Sea spray icing poses a great deal of risk to marine operations in the Arctic offshore water and in cold waters with favourable conditions for spray icing. By developing models and simulation frameworks for prediction of spray-icing rates and severity and by developing risk-informed decision-making frameworks, the SPRICE project helps decision-makers in the Norwegian maritime sector be equipped with better tools that support spray icing-related decision-making, risk perception, and operational safety.
The main role of SINTEF Nord is to lead Work Package 3 – Long term simulation. This work package develops a mathematical framework in order to predict and simulate the meteorological and oceanographic parameters and thus spray icing rates and frequencies, while accounting for their interdependence and long-term drift.
The abilities of machine learning algorithms, such as long-short-term-memory (LSTM) networks, and Bayesian-statistics methods, such as Sequential Monte Carlo (SMC) methods are explored to achieve this goal.
The simulation-based framework is suitable for scenario-based studies related to the risks associated with industrial applications in the Arctic Ocean.
The decadal changes in meteorological and oceanographic parameters and the impact of global warming and the effects of sea-ice retreat and its impact on wave heights and thus on spray icing rates will also be captured in the proposed simulation framework.