Contributors: A. Hall (TVUK), J. Marsh (TVUK), J. Clark (Plymouth Marine Laboratory), S. Kay (Plymouth Marine Laboratory), J. A. Fernandes (AZTI)

Issued by: Telespazio Vega UK / Jenny Marsh

Issued Date: 05/03/2020

Ref: C3S_D422Lot2.PML.3.1_201907_Product_User_Guide_ERSEM_v1.2

Official reference number service contract: 2018/C3S_422_Lot2_PML/SC2

Introduction

Climate change is likely to have a significant impact upon the seas and oceans, affecting oceanic ecosystems and marine and coastal resources, such as natural fisheries, aquaculture; and services, such as tourism and recreational activities. In general, global climate models do not give the level of spatial and temporal detail needed to understand the ecological, social and economic impacts of climate change on regional shelf seas. To address this problem, and to allow for spatially detailed assessments of regional seas to be produced, an approach known as regional climate downscaling is often used. Regional downscaling involves running relatively high resolution models of limited areal extent that are driven at the boundaries by the outputs of coarser resolution models that are often global in extent.

This data set contains the outputs of two, regionally downscaled projections for European seas that have been generated using coupled hydrodynamic-biogeochemical models. In both cases, the European Regional Seas Ecosystem Model (ERSEM) is used to simulate marine biogeochemical processes. Two different hydrodynamic models are used: The Nucleus for European Modelling of the Ocean (NEMO) model, and the Proudman Oceanographic Laboratory Coastal Ocean Modelling System (POLCOMS).

ERSEM is an established ecosystem model for the lower trophic levels of the marine food web that has been widely used in European waters. Since its original development in the early nineties (RD.6), it has evolved significantly from a coastal ecosystem model for the North Sea to a generic tool for ecosystem simulations from shelf seas to the global ocean (RD.8). The ERSEM model is composed of a set of modules that compute the rates of change of its state variables given the environmental conditions of the surrounding water body, physiological processes, and the result of predator–prey interactions.

This user guide provides an overview of the NEMO-ERSEM and POLCOMS-ERSEM models and configurations used; and the climate variables which they produce. Quality assurance information regarding both models is also provided. The user guide additionally contains information on how to access and download the data.

Reference Documents

The following is a list of documents that are referenced in this user guide. Where referenced in the text, these are identified as RD-n, where 'n' is the number in the list below:
RD.1.  C3S-MFC model validation and known issues for the pan-European domain v1.
RD.2. Butenschön, M., Clark, J., Aldridge, J. N., Allen, J. I., Artioli, Y., Blackford, J., Bruggeman, J., Cazenave, P., Ciavatta, S., Kay, S., Lessin, G., van Leeuwen, S., van der Molen, J., de Mora, L., Polimene, L., Sailley, S., Stephens, N., and Torres, R. (2016). ERSEM 15.06: a generic model for marine biogeochemistry and the ecosystem dynamics of the lower trophic levels, Geoscientific Model Development, 9, 1293-1339, https://doi.org/10.5194/gmd-9-1293-2016.
RD.3. Butenschön, M. and Kay, S. (2016). Projections of change in key ecosystem indicators for planning and management of Marine Protected Areas: An example study for European seas, Estuarine Coastal and Shelf Science. DOI: 10.1016/j.ecss.2016.03.003
RD.4. Chantal Donnelly, Jafet C.M. Andersson & Berit Arheimer (2016). Using flow signatures and catchment similarities to evaluate the E-HYPE multi-basin model across Europe, Hydrological Sciences Journal, 61:2, 255-273, DOI: 10.1080/02626667.2015.1027710
RD.5. OSPAR. (2017). 'Third Integrated Report on the Eutrophication Status of the OSPAR Maritime Area'. OSPAR Publication 694.
RD.6. C3S-MFC Model Validation for the Northeast Atlantic Domain v1.
RD.7. Baretta, J. W., Ebenhöh, W., and Ruardij, P.: The European regional seas ecosystem model, a complex marine ecosystem model, Neth. J. Sea Res., 33, 233–246, doi:10.1016/00777579(95)90047-0, 1995.
RD.8. Butenschön, M., Clark, J., Aldridge, J. N., Allen, J. I., Artioli, Y., Blackford, J., Bruggeman, J., Cazenave, P., Ciavatta, S., Kay, S., Lessin, G., van Leeuwen, S., van der Molen, J., de Mora, L., Polimene, L., Sailley, S., Stephens, N., and Torres, R. (2016). ERSEM 15.06: a generic model for marine biogeochemistry and the ecosystem dynamics of the lower trophic levels, Geoscientific Model Development, 9, 1293-1339, https://doi.org/10.5194/gmd-9-1293-2016.
RD.9. Bruggeman, J., Bolding, K., 2014. A general framework for aquatic biogeochemical models. Environmental Modelling & Software 61: 249–265. DOI: 10.1016/j.envsoft.2014.04.002"
RD.10. Madec, G.: NEMO reference manual 3_6_STABLE: "NEMO ocean engine" Note du Pôle de modélisation, Institut Pierre Simon Laplace (IPSL), France, No 27, ISSN No 1288-1619, 2016.

NEMO-ERSEM and POLCOMS-ERSEM Dataset

The dataset contains modelled projections of changes in marine physics and biogeochemical variables used to infer climate change indicators, as well as changes in the lower trophic levels of the marine food web. It covers the northwest European shelf, Mediterranean Sea and part of the North East Atlantic from 2006 up to 2049 or 2099, depending on the model output set used.

The ERSEM model is coupled to two hydrodynamic biogeochemical models configured for two different study areas: NEMO and POLCOMS. Hydrodynamic models calculate mixing and transport coefficients for each tracer in the biogeochemical model, which makes it possible to include the impact of transport and mixing processes on tracer concentrations. They also pass spatio-temporal information (e.g. temperature) about the physical environment to the biogeochemical model, which is then used to calculate rates of change for key processes (e.g. respiration, which is a temperature dependent process).

POLCOMS is coupled to ERSEM using a bespoke coupler, whereas NEMO is coupled to ERSEM using the Framework for Aquatic Biogeochemical Models coupler (RD.9). Both the NEMO and POLCOMS models require various types of input data to simulate the future climate, including data describing model initial conditions, atmospheric conditions, open ocean boundary conditions and land contributions of fresh water and nutrients. The hydrodynamic biogeochemical models are forced at the surface boundary using the outputs of atmospheric models, the details of which can be found in section 2.4.

The ERSEM Model

ERSEM is an ecosystem model of marine biogeochemistry and the lower trophic levels of the marine food web. It simulates the cycles of carbon and the major nutrient elements nitrogen, phosphorous, and silicon within the marine environment. Organisms at the bottom of the marine food web (phytoplankton) play an integral role in these cycles and are represented in the model. The four types of phytoplankton included in ERSEM are diatoms, which use silicon to build their outer cell walls; and three groups that are primarily distinguished by their size: the pico-, nano- and micro-phytoplankton. Zooplankton are heterotrophic organisms, meaning that they cannot produce their own food and instead rely on nutrition from plant and animal matter, mainly phytoplankton. This provides a vital link to commercially exploited species higher up the marine food web. Three types of zooplankton are represented in the model, which are again grouped according to their size: heterotrophic nanoflagellates, microzooplankton and mesozooplankton. A schematic of ERSEM is shown in Figure 1.

Figure 1: Schematic of ERSEM, the European Regional Seas Ecosystem Model (RD.2)

NEMO-ERSEM

NEMO is a physical ocean circulation model that has been adapted to run in either regional or global configurations (RD.10). The NEMO-ERSEM dataset is based on the Atlantic Margin Model 7 km NEMO configuration. It extends over the northwest European shelf and northeast Atlantic Ocean, as shown in Figure 2. Although the domain extends beyond the shelf to include some of the adjacent Northeast Atlantic, the focus of this system is on the shelf itself and the deep water is primarily included to ensure there is appropriate cross-shelf exchange. Grid-points near to the model boundaries are strongly affected by the model boundary conditions and so products are provided for the interior of the domain only. The outermost 10 grid-points and points East of 10°E on the Baltic boundary are masked. Table 1 provides the dataset description for NEMO-ERSEM.

Due to the nature of NEMO's computational grid, horizontal velocities are evaluated at coordinate positions offset from those at which scalar variables (e.g. salinity) are evaluated. In order to make the dataset easier to use, all outputs have been interpolated onto the same grid as that on which the scalar variables are defined. In the vertical dimension, outputs have been interpolated onto 43 fixed depth levels from the original 51 sigma levels used in the model simulation.

Figure 2: The northeast Atlantic domain.

Table 1: NEMO-ERSEM dataset description. Conventions are used in NetCDF files, and are required to ensure conforming datasets follow metadata standards e.g. each variable must have an associated description of what it represents, including physical units if appropriate.

POLCOMS-ERSEM

POLCOMS is a physical ocean circulation model which is tailored for the simulation of shelf-sea and coastal areas. To create the POLCOMS-ERSEM dataset POLCOMS has been coupled to ERSEM and run on a pan-European domain. As for the NEMO-ERSEM dataset, all variables have been interpolated on to the same vertical and horizontal grid in order to make analysis and visualization easier for users. The domain extends over the northwest European shelf and the Mediterranean Sea. Grid-points near to the model boundaries are strongly affected by the model boundary conditions and so points on the Atlantic boundary have been masked. Figure 3 shows the extent of the pan-European domain, not including these masked points. Table 2 provides the dataset description for POLCOMS-ERSEM.

Figure 3: The pan-European domain.

Table 2: POLCOMS-ERSEM dataset description

Dataset Production

Indicators from NEMO-ERSEM and POLCOMS-ERSEM are provided on the basis of the Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment Report scenarios with two Representative Concentration Pathways (RCPs), RCP 8.5 and RCP 4.5. An RCP is a greenhouse gas concentration (not emission) trajectory used by the IPCC to describe different climate scenarios, all of which are considered possible depending on how much greenhouse gas is emitted in the future.

RCP 8.5 is a 'business as usual' scenario, with high concentrations of greenhouse gases in the atmosphere, associated with unmitigated emissions and concentrations continuing to rise throughout the century. RCP 4.5 is a more moderate scenario that envisions peak concentrations at 2040 before declining. The products are available under both RCPs for 2006 onwards to provide continuity over the whole modelling period, but in practice differences between the RCPs do not emerge until the 2030s.

The POLCOMS-ERSEM and NEMO-ERSEM models were driven by global climate model projections generated for the Coupled Model Inter-comparison Project Phase 5 (CMIP5). Global model outputs, interpolated onto the respective regional grids, were used at the open ocean boundary for both physical and biogeochemical conditions. Initial conditions for physical variables were taken from the driving global models. Initial conditions for biogeochemical variables were taken from a combination of global datasets and hindcast simulations. For POLCOMS-ERSEM, river discharge and N and P loadings were taken from E-HYPE model outputs (RD.4), using the same global climate model and a business-as-usual nutrient scenario. E-HYPE is a hydrological model from the Swedish Meteorological and Hydrological Institute (hypeweb.smhi.se). Climatological water and nutrient flows were used at the Baltic boundary, and these were kept constant through the modelled period. For NEMO-ERSEM, climatological river inputs were used.

The two model configurations used downscaled atmospheric forcing data generated using the Swedish Meteorological and Hydrological Institute (SMHI) Rossby Centre Regional Atmospheric Model (RCA4). These regional atmospheric model outputs were taken from the Coordinated Regional Climate Downscaling Experiment (CORDEX) initiative; for further information see www.cordex.org. Below is a short summary of the surface atmospheric forcing data used with each model:

• The Northeast Atlantic domain (NEMO model)
  • • • CMIP5 driving model: Met Office Hadley Centre (MOHC) Hadley Global Environment Model 2 – Earth System (HadGEM2-ES). Further information regarding the model is available here: https://portal.enes.org/models/earthsystem-models/metoffice-+%20hadley-centre/hadgem2-es
      • Regional climate model: SMHI RCA4, as for the pan-European domain
      • Greenhouse gas concentration scenarios: RCP 4.5 and RCP 8.5
      • Variables: 6 hourly surface forcing 10 m wind components, sea level pressure, 2 m air temperature and relative humidity, daily precipitation, shortwave and longwave radiation flux and cloud cover
• The pan-European domain (POLCOMS model)
  • • • CMIP5 driving model: Max Planck Institute-Earth System Model-Low Resolution (MPI-ESM-LR). Further information regarding the model is available here: https://www.mpimet.mpg.de/en/science/models/mpi-esm/
      • Regional climate model: Swedish Meteorological and Hydrological Institute Rossby Centre Regional Atmospheric Model. Further information regarding the model is available here: https://www.smhi.se/en/research/research-departments/climate-research-at-the-rossby-centre/rca4-nemo-a-high-resolution-regional-coupled-model-system-1.121105
      • Greenhouse gas concentration scenarios: RCP 4.5 and RCP 8.5
      • Variables: 6 hourly surface forcing 10 m wind components, sea level pressure, 2 m air temperature and relative humidity, daily precipitation, shortwave and longwave radiation flux and cloud cover

The dataset production workflow is available in Figure 4.

Figure 4: Production workflow

The datasets are provided through the Copernicus Climate Change Service (C3S) Climate Data Store (CDS): 

• https://cds.climate.copernicus.eu/cdsapp#!/dataset/sis-fisheries-abundance?tab=overview
• https://cds.climate.copernicus.eu/cdsapp#!/dataset/sis-fisheries-eutrophication?tab=overview
• https://cds.climate.copernicus.eu/cdsapp#!/dataset/sis-fisheries-ocean-fronts?tab=overview

Quality Assurance

POLCOMS-ERSEM

The POLCOMS-ERSEM model has been validated through comparison to satellite data, using monthly values for years 1998-2015. Satellite chlorophyll data came from the ESA Climate Change Initiative (CCI) Ocean Colour project. Sea surface temperature is from the Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA) dataset. The model broadly reproduces the temporal and spatial patterns of chlorophyll concentration across the region. However, model estimates tend to be higher than those from satellite, especially in regions of high chlorophyll. Sea surface temperature also produced a good spatial and temporal match between the model and the satellite data. The POLCOMS-ERSEM model has been extensively used in studies of the Northwest European Continental Shelf, and has been applied successfully in the Mediterranean Sea (RD.2, RD.3).
The POLCOMS-ERSEM model contains the following known issues in model outputs:

• Mediterranean phosphate values are too high. The effect of this is that the modelled ecosystem is limited by nitrogen instead of phosphorus. However, tests suggest that the resulting productivity is similar.
• Nutrients and pH near river mouths are inaccurate. River inputs of nutrients and dissolved inorganic carbon were inaccurate affecting biogeochemical model outputs around river mouths. The pH in the western Mediterranean, influenced by the Rhône output, is very affected and should not be used. Other biogeochemical variables should be treated with caution near large river mouths. There is also excess nitrate input from the Baltic Sea leading to over-high production in the Norwegian Trench. River discharge volume and physical model outputs such as temperature and salinity are not affected.

Further details regarding model validation and known issues for the POLCOMS-ERSEM model covering the pan-European domain is available in RD.1.

NEMO-ERSEM

Surface chlorophyll and sea surface temperature output by the NEMO-ERSEM model have been compared to satellite values, using monthly values for years 2000-2015. Satellite chlorophyll data come from the ESA CCI Ocean Colour project (http://www.esa-oceancolour-cci.org/, v4.0). Sea surface temperature is from the OSTIA dataset, downloaded from CMEMS (http://marine.copernicus.eu/). Two model runs have been compared to the satellite values: the climate run, which was driven by a downscaled General Circulation Model (GCM), and a separate run driven by reanalysis data.

The model broadly reproduces the temporal and spatial patterns of chlorophyll concentration across the region. However, model estimates tend to be higher than satellite, especially in regions of high chlorophyll. On the other hand, model outputs are lower than satellite in the northern winter, although satellite data tends to be sparse for this period and region because of high cloud coverage, resulting in higher than usual uncertainty and the CCI Ocean Colour algorithm over predicts in turbid shelf waters in winter. Outputs from the GCM-driven model generally have a higher bias and root-mean-square-deviation than the reanalysis-driven model. Additionally, the correlation to satellite data is weaker, which is expected since the climate model cannot reproduce the month-by-month conditions as closely as the reanalysis. The model-satellite correlation is weakest in shallow coastal areas; however, satellite chlorophyll estimates are less reliable in these areas than for open seas.

There is also a good spatial and temporal match between modelled and satellite-derived sea surface temperature. The reanalysis-driven run of the model tends to overestimate sea surface temperature in shelf waters, especially in the spring and summer. The GCM-driven run has higher bias, and over-estimates winter temperatures in much of the region. This is in line with the biases in the HadGEM2-ES model driving the model. Model-satellite correlation is high in all regions, though rather lower for the GCM-driven run: as noted above, a climate model is not expected to accurately reproduce conditions month-by-month, even in a hindcast.
Copernicus Climate Change Service

Related articles

• Page:
  NEMO-ERSEM and POLCOMS-ERSEM: User Guides for Products (v1.21)
• Page:
  NEMO-ERSEM and POLCOMS-ERSEM: User Guides for Products (v1.1)
• Page:
  Summary indicators (Tier 1): Product User Guide
• Page:
  Risk indicators (Tier 3): Product User Guide
• Page:
  Loss indicators (Tier 3): Product User Guide