Annette Samuelsen, research leader of our Ocean Modelling Group, published an article in 2019 that recently made it on the “Hidden gems” list of the ICES Journal of Marine Science! These articles are hand-picked by the Editor-in-Chief for being of high quality and special interest, but not having many citations yet.

Samuelsen, together with her colleagues Ute Daewel (formerly NERSC, now Helmholtz-Zentrum Hereon, Germany) and Cecilie Wettre (Norwegian Meteorological Institute), found a way to estimate the risk that oil spills along the Norwegian coast can pose for Atlantic cod eggs and larvae. Their findings are relevant for potential future oil spills and a quick risk assessment in such a case.

Why cod?

Photo: Atlantic cod. Hans-Petter Fjeld (CC BY-SA 3.0: https://creativecommons.org/licenses/by-sa/3.0/)Atlantic cod (torsk in Norwegian) is one of the most common, but also one of the most valuable commercial fish we have in Norwegian waters. They spawn very close to the coast and in many fjords all along the entire Norwegian coastline, from the Skagerrak up to Northern Norway.

Not all fish eggs behave the same way, but Atlantic cod eggs are buoyant – they float close to the surface. Cod larvae are present in the same area as eggs, as they hatch from the eggs within a few weeks. They can maneuver up and down in the water column, but they are generally almost as helpless as the eggs.

Aside from fish in our waters, there are also many ships.

Ship traffic along the Norwegian coast

The Norwegian coast is a busy area for shipping, both cargo and due to oil production on the Norwegian shelf. Ships mainly travel along the coast instead of far out on the ocean. This means that ships are present in the same areas where fish eggs and larvae float around.

Ships use oil as fuel, and oil spills from ships are not unheard of globally, or even along our coastline. For example, the freighter “Server” partially sank close to Bergen in 2007, and oil was released into the water nearby.

Coastal and marine life can be affected by oil spills

Photo: Cod developmental stages. By Terje van der Meeren / Institute of Marine Research (CC BY-SA 4.0: https://creativecommons.org/licenses/by-sa/4.0/)

Everyone is familiar with images of birds covered in oil after marine spills, but oil does not just affect them. It also has effects on fish, especially in their early life stages – eggs and larvae. And when oil spills occur close to where large numbers of fish eggs and larvae are present, they are at risk. The weather and oceanic conditions obviously influence this risk, but to what extent? And how do the changes in weather and oceanic conditions throughout a year, as well as from year to year, affect the risk?

How can the risk be estimated? – Models!

Samuelsen and her colleagues set out to answer these exact questions. They used two different models and combined their results to do so. The first model is an oil spill model. It can simulate where and how oil from a spill location drifts and disperses. The second model is an egg- and larvae-drift model, which can simulate where and how fish eggs and larvae drift. Simulating these drift processes is crucial for the risk analysis, and for these models to work we need data on the weather and oceanic conditions. Otherwise, you cannot realistically imitate what happens in nature.

Reanalyses to strengthen the two drift models

Real life data, for example ocean data such as currents, sea surface temperature, and salinity, is included in a model to reconstruct the past conditions in the ocean under investigation. The output of this model including the real life ocean data is a so-called reanalysis, and it is much more complete than the original dataset, and can be used to enable other models to simulate what happened in the past in a realistic way. There are different kinds of reanalyses, such as for example oceanic reanalyses (for the oceanic conditions) and atmospheric reanalyses (for the weather conditions). Both those types were used in this study to properly simulate real conditions in the past. The oceanic reanalysis used is the TOPAZ reanalysis, based on HYCOM – an ocean model making use of data assimilation to include remote sensing ocean data and in-situ observations. The data assimilation scheme for TOPAZ has been developed at NERSC over the past years.

Samuelsen and her colleagues chose to look at a 20 year timeframe, from 1991 to 2010. They focused on the southern and western coast of Norway. Here, the general ocean current direction along the coast is from the south (by the Skagerrak) northwards, along the coast. Local variations are possible though.

Highest risk when?

According to their findings, the months March and April are the time of year with the highest risk for early life stages of cod being contaminated by oil along the Norwegian coast! This is because in these months, the amount of cod eggs and larvae in the coastal waters is extremely high, independent of oceanic and weather conditions. Large variations in between years are also possible (see figure below).

Highest risk where?

Oil spills that would occur closest to the coast have the highest potential for posing a risk to fish eggs and larvae. This is mainly due to the spawning grounds being right by the coast as well, making overlap very likely. 

Highest risk why?

Two factors stand out as the most important ones. Samuelsen and her colleagues found that both wind and currents are the driving factors for large or little overlap between areas with oil spill and fish eggs/larvae! They are mainly responsible for changes in the risk throughout the year, as well as for differences between years.

Two different years showing different levels of overlap during the same time of year, Figure 10 in Samuelsen et al. (2019). Left: In March 1994, an oil spill in the hypothetical location would have led to massive overlap of oil and cod eggs/larvae in the same region. Right: In March 2001, a hypothetical spill in the exact same location would have resulted in a much lower risk for cod eggs/larvae to be contaminated with oil. This shows that variations between years can be significant, and according to the recent study, this is mainly caused by variations in winds and currents.

How can these findings be used?

In the event of an actual oil spill along the southern or western coast of Norway, the developed model setup by Samuelsen et al. (2019) and their findings can come in handy. When taking current weather and ocean data into account, it is possible to make a quick risk analysis. That way we can see where the oil spill is most likely to overlap with the area where cod eggs and larvae are present! Then further measures can be taken to mitigate the risk of the oil spill affecting marine and coastal life.

Samuelsen on the relevance of this study from 2019: “We now have 10 more years of reanalysis data for the ocean and atmosphere with improved models. The methods presented in this work are still relevant today and can also be extended and applied to other marine organisms.”

Reference:

Samuelsen, A, Daewel, U, & Wettre, C. Risk of oil contamination of fish eggs and larvae under different oceanic and weather conditions, ICES Journal of Marine Science, Volume 76, Issue 6, November-December 2019, Pages 1902–1916, https://doi.org/10.1093/icesjms/fsz035

The full hidden gems list by the ICES Journal of Marine Science can be found here: 

https://academic.oup.com/icesjms/pages/high_impact_articles#e7fa4985199cec337ee6