SINMOD
The ocean model SINMOD has been developed by SINTEF in close collaboration with NTNU, UiT and other national and international research groups since 1987. What particularly distinguishes SINMOD from other Norwegian and international ocean models is its holistic approach to the marine ecosystem. SINMOD is a coupled ecosystem model that integrates physical, chemical and biological processes in the ocean, including the carbon system and the exchange of oxygen and CO₂ with the atmosphere.
SINMOD includes specialised modules for sea lice, virus dispersal, seaweed and jellyfish. In addition, it serves as a flexible research tool where new processes can be implemented easily to test hypotheses and explore different scenarios, such as “what happens if…?”. A key driver behind its development has been understanding how changes in physical conditions affect marine ecosystems.
Hydrodynamic component
The hydrodynamic module is based on the Navier-Stokes equations, solved using the finite difference method on an Arakawa C-grid. Model inputs include atmospheric forcing, freshwater inflows and boundary conditions (hydrography, density, tides and currents).
The first SINMOD configuration was developed for the Barents Sea, but the model has since been applied in many regions worldwide, including Chile, New Zealand, Brazil and the Mediterranean. Spatial resolution varies according to the region and the scientific challenge being addressed, ranging from 20 km down to 32 m horizontally. Vertically, the model uses z-layers.
A basic description of the model is provided by Slagstad and McClimans (2005).
Biological models for mesozooplankton, seaweed, jellyfish and mussels
Alongside the hydrodynamic component, an ecosystem module has been developed to simulate the fundamental processes of the marine ecosystem. The model describes how plankton production interacts with, among other things, the nitrogen, carbon and oxygen cycles in the ocean, and how this primary production supports zooplankton and the export of organic matter to deeper waters.
A detailed population model for Calanus (Calanus finmarchicus), one of the key species in the pelagic ecosystem, makes it possible to address important questions concerning species higher up in the marine food web. This includes both natural variability and ecosystem dynamics, as well as human impacts from activities such as fisheries and aquaculture.
SINMOD has been used in numerous studies of Nordic and Arctic ecosystems. Parts of the ecosystem module are described in Wassmann et al. (2006). An updated version of the Calanus model is described in Alver et al. (2016) and Chamorro (2025).
The SINMOD ecosystem module can be coupled with models for species and topics of particular interest. Examples include aquaculture involving mussels (Handå et al., 2011) and seaweed (Broch et al., 2013), as well as jellyfish dispersal (Majaneva 2025, 2026).
Detailed growth models for seaweed and other macroalgal species can be used to estimate cultivation potential, identify suitable aquaculture locations (Broch et al., 2019), and assess how seaweed and other species can be integrated into aquaculture as sustainably as possible. More information about seaweed can be found through Norwegian Seaweed Centre and SFI Seaweed.
Data assimilation
Ocean models are associated with uncertainties arising from spatial resolution, mathematical simplifications, parameterisations, and uncertain forcing and boundary conditions. Because of the ocean's non-linear and chaotic dynamics, small errors in the model's initial state can grow throughout the forecast period. Accurate initial conditions are therefore essential for reliable ocean forecasting.
Data assimilation combines observations with the model's calculated state while accounting for uncertainty in both the model and the observations. SINTEF has implemented ensemble-based data assimilation using the Ensemble Kalman Filter and the open-source software library PDAF. The ensemble represents multiple plausible ocean states and provides estimates of model uncertainty in both space and time.
Observations from satellites, buoys, vessels and autonomous platforms are used to update the model and establish a more accurate starting point for further forecasts. The uncertainty estimates can also be used as a priori information when planning autonomous operations, for example to select routes and sampling locations where new observations are expected to provide the greatest value.