NEMO: a numerical ocean model

Ocean models store physical  properties such as salinity, temperature and currents on a three-dimensional grid. Picture: Adobe Stock

Smaller ocean features can be resolved by using a finer grid with more points, but this requires more computational power. The model evolves these physical properties forward in time using its equations of motion. Models of sea ice and biogeochemistry work using similar principles.

Why do we need a numerical ocean model?

We need these models to predict the state of the ocean within short and long timescales for a variety of purposes, ranging from support to operations at sea (for example, search and rescue) to understanding the role of the ocean in the Earth’s climate system. As the ocean sits beneath the atmosphere, sea-surface temperature patterns have widespread impact on the weather over land. Largely because two-thirds of the Earth is covered by ocean and the heat capacity of water considerably outweighs that of the air, the ocean acts as a regulator of the atmosphere.

In polar regions temperatures become cold enough for seawater to freeze and sea ice forms on the surface of the ocean. Sea ice plays an important role in the climate system because it insulates the ocean from the colder atmosphere in winter and, being whiter than the ocean, reflects sunlight in the summer.

The NEMO modelling framework includes a sea-ice model component, known as SI³ (Sea Ice modelling Integrated Initiative). The sea-ice component is run along with the ocean component in a similar manner but using a different set of equations. To understand and prepare for climate change we need to account for the role of the ocean and sea ice.

How is the NEMO model developed?

Nucleus for European Modelling of the Ocean (NEMO) is a state-of-the-art ocean modelling framework. NEMO is developed by a European consortium with the objective of ensuring long-term reliability and sustainability of the code. In other words, the task of maintaining and developing such a complex computer programme requires a well-coordinated team effort, involves tens of developers and hundreds of users.

In the UK there are two member organisations: the Met Office and the National Oceanography Centre (NOC). Met Office Scientific Manager in Ocean Modelling, Ana Aguiar explains: “We work in partnership through the Joint Marine Modelling Programme, contributing to the development of NEMO. The code is publicly available for use in research and commercial applications. It is imperative to reach as many users as possible, to ensure the code gets tested and pushed to the limits of its usability. User requirements then prompt further advances.”

NEMO benefits from continual work to improve its performance (scientific and computational efficiency), to incorporate new scientific and process understanding, and to exploit the increase in supercomputer resources. When the developments are sufficiently mature and can provide significant scientific or technical improvements, a new NEMO version is released. Along with scientific upgrades (which tend to be increasingly computationally demanding), we must deliver code optimisation to make the best use of the available computing resources.
This video presents how NEMO is used by the Copernicus Marine Environment Monitoring Service.

What’s next?

The next NEMO release (expected to be rolled out this summer) will deliver significant improvements to model performance allowing it to run considerably faster. In the long term, among other things, we are also working towards porting the NEMO code to Graphical Processing Units (GPUs) to ensure continuity of the code in future mainstream High Performance Computing architectures

During April we are exploring the topic of the ocean and climate. Follow the #GetClimateReady hashtag on X (formerly Twitter) to learn more throughout the month.