Transcript Document
Presentation to Marine Strategy Coordination Group (MSCG) J Icarus Allen (on behalf of the MEECE consortium) 22th February 2012 Brussels 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 Scientific Challenge MEECE is a FP7 Integrated Project which aims to push forward the state-of-the-art of our understanding of impacts of global climate change and direct anthropogenic drivers on marine ecosystems end to end. 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 GOALS The specific goals of MEECE are: To improve the knowledge base on marine ecosystems and their response to climate and anthropogenic driving forces and To develop innovative predictive management tools and strategies to resolve the dynamic interactions of the global change driver, changes in ocean circulation, climate, ocean acidification, pollution, over fishing and alien invasive species on the structure and functioning of marine ecosystems To expand the knowledge based and provide scientific tools for the implementation of the European Marine Strategy 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 The Policy Challenge • To improve the knowledge base on marine ecosystems and how they are impacted by anthropogenic and natural driver. • To provide input to governmental and non-governmental actors in the development of innovative tools and strategies for the rebuilding degraded marine ecosystems, protection and the sustainable use of the sea and its resources, in the perspective of the ecosystem approach. • To improve the knowledge base for protection and management scenarios aimed at reconciling the interests of the many economic groups benefiting from the marine resource (including coastal). • To support to EU Marine Strategy (long term ecological objectives), the EU Maritime Policy and the EU Common Fisheries Policy (ecosystem approach to the management of marine resources). 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 The MEECE The MEECE ApproachApproach Observations Meta Analysis Parameterisations Scenario Definition Simulations & Synthesis Management Strategy Evaluation Indicators Model Systems Experiments Knowledge Transfer Experiments and Parameterisations Exposure Experiments Effects of Herbicides on phytoplankton Experiments to inform models Multiple Stressors T, CO2, on phytoplankton, zooplankton, fish larvae T and Pollutants on Phytoplankton, zooplankton, benthic invertebrates Model response parameterised as penalty function on growth. Impacts of Cu and T on Copepod egg production 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 Coupled End to End Models Generic Coupler Two Way Coupled Models ERSEM-ECOSIM ROMS-NPZD-OSMOSE ROMS-PICSES-APECOSM NORWECOM.E2E “a thin layer of code for communication and data exchange, enveloped by explicit programming interfaces through which a physical host and any number biogeochemical models can pass information” 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 A Regional Modelling Approach Modelling allows us •Describe the state of the system and how it may evolve •Represent the dynamics of the pressure - state relationship •Assess the risk on a negative indicator event Barents Sea Baltic Sea GLOBAL Benguela Upwelling North Sea Atlantic Margin Black Sea Biscay Adriatic Sea Aegean Sea Model Library Mapping Model Outputs (Characteristics) to Descriptors Biodiversity Physio chemical Temperature Salinity Nutrient Nitrate Phosphate Silicate pH Biological Features Phytoplankton Small Large Zooplankton Small Large Fish Chlorophyll Net Primary Production Community production Bottom fauna* 1960 Invasive Species Commercial Fisheries Foodwebs X X X (x) X (x) Eutrophication Seabed integrity Contaminants (x) (x) X X (x) (x) X (x) x X (x) X X (x) x X (x) X X (X) x x (X) X (x) X X (x) Evaluated Hindcast 2000 *ERSEM NW European Shelf Only X Forecast……… (x) X X x 2050 (X) (X) x X 2100 Fit for Purpose If we are to use our simulations either for science or policy applications we need to understand and be able to articulate their quality. Validation and verification Model Skill Assessment Relationships between model and data Quantification of Uncertainty Observational Error Predictive Error Truth T Prediction P O Observation Residual f(O-P) Predictive uncertainty (e.g. numerical error, parameter uncertainty) Data assimilation is the art of reducing this distance Observational accuracy, (e.g. measurement error, range of replicates etc.) Range of Drivers Range of models 1960 Evaluated Hindcast 2000 Sources Scenario uncertainty Structural/parameter uncertainty Natural variability (attribution to global change) Forecast……… 2050 Reference Hindcasts 2100 Scenarios Past Policy relevant Future Climate Change Climate forcing 2040-2050 1980-2000 Reanalysis Hindcast 1960 Evaluated Hindcast 2000 Forecast……… 2050 Anthropogenic Sensitivity Eutrophication Fishing Pollution Eutrophication Fishing Pollution 2100 Changes in State D1 Biodiversity D2 Alien Invasive Species Biogeographic Approach Distribution of potential Habitats for Procentrum Minimum 2100 2000 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 i Pressure – State Relationships Eutrophication changes D5 Eutrophication Impact of terrestrial N load changes reduction on Primary Production in the Baltic Sea +50% -50% 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 Probability of a Negative indicator event D5 Eutrophication • • • • How well can we resolve the observed frequency distribution? How well can we resolve the thresholds? How might the frequency distribution evolve in the future? What are the consequences for GES? Winter Nitrate 18% of events above the reference level 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 Risk Assessment Knowledge about Probability Knowledge about outcome High Low Ambiguity (known unknowns) Known outcomes High D1 Biodiversity (habitats) D5 Eutrophication Acidification D3 Commercial Fishing D4 Foodwebs Low Pollutants D8 Pollutants Foodwebs D2 Invasive Species AIS Uncertainty (known unknowns) Ignorance (unknown unknowns) Relating IEA to Models Relating an Models to the Integrated Assessment Integrated Ecosystem Assessment –main outputs of models in Relation to importance of ecosystem components as viewed by contributing experts to the IEA 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 Implications for Resource Management 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 MEECE Model Atlas Web-based Tool online autumn 2012 Descriptor Fact Sheets: June 2012 Pollutants: Eutrophication: Biodiversity: Invasive species: Commercial species:. Food webs: Hydrography (climate change): 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100 Toward Operational Models MSE MEECE Experi ments Scenarios Climate & Anthropogenic Drivers New Models Data Research Fisheries Eutrophication Pollution? Mulitple driver Context Downstream Services Users Decision Support Tools GMES Research models Operational Models (Core Service) Operational Lesson Learnt and Future Challenges MEECE was conceived before the MSFD implementation plan – hence retro fitting to descriptors. • Models have a degree of maturity where a range of outputs are ‘fit for purpose’ • Model development dominated by research push – need user pull • Challenge to improve model skill (requires observations and monitoring) • Challenge , to translate Tbytes of model output into useful information for users • Challenge to demonstrate the usefulness of model outputs to users • Challenge to bring MEECE models to the operational arena • Challenge to develop the next generation of models (experiments, observations etc..) 1960 Evaluated Hindcast 2000 Forecast……… 2050 2100