The Baltic Sea Atlantis: An integrated end-to-end modelling framework evaluating ecosystem-wide effects of human-induced pressures

Abstract

Achieving good environmental status in the Baltic Sea region requires decision support tools which are based on scientific knowledge across multiple disciplines. Such tools should integrate the complexity of the ecosystem and enable exploration of different natural and anthropogenic pressures such as climate change, eutrophication and fishing pressures in order to compare alternative management strategies. We present a new framework, with a Baltic implementation of the spatially-explicit end-to-end Atlantis ecosystem model linked to two external models, to explore the different pressures on the marine ecosystem. The HBM-ERGOM initializes the Atlantis model with high-resolution physical-chemical-biological and hydrodynamic information while the FISHRENT model analyses the fisheries economics of the output of commercial fish biomass for the Atlantis terminal projection year. The Baltic Atlantis model composes 29 sub-areas, 9 vertical layers and 30 biological functional groups. The balanced calibration provides realistic levels of biomass for, among others, known stock sizes of top predators and of key fish species. Furthermore, it gives realistic levels of phytoplankton biomass and shows reasonable diet compositions and geographical distribution patterns for the functional groups. By simulating several scenarios of nutrient load reductions on the ecosystem and testing sensitivity to different fishing pressures, we show that the model is sensitive to those changes and capable of evaluating the impacts on different trophic levels, fish stocks, and fisheries associated with changed benthic oxygen conditions. We conclude that the Baltic Atlantis forms an initial basis for strategic management evaluation suited for conducting medium to long term ecosystem assessments which are of importance for a number of pan-Baltic stakeholders in relation to anthropogenic pressures such as eutrophication, climate change and fishing pressure, as well as changed biological interactions between functional groups.

Fig 1. A schematic diagram of the integrated ecosystem modelling and strategic management evaluation framework.

The diagram illustrates the current linking within the integrated modelling framework for investigating ecosystem-wide effects of human-induced pressures in the Baltic Sea.

Fig 2. Baltic Atlantis spatial polygon structure.

Fig 3. Baltic Atlantis model–biological structure.

Detail list of species included in each group is found in Table G in S1 File. The figure illustrates the main interactions focused upon in the current context, i.e. a comprehensive diagram of the full biological interactions between functional groups in the Baltic Atlantis model will be way too complex to overview in one diagram.

Fig 4. Spatial distribution of annual average surface Chl-a [mg m^-3].

As simulated by HBM-ERGOM for the annual average over the period 2005–2014 (A-B) and Baltic Atlantis for the annual average for the last five projection years (C). Areas marked in yellow represent concentrations of 5 mg m^-3 or higher.

Fig 5. Time series evolution of Baltic Atlantis functional groups.

Total biomass in metric tons of biological functional groups and total DIN (dissolved inorganic nitrogen) obtained from a 60-year reference run initialized with 2005 data.

Fig 6. Emergent diet composition of sprat, cod, and seals from the whole of the Baltic Sea.

Results are average from last 5 years of model 60-year calibration run for both juvenile and adult groups, with the prey being arranged on the x-axis according to the trophic level (A, C, E) and the dynamics of the diet composition simulated over the 60 years–combined for adults and juveniles (B, D, F).

Fig 7. Biomass, condition and demography of Baltic cod and sprat, taken as an annual average from last 5 years of calibration run for the whole Baltic Sea.

The individual and population metrics per age class include: (A, C) biomass for 3 age groups over the 60 year simulation period (B, D) number of individuals.

Fig 8. Spatial distribution of Baltic sprat biomass for the 60 year simulation period.

Fig 9. Spatial distribution of bottom oxygen concentration for the Baltic Sea for the 60 year simulation period.

Fig 10. Results of the four river load scenarios.

Percentage of change of the biomass for the different biological functional groups compared to the status-quo scenario #1.

Fig 11. Oxygen concentration of bottom layer: Baseline (scenario #1) vs. scenario #5.

Fig 12. Results of the five fishing mortality sensitivity analyses (scenario #6–#10 respectively).

Percentage of change of the biomass for the different biological functional groups compared to the status-quo scenario #1.