Accounting for forecast uncertainties in communicating sea ice and weather information in the Arctic

There is a growing demand for accurate Arctic environmental predictions which puts great expectations on our current global and regional forecasting systems. However, the forecast accuracy is limited by the chaotic nature of the atmosphere and ocean and by the imperfection of our model systems. A careful and accurate estimation of the forecast accuracy, by means of a probabilistic forecast, is thus needed to provide the end-user high-quality information for decision-making. In FOCUS an interdisciplinary and international team of natural and social researchers, work together with intermediate- and end-users from business and governmental sectors, to advance the value chain for environmental predictions and associated uncertainties in the Arctic. The main objective is to operationalize advanced probabilistic weather and sea-ice Arctic forecast information, tailored to key social, environmental and economic needs, in order to support sustainable development in the Arctic. FOCUS will significantly advance (1) the environmental probabilistic forecasts for Arctic maritime operators, (2) the technology and infrastructure used to produce and disseminate information to sector-specific end-users, and (3) our understanding of how to link present-day and future requirements of maritime operators to relevant research questions and policy focus. The FOCUS website can be found here https://focus-arctic.com. In 2021, the work on the forecasting systems, the ingestion of forecast information into the dissemination systems, and the discussions with end-users had been initiated. The project team decided that there will be a focus on probabilistic sea ice and sea spray icing forecasting in the project. Furthermore, the Stakeholder Advisory Group, consisting of representatives of the Association of Arctic Expedition Cruise Operators, BarentsWatch, Royal Norwegian Navy and Fiskebåt has been established. In 2022, in WP-1 extensive work on coupled forecasting systems has been done. First, the ensemble sea-ice prediction system (based on neXtSIM) has been developed and a paper is under review. Second, the atmospheric model system (AROME Arctic) has been coupled to the ocean model GOTM, and a paper has been published in Ocean Modelling which highlights the use of thousands of ARGO and CTD ocean profiles for model verification by means of a machine learning approach. In addition, a paper has been published on the development of an ice-wave-ocean buoy for advanced measurements of the coupled Arctic system. In WP-2 the Norwegian Meteorological Insitute and NAVTOR collaboratively work on a probabilistic sea spray icing product. A 2-day workshop, which included the Stakeholder Advisory Group, as well as, other end-users from Hurtigruten and the Coast Guard, took place May 2022 in Tromsø. At this workshop the needs and opportunities for probabilistic sea-ice and weather forecasting has been discussed. In WP-3 an agent-based model has been developed to explore the uptake of a new sea ice information product by Arctic cruise operators. The model can be found here: ttps://berillblair.github.io/sea-ice-demo/ and a paper is currently in preparation.

Due to the Arctic's extreme environmental conditions and remoteness, maritime operators are strongly demanding user-specified weather and sea-ice predictions. Because of the chaotic nature of the coupled atmosphere-ocean-sea-ice system and systematic errors inherent in our model systems, careful and accurate estimation of the forecast accuracy is needed to provide the end-user with high-quality information for decision-making. Probabilistic forecasts based on ensemble prediction systems can provide the necessary information. In previous end-user workshops and training sessions, we have gained extensive knowledge where advanced services are needed and have the potential to drive innovation. We will built-up on this knowledge and create a consortium of natural and social scientists, commercial and public suppliers of navigational products and services, and end-users from the maritime sectors. For additional guidance and outreach to end-users not part of the project we establish a Stakeholder Advisory Group with key representatives from the shipping, tourism, military, and fisheries sectors. In order to allow for advanced probabilistic weather forecasting, the operational weather prediction system of MET Norway will be enhanced by coupling to a 1D ocean turbulence model allowing a physically consistent error propagation and operationally feasible coupling strategy. The ocean sea-ice prediction system, as part of the Copernicus Marine Environmental Monitoring Service, will also be enhanced using better knowledge of atmospheric forecast uncertainties. Novel ensemble based forecast products will be analysed, designed and refined in a co-production process. The involvement of partners from the public and commercial sectors, will warrant an interactive and flexible development of forecast products and ensure a user-oriented development of the system. present-day and future forecast information needs will be assessed with novel methodologies from the social sciences.