Rapid changes are occurring in the Arctic, including a reduction in sea ice thickness and coverage and a shift towards younger and thinner sea ice. Snow and sea ice models are often used to study these ongoing changes in the Arctic, and are typically forced by atmospheric reanalyses in absence of observations. ERA5 is a new global reanalysis that will replace the widely used ERA-Interim (ERA-I). In this study, we compare the 2 m air temperature (T2M), snowfall (SF) and total precipitation (TP) from ERA-I and ERA5, and evaluate these products using buoy observations from Arctic sea ice for the years 2010 to 2016. We further assess how biases in reanalyses can influence the snow and sea ice evolution in the Arctic, when used to force a thermodynamic sea ice model. We find that ERA5 is generally warmer than ERA-I in winter and spring (0–1.2 ∘C), but colder than ERA-I in summer and autumn (0–0.6 ∘C) over Arctic sea ice. Both reanalyses have a warm bias over Arctic sea ice relative to buoy observations. The warm bias is smaller in the warm season, and larger in the cold season, especially when the T2M is below −25 ∘C in the Atlantic and Pacific sectors. Interestingly, the warm bias for ERA-I and new ERA5 is on average 3.4 and 5.4 ∘C (daily mean), respectively, when T2M is lower than −25 ∘C. The TP and SF along the buoy trajectories and over Arctic sea ice are consistently higher in ERA5 than in ERA-I. Over Arctic sea ice, the TP in ERA5 is typically less than 10 mm snow water equivalent (SWE) greater than in ERA-I in any of the seasons, while the SF in ERA5 can be 50 mm SWE higher than in ERA-I in a season. The largest increase in annual TP (40–100 mm) and SF (100–200 mm) in ERA5 occurs in the Atlantic sector. The SF to TP ratio is larger in ERA5 than in ERA-I, on average 0.6 for ERA-I and 0.8 for ERA5 along the buoy trajectories. Thus, the substantial anomalous Arctic rainfall in ERA-I is reduced in ERA5, especially in summer and autumn. Simulations with a 1-D thermodynamic sea ice model demonstrate that the warm bias in ERA5 acts to reduce thermodynamic ice growth. The higher precipitation and snowfall in ERA5 results in a thicker snowpack that allows less heat loss to the atmosphere. Thus, the larger winter warm bias and higher precipitation in ERA5, compared with ERA-I, result in thinner ice thickness at the end of the growth season when using ERA5; however the effect is small during the freezing period.
The accuracy of the initial state is very important for the quality of a forecast, and data assimilation is crucial for obtaining the best-possible initial state. For many years, sea-ice concentration was the only parameter used for assimilation into numerical sea-ice models. Sea-ice concentration can easily be observed by satellites, and satellite observations provide a full Arctic coverage. During the last decade, an increasing number of sea-ice related variables have become available, which include sea-ice thickness and snow depth, which are both important parameters in the numerical sea-ice models. In the present study, a coupled ocean–sea-ice model is used to assess the assimilation impact of sea-ice thickness and snow depth on the model. The model system with the assimilation of these parameters is verified by comparison with a system assimilating only ice concentration and a system having no assimilation. The observations assimilated are sea ice concentration from the Ocean and Sea Ice Satellite Application Facility, thin sea ice from the European Space Agency's (ESA) Soil Moisture and Ocean Salinity mission, thick sea ice from ESA's CryoSat-2 satellite, and a new snow-depth product derived from the National Space Agency's Advanced Microwave Scanning Radiometer (AMSR-E/AMSR-2) satellites. The model results are verified by comparing assimilated observations and independent observations of ice concentration from AMSR-E/AMSR-2, and ice thickness and snow depth from the IceBridge campaign. It is found that the assimilation of ice thickness strongly improves ice concentration, ice thickness and snow depth, while the snow observations have a smaller but still positive short-term effect on snow depth and sea-ice concentration. In our study, the seasonal forecast showed that assimilating snow depth led to a less accurate long-term estimation of sea-ice extent compared to the other assimilation systems. The other three gave similar results. The improvements due to assimilation were found to last for at least 3–4 months, but possibly even longer.
22 September 2023 in Geoscientific Model Development
An operational ocean and sea ice forecast model, Barents-2.5, is implemented for short-term forecasting at the coast off northern Norway, the Barents Sea, and the waters around Svalbard. Primary forecast parameters are sea ice concentration (SIC), sea surface temperature (SST), and ocean currents. The model also provides input data for drift modeling of pollutants, icebergs, and search-and-rescue applications in the Arctic domain. Barents-2.5 has recently been upgraded to include an ensemble prediction system with 24 daily realizations of the model state. SIC, SST, and in situ hydrography are constrained through the ensemble Kalman filter (EnKF) data assimilation scheme executed in daily forecast cycles with a lead time up to 66 h. Here, we present the model setup and validation in terms of SIC, SST, in situ hydrography, and ocean and ice velocities. In addition to the model's forecast capabilities for SIC and SST, the performance of the ensemble in representing the model's uncertainty and the performance of the EnKF in constraining the model state are discussed.
The Ocean and Sea Ice Satellite Application Facility (OSISAF) of the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT) distributes satellite observations operationally. In this paper, we describe the development of a combined optimal interpolation and nudging (COIN) scheme in the Norwegian Meteorological Institute that assimilates OSISAF sea-ice concentration (SIC) into a coupled sea-ice–ocean Regional Ocean Modeling System (ROMS). The scheme modifies the modeled SIC at every time-step, based on the difference between observation and forecast as well as on the ratio of observation error to model error. Snow and sea-ice thickness are diagnostically modified to match the ice concentration field. Hindcast experiments of Arctic sea ice are performed for the whole of 2009. The results are compared with OSISAF and Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E) SIC, showing a significant improvement in the analyzed sea-ice edge and summer SIC.
The Los Alamos Sea Ice Model (CICE) is used by several Earth system models where sea ice boundary conditions are not necessary, given their global scope. However, regional and local implementations of sea ice models require boundary conditions describing the time changes of the sea ice and snow being exchanged across the boundaries of the model domain. The physical detail of these boundary conditions regarding, for example, the usage of different sea ice thickness categories or the vertical resolution of thermodynamic properties, must be considered when matching them with the requirements of the sea ice model. Available satellite products do not include all required data. Therefore, the most straightforward way of getting sea ice boundary conditions is from a larger-scale model. The main goal of our study is to describe and evaluate the implementation of time-varying sea ice boundaries in the CICE model using two regional coupled ocean–sea ice models, both covering a large part of the Barents Sea and areas around Svalbard: the Barents-2.5 km, implemented at the Norwegian Meteorological Institute (MET), and the Svalbard 4 km (S4K) model, implemented at the Norwegian Polar Institute (NPI). We use the TOPAZ4 model and a Pan-Arctic 4 km resolution model (A4) to generate the boundary conditions for the sea ice and the ocean. The Barents-2.5 km model is MET's main forecasting model for ocean state and sea ice in the Barents Sea. The S4K model covers a similar domain but it is used mainly for research purposes. Obtained results show significant improvements in the performance of the Barents-2.5 km model after the implementation of the time-varying boundary conditions. The performance of the S4K model in terms of sea ice and snow thickness is comparable to that of the TOPAZ4 system but with more accurate results regarding the oceanic component because of using ocean boundary conditions from the A4 model. The implementation of time-varying boundary conditions described in this study is similar regardless of the CICE versions used in different models. The main challenge remains the handling of data from larger models before its usage as boundary conditions for regional/local sea ice models, since mismatches between available model products from the former and specific requirements of the latter are expected, implying case-specific approaches and different assumptions. Ideally, model setups should be as similar as possible to allow a smoother transition from larger to smaller domains.
Rapid changes are occurring in the Arctic, including a reduction in sea ice thickness and coverage and a shift towards younger and thinner sea ice. Snow and sea ice models are often used to study these ongoing changes in the Arctic, and are typically forced by atmospheric reanalyses in absence of observations. ERA5 is a new global reanalysis that will replace the widely used ERA-Interim (ERA-I). In this study, we compare the 2 m air temperature (T2M) and precipitation between ERA I and ERA5, and evaluate these products using buoy observations from Arctic sea ice. We further assess how biases in reanalyses influence the snow and sea ice evolution in the Arctic, when used to force a thermodynamic sea ice model. We find that both reanalyses have a warm bias over Arctic sea ice in relation to the buoy observations. The warm bias is smaller in the warm season, and larger in the cold season, especially when the T2M is lower than −25 °C. Interestingly, the warm bias in the new ERA5 is on average 2.1 °C (daily mean) larger than ERA-I during the cold season. While ERA-I is drier than most modern reanalyses in the Arctic, the total precipitation along the buoy trajectories is often lower in ERA5 than in ERA-I. Nonetheless, the snowfall products are broadly similar for both ERA I and ERA5. ERA-I had substantial anomalous Arctic rainfall, which is greatly reduced in ERA5. Simulations with a freezing degree days (FDD) model and a 1D thermodynamic sea ice model demonstrate that the warm bias in ERA5 acts to reduce thermodynamic ice growth. However, the lower precipitation in ERA5 results in a thinner snow pack that allows more heat loss to the atmosphere. Thus, the larger warm bias and lower precipitation in ERA5, compared with ERA I, compensate in terms of the effect on winter ice growth. Ultimately, we find slightly thicker ice at the end of growth season when using ERA5 forcing, compared with ERA-I. Thus differences in the precipitation fields of the two reanalyses have a larger influence on the sea ice evolution than the T2M.