Keystone species: toward an operational concept for marine biodiversity conservation

Abstract

Abstract. Various definitions and indices have been proposed in the literature to identify keystone species. In this study, we intended to make the concept of keystone species operational for marine biodiversity conservation. We used an exclusive definition of keystone species, based on the original concept of keystone predator, and derived a new functional index of keystoneness (KS) from an ecosystem-modeling approach. First, several KS indices were formulated, by combining measures of the mixed-trophic impact (MTI) and biomass of species. Then, a meta-analysis was performed, based on 101 published Ecopath food-web models, selected with a scoring method, and representative of the variety of marine ecosystems worldwide. The indices were applied to the models, and two statistical methods were compared to select the most promising KS index. Rank correlation tests were performed to assess the balance between the contribution of the impact and biomass components to the different KS indices. In addition, a classification tree was implemented, based on ecosystemspecific thresholds applied to the latter species traits, and used to confirm the identified keystone species. The selected index obtained the highest number of models with positive results from both the rank correlation tests and the classification tree. We also demonstrated the limitations of existing KS indices previously applied in the literature. Species were ranked according to their estimates of keystoneness with the selected KS index, so that potential keystone species were quantitatively identified in the 101 modeled food webs. The standardized modeling approach allowed for a comparison of the identified keystone species across models: cartilaginous fishes and toothed whales obtained the highest occurrences. Finally, the selected KS index was applied to the well-known case study of Prince William Sound (Alaska, USA). Potentially significant anthropogenic (fishing) impacts on keystone species were also considered and discussed. The operational methodology presented is directly applicable to marine food webs, and may be adapted to other (freshwater or terrestrial) systems.

Global overview of the applications of the Ecopath with Ecosim modeling approach using the EcoBase models repository

“…By applying a scoring method on these selection criteria, a pool of models of potential interest would be defined, and could then be reused to perform EwE-based meta-analyses. Such approach has already been implemented in the literature (e.g., Valls et al, 2015), and many more studies could exploit the wide range of information that can be derived from published EwE models, once made available in EcoBase.…”

Modeling Predator–Prey Linkages of Diadromous Fishes in an Estuarine Food Web

“…The KS 1 index is highly influenced by inputs from the mixed trophic impact analysis and identifies species of high biomass, whereas KS 2 assigns high keystoneness values to groups with low biomass and low overall effect, which can be considered rare (Power and Mills 1995;Valls et al 2015). Due to limitations in comparison of these functional indices across models, we calculated an additional keystone index (KS 3 ) that was developed by Valls et al (2015) from a metaanalysis of 101 Ecopath models.…”

“…Due to limitations in comparison of these functional indices across models, we calculated an additional keystone index (KS 3 ) that was developed by Valls et al (2015) from a metaanalysis of 101 Ecopath models. The KS 3 index is calculated by using model outputs and highlights species that have a greater balance between their trophic impacts and biomass contributions (Valls et al 2015):…”

“…where the relative total impact (ε i ) is multiplied by the biomass contribution (B i ) ranked in descending order for each species group (Valls et al 2015).…”

“…and benthic FIGURE 7. Plots of keystone index 1 (KS 1 ), keystone index 2 (KS 2 ; e.g., Libralato et al 2006), and keystone index 3 (KS 3 ; e.g., Valls et al 2015) versus the relative total impact of each functional group (abbreviations are defined in Figure 2). Groups that are oriented toward the top right of a plot play a greater role as keystone species.…”