Science Highlights

Understanding how marine species might expand their ranges is crucial for preparing and managing ecosystems in the face of climate change. However, many projections lack clarity on how uncertain these future changes might be, making planning difficult. We looked at uncertainties in climate models to predict the spread of the Pacific oyster in Europe. Using different climate change scenarios and a mix of climate models, we found that the oyster's future recruitment areas could vary greatly depending on the emissions scenario. Particularly, under high emissions, oyster recruitment areas could double compared to low emissions. Uncertainties in climate models and natural variability play significant roles in these projections, showing that considering different scenarios is essential for planning for the future of marine ecosystems. Our study emphasizes the need to account for these uncertainties when making decisions.

This work has been published and the full article is available here

Spatial fisheries management is widely used to reduce overfishing and protect biodiversity. However, the effectiveness and optimization of spatial measures depend on accurately identifying ecologically meaningful areas, which can be difficult in mixed fisheries. To apply a method generally to a range of target species, we developed an ensemble of species distribution models (e-SDM) combining general additive models, generalized linear mixed models, random forests, and gradient boosting machines.

The e-SDM was used to integrate abundance indices from scientific bottom trawl surveys with the geopositional data and essential oceanographic variables (EOVs) from the three-dimensional physical-biogeochemical operational model applied to juvenile and adult stages of 10 marine demersal species representing 60% of the total demersal landings in the central areas of the Mediterranean Sea. Using the e-SDM results, hotspots of aggregation and potentially more selective areas were identified for each target species of otter trawl.

This work explores the current comprehension of oil spill pollution in the Mediterranean.

The Lagrangian oil spill model MEDSLIK-II was suggested as a reliable and powerful tool for the prediction of oil transport and fate at sea. The Haven oil spill occurred off the coast of Genoa in April 1991 was proposed as a test case for the model validation. Oil spill drift and transformation will be forced by the datasets from the Copernicus Marine Service reanalysis.

Stochastic oil spill modeling will be developed to generate statistics for multiyear products. Hazard maps illustrated by the previous implementations (i.e., in the Port of Taranto) were presented as the NECCTON’s prototypes. Further usage of hazard maps, the potential of their upgrading and extension of applicability were discussed. 

The UK Regional Environmental Prediction project systematically evaluates the interactions between components of the Earth system. As part of this work we have assessed the impact of coupling the WAVEWATCHIII wave model to a NEMO-ERSEM simulation of biogeochemistry on the Northwest European Shelf. We observed that the inclusion of waves predominantly influences the off-shelf region, with delays to the spring bloom and reduced total net primary production over the bloom season due to increased vertical mixing. On-shelf, where the waters are typically more mixed sees a much smaller influence from the inclusion of waves. The suspended particulate matter models being developed under NECCTON will be heavily impacted by the vertical structure of the water column, meaning the inclusion of wave coupling is vital to producing an accurate simulation.

Chromophoric dissolved organic matter (CDOM) interacts with the ambient light and gives the waters of the Mediterranean Sea their colour. We propose a novel parameterization of the CDOM cycle, whose parameter values have been optimized by using the data of the monitoring site BOUSSOLE. 

Nutrient and light limitations for locally-produced CDOM caused aCDOM(λ) to covary with chlorophyll, while the above-average aCDOM(λ) values observed at this site were maintained by allochthonous CDOM. The configuration, outputs and sensitivity analyses from this 1D model application will be instrumental for future applications of BFM to the entire Mediterranean Sea in a 3D configuration.

The development of an Artificial Neural Network (ANN) to alleviate nutrient bias in a conventional biogeochemical (BGC) model is important as the available observation (mostly surface chlorophyll) are insufficient to constrain nutrients without multi-variate Data Assimilation (challenging and computationally expensive). Furthermore, an ANN can handle large amounts of data in parallel with a relatively shorter execution time, unlike the grid point-to-grid point integration of a conventional model. Additionally, there are possibilities to resolve existing non-linear relationships between various parameters which are challenging for a numerical model too.
 
In this research, an ANN model is trained to predict nitrate concentration at the L4 station in the western English Channel. Input training features include atmospheric, ocean BGC and physical variables. Preliminary results exhibit good correlations with independent testing data. This ensures an improvement in representing complex BGC processes and thereby forecasting harmful events i.e., eutrophication with more accuracy.
 

This research explores how ocean environmental conditions affect benthic functional biodiversity and hence biogeochemical cycles and ecosystem services at shelf scales.

The study area covers the northwestern continental shelf of the Black Sea where macrobenthos samples were collected from in situ measurements between 1995 and 2017 for a total of more than 280 taxa sampled at 237 stations. Thirty functional traits (i.e morphological, physiological features defined at the species level) were determined for a representative subset of 150 taxa. 

Through multivariate analysis the traits were linked to the environmental conditions. Traits will then be mapped with a set of approaches combining the dynamic interpolation variational analysis tool (DIVA) combined with Neural Network and Habitat Suitability Models.

Marine ecosystems are currently under pressure from human activities and changes in their environment, linked to global warming. A better understanding of the impact of climate change on the ocean (temperature, currents, acidification, oligotrophication,…) is necessary to mitigate the decrease of biomass in the future.  

In the framework of NECCTON, climate projections will be conducted, and evaluated to assess the needs for fisheries management and biodiversity conservation. An ensemble of 21st century projections of the ocean’s state, from physics and biogeochemistry to biology, will be driven. Atmospheric forcings from coupled models part of CMIP6 (Couple Model Intercomparison Project) will serve as input for the projections, and will be debiased to better fit the mean state and variability of the hindcast. Different warming scenarios, based on different socio-economic pathways (SSPs) from various models will be tested, to encompass the uncertainty on climate change efforts that will be taken in the coming years.

Micronekton is a group of small organisms (2-20 cm) located at the intermediate level of the oceanic trophic chain (crustaceans, cephalopods, fish, etc.). They make daily vertical migrations between the surface at night to feed, and the deep ocean during the day to hide from predators. Through this migratory behaviour, micronekton is responsible for an active export of carbon to the depths, where it remains sequestered for very long periods. The study carried out as part of NECCTON focuses on micronekton modeling in its environment, by coupling the numerical model of micronekton population dynamics SEAPODYM-LMTL (Lehodey et al., 2010, 2015) and the biogeochemical model PISCES, which represents carbon fluxes in the lower trophic levels (Aumont et al., 2015). 
This coupling will enable the quantification of micronekton impact on the ocean carbon cycle. It will be carried out using the FABM framework, in one dimension thus modeling a simple water column. The implementation of SEAPODYM-LMTL in FABM has just started, PISCES already is, so the next step will be the proper coupling of these models.