Temporal-Spatial Features Of Landscape Change, Disturbance Simulation And Ecological Security Early Warning In World Mixed Heritage Site Wuyishan Scenery District

A world Heritage Site is a place recognized by the World Heritage Committee of the UNESCO as akind of cultural and natural wealth of universal value, infrequency and irreplaceability.. For a long time,world heritages sites suffer serious damages or even devastation from a variety of human and naturalfactors, for example, earthquake, hurricane, tsunami, war, urban construction, tourism development andenvironmental destruction. Therefore, it is urgent to provide more protection for world heritage sites.Located in the northwestern part of Fujian Province, Wuyi Mountain is the most outstanding area forbiodiversity conservation in southeast China and a refuge for a large number of ancient, relict species,many of them endemic to China. In December1999, Wuyi Mountain was included on the World Naturaland Cultural Heritage List by the23rd Session of the World Heritage Committee of the UNESCO, beinggiven the comment "a natural landscape so unique, rare and marvelous, is the beauty of nature andembodiment of the harmonious relations between human being and environment". Wuyi Mountain is nowthe largest of China' s World Heritage sites which covers a total area of999.75square kilometers. Thenominated area is divided into4parts, with the Biodiversity Preserve to the west, Nine-twist StreamEcological Preserve in the center, Natural Beauty and Cultural Landscape Preserve to the east (i.e.Wuyishan Scenery District) and the ancient Chencun Minyue Relics to the southeast.With the tourism of world heritage sites developing rapidly, human beings are producing more andmore disturbances to the nature, society, economy and local culture in these sites. As a world-renowedmountain, Wuyi Mountain has attracted numerous domestic and foreign tourists for its Danxia landform,natural landscape, historically and culturally famous mountains. However, at the same time, the compositeecosystem of nature, society and economy in Wuyi Mountain is facing tremendous pressure, which cast aheavy burden on the protection of world heritage. Therefore,, a study on ecological security status and itschange mechanism and trend in Wuyi Mountain will have great significance to the protection and rationalutilization of the cultural and natural mixed heritage sites.Landscape ecology is considered as an emerging interdiscipline between ecology and geography,focusing on ecological issues on large spatial scale and temporal scale. Landscape ecology theories providenew idea and approaches to ecological security research on world heritage sites. Wuyishan Scenery Districtis chosen as a study region in the paper, as have been disturbed most severely compared to the other threesub-regions, although in general all of the four regions have been well protected under strict managementmeasures required for protecting a world heritage site. Utilizing3S technology and combining with basicdata, remote sensing image, social and economic material and previous research results, we conduct aseries of research in the three critical periods of1886,1997and2009on the basis of the theories andmethods on landscape ecology and ecological security. An analysis is made for the law of change oflandscape pattern, and an explanation of main interference factors and driving mechanism is given.Canonical correspondence method is used to analyze a multi-scale effect between landscape pattern andenvironmental factors. A risk indicators system is constructed to identify risk pattern of main naturalsources including forest fires and dendrolimus punetatus walker damage. Spatial statistics method isadopted to deal with temporal-spatial differentiation and change rules of landscape ecological security degree. Step regression is used to analyze characteristics of temporal-spatial change of ecosystem servicesvalue and its responses to environmental factors. Further, a composite ecological security index is firstestablished, and the simulation of nonlinear relationship between this index and driving factors performswell by artificial neural network (ANN) method. Compared with the currently main models, a CLUE-Smodel is chose to simulate and forecast the future change of landscape pattern. In addition, the paper putsforward a concept of ecological security on cultural and natural heritage sites and then constructs anearly-warning frame and system index of ecological security in mixed sites; an extension analysis methodis introduced to give an analysis of ecological security early warning in Wuyishan Scenery District. At theend, an ordering rule for protection and management in world heritage sites is discussed on the basis ofdegree of loss (DL) for the first time. Primary results are as follows:(1) Landscape types in the study area are divided into eleven categories including Pinus massonianaforest, Camellia sinensis plantation, farmland, bare area, construction land, Cunninghamia lanceolataforest, broad-leaved forest, shrub and grassland, bamboo forest, economic forest, and river. The mainlandscape types are Pinus massoniana forest (matrix landscape), Camellia sinensis plantation, farmland,construction land and river. Landscape pattern from1986to2009changed significantly with thecontinuous increase in the area of Camellia sinensis plantation and construction land, while the large-scaledecrease in the area of Pinus massoniana forest, farmland, and bare area. With the fragmentation of Pinusmassoniana forest and farmland, the construction land became more regular in a patch shape. Significantdisturbances took place in Camellia sinensis plantation while the river landscape maintained stable. Thedifferent characteristic of the change of landscape pattern occurred in different development periods (1986-1997and1997-2009).(2) Total lengths of the scenic corridors in1986,1997and2009were94.218km,156.725km and197.574km, respectively. As a major change in elements of scenic corridor, the roads increased by64.659km from1986to1997. From1997to2009, the length of roads had only a increase by22.172km while thatof trails by18.687km. Between1986and2009, the construction ratio of roads vary from0.31to0.60; thedensity of corridors increased from1.34km/km2to2.81km/km2; curvature increased from1.19to1.56.The ratio of line to node, circuitry and connectivity had an increasing trend during the period of1986to2009, which leaded to a more complicated network structure, especially at the mountain's north scenicregion,. The three above indices of this region were1.387,0.494,0.228in2009, respectively, all higherthan that of other scenic regions. Visitor trails, curvature, density, and construction ratio of roads weresignificantly associated with garbage quantity, Shannon-Wiener diversity index, plant Simpson diversityindex, plant evenness index and landscape important value. The network structure indicators were notcorrelated with the ecological environment indicators.(3) In terms of unit area of ecosystem services value in study area, river bears a maximum of93707RMB/hm2a and bare area bears a minimum of354RMB/hm2a. Although ecosystem services valueachieved a certain degree offset from1997to2009, it still remained a deficit trend in the period of1986to2009. This is due to that vegetation landscapes were transformed to construction land in that time.Ecosystem services value has a medium correlation level with environmental factors. Correlation coefficient between canopy density and service value is the maximum0.7**, and the one between canopydensity and volume was the maximum (0.86**) in various environmental factors. The fitting results ofvegetation landscape types of ecosystem services value(R~2=0.7524, p＜0.0001) has an advantages overnon-vegetation one (R~2=0.5370, p＜0.0001).(4) Multi-scale effect between landscape pattern and environmental factors, and the response law ofenvironment factors to ordination axes could be scaled within six sampling scales(30m~450m). Theaverage correlation coefficients of environment factors with landscape pattern in a descending order werecanopy density (0.7989**), humus thickness (0.7248**), elevation (0.6083**), slope (0.6132**), longitude(-0.4064**), soil thickness(0.3696**), latitude (0.0614), aspect(0.0295). The canopy density is mostcorrelated with landscape pattern while the slope is least correlated. According to the relationship betweenthe landscape types and environmental factors, landscapes in the area can be classified into three categories:the elevation related landscapes including bare land, rivers, construction land, and farmland,; the soilrelated landscape including Cunninghamia lanceolata forest, Pinus massoniana forest, broad-leaved forest;and human managed landscape including Camellia sinensis plantation, economic forest, bamboo forest,shrub and grassland.(5) The main reasons causing the landscape succession in Wuyishan Scenery District came from fouraspects: the restrictive function of the natural environment, the behavior pattern change driven by theeconomy profit, development construction caused by the tourism development and population growth, andthe orientation function of laws and regulations and management. In order to simulate the landscape change,both natural driving factors (altitude, slope, slope etc.) and human driving factors (the closest distance toroads, to river, to residential areas, to the parking areas) are selected as quantitative driving factors. Underthree scenarios, landscape pattern in2020is simulated by CLUE-S model to forecast the change ofsensitive landscape types.(6) During the period of1986to2009, the ecological security degree displayed a gradual increase inWuyishan Scenery District. Morans's I of ecological security degree showed a certain degree positivecorrelation, and this correlation was most distinct between1986and1997. General auto correlation oflandscape ecological security degree had a scale response. In the period of1997to2009, localautocorrelation pattern was consistent. However, the cluster structure and significant level of ecologicalsecurity degree presented obvious change between1986and1997. The strong spatial correlation forecological security degree was the result of combined influence from structural factors and non-structuralfactors. Structural factors such as topographic and landform and agrotype played a conclusive role in itsspace distribution. And non-structural factors including tourism development and construction,deforestation and shifting farmland to plant tea, had an important influence on its evolution.(7)There is little possibility of forest fire, and forest fire results mainly from human activities. Risklevel of forest fire the this area is at better safety grade and general safety grade. The area of both gradesaccounts for98.79%of the scenic forests. Both the South Stream Scenic Zone and the Travel Service Zoneare the focus of fire prevention work. Besides, there are1393.83hm2of Pinus massoniana forest exposes inhigh-risk area, accounting for34.26%of the total Pinus massoniana forest area. The high-risk areas basically covers the essence scenic spots in the district., And the risk on the north bank of the JiuquxiStream is higher than that of the south bank. It is very important to implement forest landscape pest controlin both sides of Jiuquxi Stream in Wuyishan Scenery District(8) Composite ecological security index better explains the ecological security spatial pattern.Prediction92.04%of accuracy can be achieved by artificial neural network (ANN) method to simulate thenonlinear relationship between this index and driving factors, with high precision and good effect.Meanwhile, It gives a good explanation of mechanism driving factors affecting ecological security pattern.(9) The accuracy result of extension analysis method introduced to perform ecological security earlywarning is84%. Early warming rank is at a Ⅰ level (no alarm status) in both the year of1997and2009.Nonetheless, the rank in2009is easier than the one in1997to transform to the Ⅱ rank that has a more risk.At the same time, early-warning indices belong to sensitive ranks have some difference between1997and2009.The study not only agrees with hot topics and trend in the field of ecological security, but also meetsthe need of current world heritage protection. Meanwhile, this study provides certain reference value andpractical significance in the aspects of developing ecological security research objects and enriching worldheritage research content.