model was 89.89%,and the Kappa coefficient was0.87.The total area of marsh wetlands in 2020 was 2778.24 km²,which was mainly distributed in the headwaters area of the watershed,showing patchy distribution according to the rivers.Normalized difference water index（NDWI）,fraction vegetation coverage（FVC）,annual minimum temperature（AMNT）,normalized difference vegetation index（NDVI）,and digital elevation model（DEM）were the most critical factors in marsh wetland simulation,with the relative importance of 11.94%,10.19%,8.06%,7.00%,and 6.63%,respectively.The topography acted on the marsh wetlands’distribution by influencing the nutrient input function of potential marsh wetlands（the indirect effect was 0.45）.The climate was the fundamental factor that directly controlled the distribution of marsh wetlands（the direct effect was 0.43）,and vegetation had a total effect of 0.23.Climate and topographic factors dominated the formation and distribution of marsh wetlands in QMQ.It cannot be ignored that the impact of human activities on the marshes even exceeded the influence of soil factors in QMQ.（2）Characteristic changes in ecohydrology-plant communities-soil physicochemical properties and their relationships of alpine marsh wetlands in QMQAs the degradation degree of the marsh wetlands increased,there were significant differences in precipitation（PRE）and intermound water（Hwater）of hydrologic characteristics.The plant height,cover,density,aboveground biomass（AGB）,soil moisture content（SMC）,and nutrients showed an obvious decreasing trend.The Shannon-Wiener diversity index（H’）,the Pielou evenness index（E）,and the Simpson dominance index（D）showed an increasing trend.However,soil total potassium（TK）showed an apparent enrichment phenomenon.The relationships between hydrologic-vegetation-soil characteristics of marsh wetlands were slightly different at different degradation levels,which were mainly related to water stress in marsh wetlands.PRE significantly affected wetland plant communities at the non-degraded and lightly degraded levels,while TK and SMC significantly affected wetland plant communities at the moderately degraded level.（3）Spatial and temporal characteristics of ecosystem service functions and their drivers of alpine marsh wetlands in QMQBased on the coupled In VEST-Random Forest（RF）-Geographically and Temporally Weighted Regression models（GTWR）,it was calculated that the water yield of the marsh wetlands in QMQ increased by 67.46%from 2000 to 2020.The soil conservation and habitat quality service functions showed a decreasing trend,which decreased by 65.08%and 30%,respectively.Three ecosystem service function values showed a pattern of high in the south and low in the north.The water yield service function was primarily driven by PRE（25.86%）,the soil conservation function was most significantly affected by NDVI（20.94%）,and the disturbance of habitat quality services was exacerbated by human activities（15.79%）.There was significant spatial heterogeneity in the drivers of the various ecosystem service functions,showing a northwestern-southeastern strip distribution.（4）The coupling relationship between patterns-characteristics-functions of marsh wetlands in QMQThe moisture-related PRE（x₄）,surface runoff depth（Runoff,x₅）,SMC(x₁₀),and soil cation exchange capacity(CEC,x₁₃)were the decisive characteristic factors affecting the water yield service function（y₁）of marsh wetlands in QMQ.The Multiple Linear Stepwise Regression model（MLSR,R²=0.720）was able to fit the relationship between water yield service function and patterns-characteristics of marsh wetlands(y₁=-73.457+0.712x₄-0.671x₅+0.638x₁₀+0.638x₁₃).Runoff（x₅）and AGB（x₉）were important characterization factors affecting the soil conservation function（y₂）of marsh wetlands,and MLSR（R²=0.584）was the optimal model fitting the relationship between soil conservation function and patterns-characteristics of marsh wetlands in QMQ（y₂=-247.985+1.421x₅-0.438x₉）.The habitat quality function of marsh wetlands was significantly and positively correlated with TK in QMQ,but none of the three models,MLSR,Partial Least Squares Regression（PLSR）,and Weighted Least Squares Regression（WLSR）,could better reflect the relationship between habitat quality function and patterns-characteristics.The formation and distribution of marsh wetlands in the region were prerequisites for influencing their ecosystem characteristics and functions,and the deterioration of hydrological conditions was the main cause of the degradation of marsh wetlands,which directly affected the characteristics of wetland plant communities and soil nutrients,leading to changes in ecosystem service functions.（5）Adaptive strategies of ecological conservation and management for alpine marsh wetlands in QMQBased on the ecological source,ecological corridor,and ecological node paradigm,this study constructed an ecological resistance evaluation system by applying the coupling relationship of patterns,characteristics,and functions.The key areas for the ecological protection of marsh wetlands were identified in QMQ,with a total area of996.53 km²,including 40 ecological sources,39 ecological corridors,and 16 ecological nodes.Regarding the two types of disturbances（grazing and mineral exploitation）that degrade the marsh wetlands in QMQ,control measures of ecological protection for marsh wetlands were proposed according to local conditions,mainly focusing on natural restoration,adjustment of grazing strategies,implementation of mines ecological restoration projects,and delineation of protected areas.