Identifying Keystone Species Based On Centrality Of Complex Networks

At present,the Earth's ecosystem is facing prominent problems such as accelerated species extinction and sharp decline in biodiversity.Some scholars even believe that the Earth is entering the sixth era of extinction of biological species.What is more noteworthy is that we humans may also be extinct.Biodiversity plays an important role in maintaining ecological balance and protecting regional environment.It is the laterial basis for human survival.The food web is a complex network,and the extinction of a single species is likely to be infinitely magnified along this network.The extinction of an initial species affects other species through the complex interrelationships of species in the food web,and even leads to a series of secondary extinctions that further magnify the decline in biodiversity.Loss of keystones species will result in greater than average changes in the development of other communities or ecosystems of ecological species.A very small change in the keystones itself can also have a major impact on the community or ecosystem processes,and reloval of such species is likely to trigger extinction of other species,other populations or functional processes in the ecosystem.Therefore,the keystones to protection is to protect more other species.Identification of keystone species has a guiding reference for the priority protection of the calibrated species and the conservation of biodiversity.This study collected and analyzed food web data containing carbon flux between species,and selected 20 food webs as the research object.This study was based on the binary network and weighted network of 20 food webs.We compared the betweenness centrality,closeness centrality,Page Rank centrality and traditional degree centrality of complex networks from two aspects.On the one hand,we used topological method to study the extinction caused by species removal of specific sequences,the importance of different species in the structure of food webs and judging the effectiveness of three new centrality indicators in identifying keystone species and comparing traditional centrality with the new indicators.On the other hand,we divided the food webs into core-periphery structure based on different centrality indicators,and verified the feasibility of the partitioning method by sequence removal.We got four major results:(1)The betweenness centrality,closeness centrality and Page Rank centrality all can be used as effective indicators in identifying keystone species in both binary networks and weighted networks.(2)Iin the binary network,the degree centrality is better in identifying keystone species,and the three new indicators measure the importance of species better in the weighted network.(3)Closeness centrality and Page Rank centrality can be used to divide the food webs into core-periphery structure in both binary networks and weighted networks.In general,it is not recommended to use betweenness centrality for division.(4)There are some differences between the core and periphery of the network when the structure is divided based on different indicators.However,since there is no significant difference between the core species in different divisions,we can not judge the advantages and disadvantages of divisions based on different indicators and the traditional centrality,it can be further validated in real food webs.This study analyzed the application of complex network centrality in identification of species importance in food webs from two different scales.We used topological method to verify the keystone species of betweenness centrality,closeness centrality and Page Rank centrality,and found that they could be used as indicators to measure the importance of species,and the recognition effect in weighted networks was better than degree centrality.If an ecosystem 1s a functional unit and has only one key species,it is intuitive to show that the network collapses easily when the species is removed.So for a strong stable ecosystem,it should have several keystone species,so that one species can be compensated by other keystone species after loss.Therefore,the division of the network core-periphery structure is very important.The research in this paper provided a new idea for the subsequent research through theoretical discussion.Most of the current research focuses on the identification of keystone species in the binary network.This paper demonstrates the indicator recognition effect in the weighted network to protect the ecological diversity.Sexuality provides theoretical guidance and is important for protecting the diversity and stability of ecological species.