Chemical contamination-mediated regime shifts in planktonic systems

Abstract

Abrupt transitions leading to algal blooms are quite well known in aquatic ecosystems and have important implications for the environment. These ecosystem shifts have been largely attributed to nutrient dynamics and food web interactions. Contamination with heavy metals such as copper can modulate such ecological interactions which in turn may impact ecosystem functioning. Motivated by this, we explored the effect of copper enrichment on such regime shifts in planktonic systems. We integrated copper contamination to a minimal phytoplankton–zooplankton model which is known to demonstrate abrupt transitions between ecosystem states. Our results suggest that both the toxic and deficient concentration of copper in water bodies can lead to regime shift to an algal-dominated alternative stable state. Further, interaction with fish density can also lead to collapse of population cycles thus leading to algal domination in the intermediate copper ranges. Environmental stochasticity may result in state transition much prior to the tipping point and there is a significant loss in the bimodality on increasing intensity and redness of noise. Finally, the impending state shifts due to contamination cannot be predicted by the generic early warning indicators unless the transition is close enough. Overall the study provides fresh impetus to explore regime shifts in ecosystems under the influence of anthropogenic changes like chemical contamination.

Appendices

Appendix

Effect of stochasticity in low copper concentrations

Fig. 9

Probability density estimates of the (A, B, C) phytoplankton and (D, E, F) zooplankton populations under three different external copper concentrations: 5.085 $\mu g L^{-1}$, 5.095 $\mu g L^{-1}$ and 5.1 $\mu g L^{-1}$; $K=2$ $mgCL^{-1}$

Deficient copper concentrations also lead to bistable system dynamics resulting in planktonic regime shifts. The effect of stochasticity on such low ranges of copper concentration is examined when carrying capacity $K=2$. Similar to the toxic concentration case, the system switches to phytoplankton-dominated state prior to the fold bifurcation. The probability density of the observed values from the simulation is unimodal with mode around zooplankton-dominated equilibrium at copper concentration 5.1 $\mu g L^{-1}$. Subsequent small decrease of copper results in the system demonstrating bimodality at concentration 5.095 $\mu g L^{-1}$ and unimodal mode around phytoplankton-dominated state at concentration 5.085 $\mu g L^{-1}$ (see Fig. 9). Increased intensity of noise leads to decreased skewness of the probability densities.

Basin of attraction for the alternative stable states

The stochastic switch between the attractors in Fig. 7 can be understood with the help of basin of attraction for the two equilibria under different carrying capacities. When $K=2$, the boundary separating the basin of attraction is very close to both the phytoplankton and zooplankton-dominated equilibrium which facilitates multiple stochastic switching. On the other hand, the boundary is relatively farther away from the two attractor in case of higher carrying capacity, i.e., $K=3$ resulting in very infrequent switch.

Fig. 10

Basin of attraction for the bistable scenario for different carrying capacities. Left panel: $K=2$, $E=14.3$ $\mu g L^{-1}$; Right panel: $K=3$, $E=16$ $\mu g L^{-1}$. The red and the blue points denote initial conditions for which the system converges to the zooplankton-dominated and phytoplankton-dominated equilibrium respectively