Identifying thresholds for ecosystem-based management

Abstract

Background: One of the greatest obstacles to moving ecosystem-based management (EBM) from concept to practice is the lack of a systematic approach to defining ecosystem-level decision criteria, or reference points that trigger management action.

Methodology/principal findings: To assist resource managers and policymakers in developing EBM decision criteria, we introduce a quantitative, transferable method for identifying utility thresholds. A utility threshold is the level of human-induced pressure (e.g., pollution) at which small changes produce substantial improvements toward the EBM goal of protecting an ecosystem's structural (e.g., diversity) and functional (e.g., resilience) attributes. The analytical approach is based on the detection of nonlinearities in relationships between ecosystem attributes and pressures. We illustrate the method with a hypothetical case study of (1) fishing and (2) nearshore habitat pressure using an empirically-validated marine ecosystem model for British Columbia, Canada, and derive numerical threshold values in terms of the density of two empirically-tractable indicator groups, sablefish and jellyfish. We also describe how to incorporate uncertainty into the estimation of utility thresholds and highlight their value in the context of understanding EBM trade-offs.

Conclusions/significance: For any policy scenario, an understanding of utility thresholds provides insight into the amount and type of management intervention required to make significant progress toward improved ecosystem structure and function. The approach outlined in this paper can be applied in the context of single or multiple human-induced pressures, to any marine, freshwater, or terrestrial ecosystem, and should facilitate more effective management.

Figure 1. Relationships between hypothetical ecosystem attributes and anthropogenic pressures.

Attribute values range from unstressed to stressed (sensu [13]), and the levels of the pressures applied have been scaled relative to a theoretical maximum. A utility threshold cannot be defined objectively for the linear model (a), but can be defined objectively for the two piecewise models (b and c) and the sigmoidal model (d). Equations for the models and the location of the utility thresholds are described in Text S1. In (b-d), the threshold pressure is indicated by the dashed lines.

Figure 2. Model-generated relationships between 4 ecosystem attributes and increasing ecosystem-wide fishing (a-d) or nearshore habitat (e-h) pressure.

The ecosystem attributes are resilience, NPP/Biomass, Shannon diversity, and mean trophic level. Open triangles indicate median values calculated from Monte Carlo simulated Ecopath with Ecosim data (n = 100), and error bars denote 95% confidence intervals. The solid lines represent best-fit functional relationships and the dotted lines designate significant utility thresholds estimated using a nonparametric bootstrap resampling procedure (n = 10,000 for each Monte Carlo data set) (parameter values and significant utility thresholds listed in Table 2). NPP = net primary production. In this and following figures, the ecosystem attributes (y-axes) have been re-scaled so that larger values are considered unstressed rather than stressed. The pressure values have been re-scaled relative to the maximum simulated pressure, and are contained within the range [0, 1].

Figure 3. Correlations between indicators and attributes with significant utility thresholds, along with indicator-pressure relationships.

The indicators are sablefish and jellyfish biomass for the fishing (a-d) and nearshore habitat (e-g) pressure simulations, respectively. Open circles (a-c, e-f) and triangles (d, g) indicate median values calculated from Monte Carlo simulated Ecopath with Ecosim data (n = 100 for each pressure level), and error bars denote 95% confidence intervals. The solid lines in (d) and (g) represent best-fit functional relationships estimated using a nonparametric bootstrap resampling procedure (n = 10,000 for each Monte Carlo data set) (parameter values listed in Table 3). r_s = median spearman rank correlation across the Monte Carlo data sets (bold indicates 95% CI did not overlap zero), NPP = net primary production.

Figure 4. Spider plots depicting trade-offs among four ecosystem attributes and two fisheries yield metrics.

The three different fishing (a-c) and nearshore habitat (d-f) pressure levels corresponded approximately to a minimum-impact scenario in which none of the utility thresholds were breached (a, d), a threshold scenario in which the simulated pressure matched that of the lowest utility threshold (median value; see Table 2) (b, e), and a maximum-impact scenario representing the maximum pressure considered (c, f). Note that for each type of pressure, all attributes have been re-scaled so that values are relative and fall within the interval [0,1], where zero corresponds to a stressed condition and one corresponds to an unstressed condition.