Application of an ice thermodynamic model to a shallow freshwater lagoon

In this study, we apply an ice thermodynamic model to a shallow freshwater lagoon in the south-eastern part of the Baltic Sea — the Curonian Lagoon. The model results were compared with the measurement data from three near‑shore stations during the period of 2004–2017. The simulation data showed the model to be capable of replicating ice thickness dynamics rather well (mean R = 0.92, RMSE = 6 cm). Although the model overestimated the number of ice days (NID) on average by one month (ranging from 3 to 40 days), the overall pattern was very similar to observations (R = 0.95). We further assessed the ice thickness and NID projections in the near (2021–2040) and far (2081–2100) future under two climate change scenarios (RCP4.5 and RCP8.5). The results showed that the mean (max) ice thickness could decrease by 10–49% (6–34%) in the near and 41–75% (22–55%) in the far future under RCP4.5, and by 2–52% (2–30%) in the near and 75–88% (50–71%) in the far future under RCP8.5 compared to the baseline period of 1986–2005. The NID will shorten by 9–19% (9–22%) in the near and 15–36% (46–57%) in the far future under RCP4.5 (RCP8.5) scenarios, compared with the baseline period.