Co-adaptation enhances the resilience of mutualistic networks

Abstract

Mutualistic networks, which describe the ecological interactions between multiple types of species such as plants and pollinators, play a paramount role in the generation of Earth’s biodiversity. The resilience of a mutualistic network denotes its ability to retain basic functionality when errors and failures threaten the persistence of the community. Under the disturbances of mass extinctions and human-induced disasters, it is crucial to understand how mutualistic networks respond to changes, which enables the system to increase resilience and tolerate further damages. Despite recent advances in the modelling of the structure-based adaptation, we lack mathematical and computational models to describe and capture the co-adaptation between the structure and dynamics of mutualistic networks. In this paper, we incorporate dynamic features into the adaptation of structure and propose a co-adaptation model that drastically enhances the resilience of non-adaptive and structure-based adaptation models. Surprisingly, the reason for the enhancement is that the co-adaptation mechanism simultaneously increases the heterogeneity of the mutualistic network significantly without changing its connectance. Owing to the broad applications of mutualistic networks, our findings offer new ways to design mechanisms that enhance the resilience of many other systems, such as smart infrastructures and social–economical systems.

Advanced progress of network resilience

“…Network resilience analysis have been applied to real networks. For example, in the ecological system, Zhang et al [72] proposed a collaborative adaptation model by incorporating dynamic features into structural adaptation. The results indicated that the network resilience could be enhanced when considering the adaptivity physics.…”

Effect of Chitosan on Induced Skin Wound Healing in Rabbits: Histopathological Evaluation

“…Due to the vast range of qualities of these compounds, they are frequently employed to treat many disorders, particularly those of the skin and soft tissues. (Zhang N. et al, 2020;He et al, 2021). Numerous investigations revealed that chitosan had the ability to heal wounds.…”

Effect of network structure on the accuracy of resilience dimension reduction

“…Gao et al [1] proposed the dimension reduction method that decomposes N-dimensional networks into one-dimensional effective models and uses it to predict the global activity of the original network. Subsequently, based on the theoretical tools for large-scale networks and the advanced data analyzing techniques [10][11][12][13], the method has been extended to many aspects such as noise effects [14][15][16], reduction methods based on spectral dimension [17], sequence mean field [18] and degree weighted average [19], and has been applied to various fields [20][21][22]. The dynamic characteristics of a system strongly depend on the underlying network structure [17].…”