Ice Season forecast under ClimateChange: Tipping element approach

The Sea of Okhotsk is a marginal sea of the western Pacific Ocean. It is one of the world's richest in biological resources and famous for the fishing industry. In winter, navigation on the Sea is difficult, if not impossible, due to the harsh conditions of the North and the presence of sea ice. On average, the ice-free period lasts from June to November. However, the start and end dates of the ice season vary from year to year within a month. Such variability is impossible to capture by meteorological methods, which have a limit of predictability for 10 days. The absence of a long-term forecast of the navigational period in the Sea of Okhotsk affects the safety of navigation and the reliability of transit transport.

Most of the studies of the distribution of ice floes focus on such factors as the location, time of year, water currents, and sea temperatures. In our study, we use the distribution of temperature in the atmosphere and wind direction (NCEP/NCAR re-analyses data set) because most of the area of the Sea of Okhotsk is located in monsoon climate zone. We propose a new approach to forecast predicting the upcoming ice advance/ retreat date by developing our Tipping element approach [1] elaborated for prediction of the Indian Summer Monsoon, which proved to be successful for prediction upcoming monsoon four years in a row (2016-2019).

The physical mechanism underlying forecast is the following. There is an atmospheric feature that appears at the beginning of the transition to the ice season. We show, for the first time, the evidence in observational data that a transition from open water season to ice season begins when the near-surface air temperature crosses a critical threshold. It appears in the form of spatially organized critical transitions in the atmosphere over the see. This event happening 2-3 months before the ice season is a starting point forecasting date of ice advance. We perform forecast in critical areas - tipping elements of the spatial structure of ice formation, which we identified via data analysis.

The retrospective test (over the period 2001-2017) confirms that the methodology allows forecasting the ice advance/retreat date more than one month in advance, with a success rate in 88% of the years within the error of +/- 4 days. Forecasts of the upcoming season 2018-2019 show successful results as well.

Climate change affects the ice season in the Sea of Okhotsk in the following aspects: there has been a declining trend in sea ice cover in recent years due to delays in the ice advance date. Season shift because it takes for the atmosphere longer time cooling down in autumn. The novel approach allows for accounting climate change effects.

ES acknowledges financial support of the EPICC project (18_II_149_Global_A_Risikovorhersage) funded by BMU, RM acknowledges the Russian Foundation for Basic Research (RFBR) (No. 20-07-01071)

[1] Stolbova, V., E. Surovyatkina, B. Bookhagen, and J. Kurths (2016): Tipping elements of the Indian monsoon: Prediction of onset and withdrawal. GRL 43, 1–9 [doi:10.1002/2016GL068392]