Climate and fishing steer ecosystem regeneration to uncertain economic futures

Abstract

Overfishing of large predatory fish populations has resulted in lasting restructurings of entire marine food webs worldwide, with serious socio-economic consequences. Fortunately, some degraded ecosystems show signs of recovery. A key challenge for ecosystem management is to anticipate the degree to which recovery is possible. By applying a statistical food-web model, using the Baltic Sea as a case study, we show that under current temperature and salinity conditions, complete recovery of this heavily altered ecosystem will be impossible. Instead, the ecosystem regenerates towards a new ecological baseline. This new baseline is characterized by lower and more variable biomass of cod, the commercially most important fish stock in the Baltic Sea, even under very low exploitation pressure. Furthermore, a socio-economic assessment shows that this signal is amplified at the level of societal costs, owing to increased uncertainty in biomass and reduced consumer surplus. Specifically, the combined economic losses amount to approximately 120 million € per year, which equals half of today's maximum economic yield for the Baltic cod fishery. Our analyses suggest that shifts in ecological and economic baselines can lead to higher economic uncertainty and costs for exploited ecosystems, in particular, under climate change.

Keywords: Baltic Sea; cod; ecosystem-based management; food-web dynamics; regime shifts; shifting baseline.

Figure 1.

Regime changes in the Baltic Sea ecosystem. Demarcation between regimes is indicated by red dotted line and coloured background for past (grey) and current (red) regimes. Changing ecosystem structure based on the leading mode (PC1) of biotic data spanning three trophic levels and regime indicator (grey) (a), changes in the fish community from cod (black circles) to sprat (blue triangles) (b), exploitation history of cod given in terms of fishing mortality (c), and changes in hydroclimatic regime indicated by spring temperature (black) and salinity (green) (d).

Figure 2.

Study design to test for ecosystem regeneration pathways. The schematic describes the four steps used in our study.

Figure 3.

Individual trophic-level statistical models. Each row displays partial plots showing the main biotic and abiotic effects on cod (a–c), sprat (d–f), herring (g–i) and Pseudocalanus acuspes (j–l). Partial plots based on models without threshold effects are shown in dark blue, while non-additive interactions above and below thresholds (i,j) are shown in red and black, respectively. Associated thresholds (θ) are indicated by horizontal lines (i,l). For sprat the sea surface temperature in summer (d) and for P. acuspes the spring mid-water salinity (k) are shown.

Figure 4.

New ecosystem baseline with a lower stability. Response of cod (a), sprat (b) and Pseudocalanus acuspes (c) and the overall coefficient of ecosystem variability (d) to increased and subsequently decreased cod exploitation level (fishing mortality) under past (black) and current (red) conditions of temperature and salinity. Schematic of the variability in the strength of the prey-to-predator loop (including cod, top; sprat, middle and P. acuspes, bottom) leading to different regeneration pathways and lower baseline biomasses (indicated by sizes of squares as well as percentage changes) under past (black) and current (red) hydroclimatic conditions (e).

Figure 5.

Economic profits and societal costs. Sustainable economic yields (a) and corresponding CS and RP (b) at different cod exploitation levels (fishing mortality) under past (black) and current (red) conditions of temperature and salinity. Dashed lines indicate 95% confidence limits (a) and CSs minus RPs (b).