| In recent years,the intensification of global human activities has profoundly affected the evolution of ecosystems in the whole world as well as various regions at different scales.Human activities first affect the regional landscape patterns.As a result,the regional net primary productivity(NPP)of the vegetation is going to be changed.However,under the influence of human activities,spatio-temporal variation,spatial correlation and scale dependence of the regional landscape pattern and NPP are currently unresolved scientific issues.Therefore,the current study is conducted under the theoretical guidance of landscape ecology and sustainable development to overcome these issues.Taking Shule River basin as an example which is located in northwestern China,spatio-temporal variation characteristics,spatial correlation and scale dependence of the regional landscape pattern and NPP of watershed scale under the influence of human activities were studied.Based on the land use information extracted from Landsat TM images from 2000 to 2015 by using Fragstats software to calculate the landscape metrics,the impact of the intensity of human activity on the landscape metrics to was discussed.Secondly,based on the CASA model,the meteorological data,MODIS data and vegetation type data were used to analyze the net primary productivity of the basin in 2001-2015,and its spatial-temporal dynamics and the influence of landscape pattern on NPP.Finally,based on the above calculated data,the grid analysis and regression analysis were used to quantitatively analyze the multi-scale landscape pattern and NPP spatial correlation in the basin from 2000 to 2015.Taking 2015 as an example,the most suitable scale is selected.The NPP multiple regression model based on landscape metrics is established in the whole basin and different landscape types to provide scientific basis and technical support for the layout optimization of the basin landscape pattern and the sustainable and healthy development of the regional ecological environment.1.Spatio-temporal variation characteristics of NPP and influence of landscape pattern on NPP in Shule River Basin(1)By analyzing the dynamic evolution and fragmentation of the landscape pattern in the Shule River basin from 2000 to 2015,it is found that the area of farm land and urban land in the oasis area of the middle and lower reaches has largely increased,and the landscape types are diverse and the most fragmented.(2)The spatial distribution pattern of human activity intensity(HAILS)in different years is relatively consistent.The high value areas are mainly located at the middle and lower reaches of strip-shaped oases area in Shule River basin and the HAILS increased most significantly over time.(3)The HAILS and the landscape fragmentation index are positively correlated in space.That is,the higher the intensity of human activity,the more complex and diverse the number,shape and structure of landscape patches.It shows that the intensity of human activity affects the landscape fragmentation to a certain extent.2.Spatio-temporal variation characteristics of NPP and influence of landscape pattern on NPP in Shule River Basin(1)In general,NPP showed a slow increasing trend from 2001 to 2015 in the basin.And after 2010,the NPP showed a very significant upward trend.The multi-year average NPP spatial heterogeneity of the basin is significant.The spatial distribution of mean NPP value has a significant difference during 2001-2015 over Shule River basin.The NPP in the upstream and middle reaches of oasis areas is the largest,and the NPP in the desert is the smallest.The difference in spatial distribution of NPP between different seasons in the basin is obvious in 2015.Among them,the spatial heterogeneity of NPP in summer is the most significant,and the spatial heterogeneity of NPP in winter basin is the smallest.(2)The type of landscape changed significantly in the basin.In general,the change of landscape type promoted the increase of total NPP,and the change of farm land area contributed the most to the change in total NPP during 2000-2015.3.Spatial correlation and its scale effect between landscape pattern and NPP(1)In current study we divided the watershed into eight different scale grids.The correlation between 20 landscape metrics and NPP on different grid scales was analyzed in 2000,2005,2010 and 2015.In different years and grid scales,landscape metrics and the NPP showed different degrees of positive or negative correlation,and there are more positive correlation indexes,but the Pearson correlation coefficient had a more consistent trend.(2)On the time scale,the correlation between landscape metrics and NPP at different grid scales tends to increase with time.The correlation between the landscape metrics characterizing the number,density and diversity of landscape patches and NPP increases with significant incline.(3)On the spatial scale,as the grid scale increases,the trend of landscape metrics and NPP correlation is different.The correlation between the metrics characterizing the number,density and diversity of landscape patches and NPP increases with the increase of grids scale.The correlation between the metrics characterizing aggregation of landscape patches and NPP decreases with the increase of grids scale.It shows that there is a significant spatial scale effect on the correlation between the landscape metrics and NPP of the Shule River Basin.4.Multiple linear regression models of NPP with landscape metrics in Shule River Basin(1)Taking 2015 as an example,the Pearson correlation coefficient of the 30 × 30 km grid scale and its significance are higher than other scales,which can be determined as the most suitable grid scale in the basin.(2)On the most suitable grid scale,the PD,SHDI,PR,LPI,DIVISION,and AI indices highly correlated with NPP are selected to establish multiple linear regression model of NPP with landscape metrics.The landscape metrics explains 62.7% of NPP variations,indicating that variations in these landscape metrics have a significant impact on NPP.(3)On the most suitable grid scale,the landscape metrics with higher correlation with NPP are selected to establish the NPP multiple regression model in different landscape types.However,the parameters of the model and the results of the simulation are different.In summary,this study revealed the spatio-temporal variation,spatial correlation and scale dependence of the regional landscape pattern and NPP under the influence of human activities,and constructed multiple linear regression models of NPP with landscape metrics in Shule River Basin.It provides a new approach for understanding the impact of landscape pattern variations on NPP.The research findings not only fill the gaps in the study related to watershed landscape ecology,but also can be used to guide landscape pattern optimization and ecological environment management and monitoring the most potential application value. |
PDF Full Text Request