Pollination is an important type of ubiquitous ecological interactions in terrestrial ecosystems that form pollination networks. The interaction between pollinators and plants will form two-level dependence network, and the extinction of animals in pollination networks will cause the extinction of plants that depend on them. The cascade effects caused by species extinction may change the functional niche of dependent species. At present, there are many quantitative indicators describing the functional niche of species in network research, but they focus on different aspects, and it is not clear which indicator can best measure the functional niche. In this paper, using 45 pollination network data, simulating extinction cascade effects by using Topological Coextinction Model (TCM) and Stochastic Coextinction Model (SCM), from the most generalization to the most specialization, one pollinator is extincted at a time, generate a standard ranking distribution curve of the indicator values of the plant species before and after extinction ( NRAD), and then calculate the Manhattan distance between t the distribution curves of the functional roles of plant species before and after extinction. Based on this distance, analysis of variance and cluster analysis were carried out to identify the differences of five kinds of indicators (degree, closeness centrality, betweenness centrality, c- scores and z- scores) to evaluate the effects of extinction cascade on species in pollination networks.
Results/Conclusions
The results show that under the two models simulating a species extinction scenario, the average distances of plants’ z- scores NRAD before and after extinction is the largest, but there is no significant difference with the results of betweenness centrality index. z- scores and betweenness centrality were significantly better than other indicators, indicating that these two indicators were the best indicators reflecting the impact of extinction cascade on the functional niche of species. In the case of single species extinction, the selection of extinction model has little effect on the selection of indicators. Cluster analysis showed that z- score and betweenness centrality were divided into different groups. This paper argues that the selection of z-score and betweenness centrality can better reflect the impact of extinction cascade on the functional niche of species in poliination networks. This study provides a new method for assessing the species roles, which can be used to identify keystone species in ecosystem and provide a basis for biodiversity conservation.