Long-term changes in Baltic Sea ice conditions

In this Master's thesis the long-term changes in Baltic Sea ice characteristics were studied. In the Baltic Sea the length of ice season is 5-7 months. The amount of seasonal ice varies significantly from year to year. However, in the last 100 years there has been a decreasing trend in the ice occurrence, which has resulted mainly from climate warming. Both observations and model results were analyzed in order to find out the long-term ice statistics, changes in ice conditions and reasons behind these changes. Three stations along the Finnish coast were chosen, Kemi in the Bay of Bothnia, Utö in the Archipelago Sea and Loviisa in the Gulf of Finland. The time series were 120 years long including the dates of freezing and break-up, the length of ice season and the maximum annual ice thickness. The model used was NEMO/LIM-3 and the modeled time 1961-2007. The key questions for the thesis were the positive trend in ice thickness in Kemi station, the reasons for the 100 year long decreasing trends although the climate warming has not affected so long and the changes in drift ice thickness. The study results show that the probability of ice occurrence has been decreasing in Utö and is now 81 %. In Kemi and Loviisa the probability is still 100 %. The freezing date has become 7-24 days later, while the break-up date has taken place 11-20 days earlier in 100 years. Consequently, the observed length of ice season has become 18-46 days shorter in 100 years. The trend of the maximum annual ice thickness is not so uniform. In Kemi station, there is an increasing trend, whereas in Loviisa the trend is decreasing. According to the model, the maximum annual ice thickness has a decreasing trend also in Kemi. The maximum annual ice volume has a decreasing trend in the entire Baltic Sea and also in different basins (Bay of Bothnia, Bothnian Sea and Gulf of Finland). The modeled ice volume correlates well with the observed maximum annual ice extent even though the ice volume has higher inter-annual variations. The possible other reasons besides rise in air temperature were detected. Sea ice thickness depends in addition to air temperature also greatly on snow accumulation and ice dynamics. The observation sites are not documented, so the places may have changed. The observation sites are usually near harbors and the increasing shipping might have affected ice conditions. The land uplift has been over 1 meter in Kemi during the 120 years, so that might also have an influence on results. In Loviisa the nuclear power plant might have affected the past few decades.