Search

This paper investigates new observations from the poorly understood region between the Kara and Laptev Seas in the Eastern Arctic Ocean. We discuss relevant circulation features including riverine freshwater, Atlantic-derived water, and polynya-formed dense water, emphasize Vilkitsky Strait (VS) as an important Kara Sea gateway, and analyze the role of the adjacent ∼250 km-long submarine Vilkitsky Trough (VT) for the Arctic boundary current. Expeditions in 2013 and 2014 operated closely spaced hydrographic transects and 1 year-long oceanographic mooring near VT's southern slope, and found persistent annually averaged flow of 0.2 m s−1 toward the Nansen Basin. The flow is nearly barotropic from winter through early summer and becomes surface intensified with maximum velocities of 0.35 m s−1 from August to October. Thermal wind shear is maximal above the southern flank at ∼30 m depth, in agreement with basinward flow above VT's southern slope. The subsurface features a steep front separating warm (–0.5°C) Atlantic-derived waters in central VT from cold (

The simulation of extremes using climate models is still a challenging task. Currently, the model grid horizontal resolution of state-of-the art regional climate models (RCMs) is about 11–25 km, which may still be too coarse to represent local extremes realistically. In this study we use dynamically downscaled ERA-40 reanalysis data of the RCM COSMO-CLM at 18 km resolution, downscale it dynamically further to 4.5 km and finally to 1.3 km to investigate the impact of the horizontal resolution on extremes. Extremes are estimated as return levels for the 2, 5 and 10‑year return periods using ‘peaks-over-threshold’ (POT) models. Daily return levels are calculated for precipitation and maximum 2 m temperature in summer as well as precipitation and 2 m minimum temperature in winter. The results show that CCLM is able to capture the spatial and temporal structure of the observed extremes, except for summer precipitation extremes. Furthermore, the spatial variability of the return levels increases with resolution. This effect is more distinct in case of temperature extremes due to a higher correlation with the better resolved orography. This dependency increases with increasing horizontal resolution. In comparison to observations, the spatial variability of temperature extremes is better simulated at a resolution of 1.3 km, but the return levels are cold-biased in summer and warm-biased in winter. Regarding precipitation, the spatial variability improves as well, although the return levels were slightly overestimated in summer by all CCLM simulations. In summary, the results indicate that an increase of the horizontal resolution of CCLM does have a significant effect on the simulation of extremes and that impact models and assessment studies may benefit from such high-resolution model output.

The Ronne Ice Shelf is known as one of the most active regions of polynya developments around the Antarctic continent. Low temperatures are prevailing throughout the whole year, particularly in winter. It is generally recognized that polynya formations are primarily forced by offshore winds and secondarily by ocean currents. Many authors have addressed this issue previously at the Ross Ice Shelf and Adélie Coast and connected polynya dynamics to strong katabatic surge events. Such investigations of atmospheric dynamics and simultaneous polynya occurrence are still severely underrepresented for the southwestern part of the Weddell Sea and especially for the Ronne Ice Shelf. Due to the very flat terrain gradients of the ice shelf katabatic winds are of minor importance in that area. Other atmospheric processes must therefore play a crucial role for polynya developments at the Ronne Ice Shelf. High-resolution simulations have been carried out for the Weddell Sea region using the non-hydrostatic NWP model COSMO from the German Meteorological Service (DWD). For the austral autumn and winter (March to August) 2008 daily forecast simulations were conducted with the consideration of daily sea-ice coverage deduced from the passive microwave system AMSR-E. These simulations are used to analyze the synoptic and mesoscale atmospheric dynamics of the Weddell Sea region and find linkages to polynya occurrence at the Ronne Ice Shelf. For that reason, the relation between the surface wind speed, the synoptic pressure gradient in the free atmosphere and polynya area is investigated. Seven significant polynya events are identified for the simulation period, three in the autumn and four in the winter season. It can be shown that in almost all cases synoptic cyclones are the primary polynya forcing systems. In most cases the timely interaction of several passing cyclones in the northern and central Weddell Sea leads to maintenance of a strong synoptic pressure gradient above the Ronne Ice Shelf. This strong synoptic forcing results in a moderate to strong offshore surface wind. It turned out that these synoptic depressions lead to strong barrier winds above the northwestern Ronne Ice Shelf and along the eastern flank of the Antarctic Peninsula. The fact, that these barrier winds often appear prior or during the initial break up of sea ice at the shelf ice edge, suggest that this mesoscale wind phenomenon plays a crucial role for polynya development. Furthermore, even mesoscale cyclogenesis above the Ronne Ice Shelf and the following northeastward passage of such a system can break up sea-ice cover under large-scale stationary weather conditions.

Spatial and temporal characteristics of the Storfjorden polynya, which forms regularly in the proximity of the islands Spitsbergen, Barentsøya and Edgeøya in the Svalbard archipelago under the influence of strong north-easterly winds, have been investigated using thermal infrared satellite and airborne imagery. Thin ice thicknesses were calculated from MODIS ice surface temperatures, combined with ECMWF ERA-Interim reanalysis atmospheric data in an energy balance model for the winters of 2002/2003 to 2013/2014 (Nov.-Mar.). Based on calculated thin ice thicknesses, associated quantities like polynya area and total ice production were derived. Calculated values underline the importance of this relatively small coastal polynya system considering its contribution to the cold halocline layer through salt release during ice formation processes. Airborne thermal infrared scanner data were collected during the field campaign LEAST (Lead and ABL study in the Transpolar System) in March 2014 using the POLAR 5 aircraft (Basler BT-67). Spatia l distributions of the ice surface temperature were acquired with a swath-width of approximately 2 km and a spatial resolution of 2.7 m at nadir for an area of roughly 16x18 km². In addition, in-situ meteorological measurements were performed (including turbulent flux measurements at low levels). This aircraft–based dataset is used for an improvement of remote sensing methods, the verification of the MODIS-derived quantities and investigations of sub-grid effects on the scale of a satellite pixel. A case study is presented for 16th March 2014 in the Storfjorden polynya area.

Das Fach Umweltmeteorologie an der Universität Trier führt seit 2007 Kooperationsprojekte mit Russland durch. Diese Kooperation kann in den kommenden drei Jahren (2013-2016) im Rahmen eines durch das Bundesministerium für Bildung und Forschung (BMBF) geförderten Projekts für Wissenschaftlich-Technische Zusammenarbeit (WTZ) zwischen Deutschland und Russland fortgesetzt werden.

We present investigations of Arctic polynya dynamics for the period 2002/2003 to 2011/2012. Thin ice thicknesses were calculated from MODIS ice surface temperatures, combined with ECMWF ERA-Interim reanalysis atmospheric data in an energy balance model. Regions of interest include the North Water Polynya, located between Ellesmere Island (Canada) and Greenland, and the Laptev Sea flaw polynyas. Based on calculated thin ice thicknesses, associated quantities like polynya area and total ice production were derived for all regarded regions and compared to recent studies using passive microwave remote sensing data. Calculated ice production reaches mean values of 223 km3 for the North Water Polynya and 79 km3 for the Laptev Sea. They underline the importance of the two coastal polynya systems in the context of the Arctic sea ice budget, although their individual contribution seems to be overestimated in other satellite-based studies. For both regions, obtained polynya areas and ice production clearly exceeded the corresponding values from passive microwave studies, despite a good agreement in the overall seasonal development. Possible reasons include a hidden effect of undetected clouds and the applied parametrizations in the polynya area retrieval. The application of a simple cloud coverage-correction scheme yielded reasonable adjustments for the polynya area and accumulated ice production, while open questions originating from inherent cloud effects have to be addressed in future studies. Noticeably, the sea ice cover in both regarded polynya regions shows signs of a delayed fall freeze-up over the 10 year-period.

The Laptev Sea area of the Siberian Arctic is known as a region of high polynya activities throughout the winter season. We analyze the impact of open-water and thin-ice covered polynyas on heat and moisture fluxes and the atmospheric boundary layer (ABL) using downscaled NWP simulations. ERA-Interim reanalysis data are used as forcing for dynamically downscaled COSMO runs with 15 and 5 km horizontal resolution. Sea ice information is taken from AMSR-E data and the period of investigation is 2002-2011. Our results clearly prove that polynyas moisten and heat the air downwind the polynya (up to several hundred kilometers) and additionally increase cloudiness. The analysis of surface energy balance components shows the sensible heat flux H0 as the largest contributor to ice production. Mean monthly H0 over polynyas is about 150 W/m2 for Dec.-Feb. 2002-2011. This is about three times higher than the energy loss by net radiation. Small polynyas have the largest heat loss (and ice production) per surface unit. In comparison with most previous studies our results suggest that most preceding studies overestimated the polynya ice production of the Laptev Sea.

Polynyas and leads are key elements of the wintertime Arctic sea-ice cover. They play a crucial role in surface heat loss, potential ice formation and consequently in the seasonal sea-ice budget. We apply and evaluate different lead segmentation techniques base on sea-ice surface temperatures as measured by the Moderate Resolution Imaging Spectroradiometer (MODIS). Daily lead composite maps indicate the presence of cloud artifacts that arise from ambiguities in the segmentation process and shortcomings in the data from thermal infrared image sequences, and applied to daily MODIS data from January to April 2008. The daily lead product can be used to deduct the occurrence, structure and dynamics of wintertime sea-ice leads and to assess seasonal divergence patterns of the Arctic Ocean.

In-situ measurements of the land-fast sea ice energy balance are scarce. We present a data set that comprises eddy-covariance measurements of sensible and latent heat as well as measurements of the sea-ice temperature gradient, long-wave and short-wave radiation measurements over land-fast sea ice in Atka Bay, Antarctica. With this setup we are able to monitor all components of the sea-ice energy budget. Additionally, we also measured the turbulent flux of CO2 over sea ice. This 37 day-long data set is evaluated for the transition period from austral winter to summer (November to December 2012) with regard to atmospheric stability and the general weather conditions. Results for the eddy-covariance measurements show an average sensible heat flux of 6.45+-10.72 W/m2 and a latent heat flux of 12.71+-9.48 W/m2 (with one standard deviation respectively) for low pressure/high wind-speed conditions. The average net radiation is 44.37+-41.54 W/m2 and for the CO2 flux an average of -3.35+-3.37μmol/m2 was measured. During high pressure/low wind-speed conditions an average of -3.03+-10.48 W/m2 and 10.76+-10.52 W/m2 was recorded for the sensible and latent heat flux, while the average net radiation and the CO2 flux are 35.63+-56.70 W/m2 and -1.95+-1.72μmol/m2 respectively. The fast ice is therefore found as a sink of CO2 for both situations.

Snow cover on sea ice and its impact on radar backscatter, particularly after the onset of freeze-thaw processes requires increased understanding. We present a data set that comprises in-situ measured snow properties from the land-fast sea ice of the Atka Bay, Antarctica, in combination with high-resolution TerraSAR-X backscatter data. Both data sets are discussed for the transition period from austral winter to summer (November 2012 - January 2013). The changes in the seasonal snow cover are reflected in the evolution of TerraSAR-X backscatter. We are able to explain between 62 % and 80 % of the spatio-temporal variations of the TerraSAR-X backscatter signal with up to three snow-pack parameters by using a simple linear model. Especially after the onset of melt processes, the majority of the TerraSAR-X backscatter variations are influenced by snow depth, snow/ice-interface temperature and snow-pack grain size and thereby imply the potential to also retrieve snow physical properties from X-Band backscatter.

Atmosphere/sea-ice/ocean exchange processes over the NOW (North Water) polynya in northwest Greenland were studied during the aircraft-based experiment IKAPOS (Investigation of Katabatic winds and Polynyas during Summer) in June 2010. The measurements were performed using the research aircraft POLAR 5 of Alfred Wegener Institute (AWI, Bremerhaven). Besides navigational and basic meteorological instrumentation, the aircraft was equipped with radiation and surface temperature sensors, two laser altimeters, and video and digital cameras. In order to determine turbulent heat and momentum fluxes, POLAR 5 was instrumented with a turbulence measurement system collecting data on a nose boom with a sampling rate of 100 Hz. While a polynya is associated with a convective boundary layer during winter, the situation in early summer is quite different. With the surface temperature at the melting point of ocean water, warm air advection leads to the formation of a stable boundary layer. Over the NOW in June, a stable, but fully turbulent boundary layer with strong winds of 15 m s-1 to 20 m s-1 was found during conditions of relatively warm synoptically induced northerly winds through the Nares Strait. Strong surface inversions were found in the lowest 100 m to 200 m. As a consequence of channeling effects at Smith Sound a well-pronounced low-level jet system was documented. These channeling effects lead to an increased wind-induced sea-ice export from the Nares Strait through Smith Sound. Cross-sections of mean quantities over the polynya are presented. The multiresolution decomposition is used to identify the spectral gap for the computation of turbulent fluxes. For the NOW polynya flights gap scales are between 500 m – 1000 m. Sensible heat fluxes are around -30W/m2 in the area of highest wind speed at Smith Sound. Accordingly, the momentum flux shows also a maximum caused by the channeling, which is an important factor for the process of NOW formation. The experimental data for the fully turbulent SBL are used to test parameterizations of exchange coefficients. The results of IKAPOS are valuable for the validation of numerical models (including climate models) and will contribute to the understanding of the exchange processes over summertime Arctic polynyas and the formation of the NOW polynya.

Spatial and temporal characteristics of the Storfjorden polynya, which forms regularly in the proximity of the islands Spitsbergen, Barentsøya and Edgeøya in the Svalbard archipelago under the influence of strong north-easterly winds, have been investigated for the period 2002/2003 to 2011/2012 using thermal infrared satellite imagery. Thin ice thicknesses were calculated from MODIS ice surface temperatures, combined with ECMWF ERA-Interim reanalysis atmospheric data in an energy balance model. Based on calculated thin ice thicknesses, associated quantities like polynya area and total ice production were derived and compared to previous remote sensing and modelling studies. It appears that the sea ice in the Storfjorden area shows signs of a delayed fall freeze-up over the 10 year-period, with an increasing frequency of large polynya events until the end of December. Average ice production in the fjord is estimated with 19.9+-3.9 km3 and is therefore slightly lower compared to previously calculated values by other authors. Nevertheless it underlines the importance of this relatively small coastal polynya system considering its contribution to the cold halocline layer through salt release during ice formation processes. Application of a simple cloud coverage-correction scheme yielded reasonable adjustments for the polynya area and accumulated ice production, while some open questions originating from inherent cloud effects and the applied parametrizations in the polynya area retrieval have to be addressed in future studies.

Sea ice fastened to coasts, icebergs and ice shelves is of crucial importance for climateand ecosystems. At the same time, it is not represented in climate models and many processes affecting its energy- and mass balance are currently only poorly understood. Near Antarctic ice shelves, which fringe about 44 % of the coastline, this landfast sea ice exhibits two unique characteristics that distinguish it from most other sea ice: 1. Ice platelets form and grow in supercooled water masses, which originate from cavities below the ice shelves. These crystals rise to the surface, where they accumulate beneath the solid sea-ice cover. Through freezing of interstitial water they are incorporated into the sea-ice fabric as platelet ice. 2. A thick and partly multi-year snow cover accumulates on the fast ice, altering the response of the surface to remote sensing and affecting sea-ice energy- and mass balance. In order to improve our understanding of these processes, we perform a continuous measurement program on the landfast sea ice of Atka Bay, Antarctica, contributing to the international Antarctic Fast Ice Network (AFIN). In addition, we will intensify our measurements during two field campaigns. Here we present our major research questions, introduce our methods and present first results.

Dense shelf water is an essential ingredient to the formation of Antarctic Bottom Water (AABW). It is formed on the continental shelves surrounding Antarctica, when freezing rates are sufficiently high to push ocean salinity to values of 34.65 and higher. Coastal polynyas, where the ice is driven away from the coastline, maintain the highest freezing rates in Antarctic winter. Since theWeddell Sea is considered the most productive source region of AABW, we investigate the dense water formation on the continental shelves of the southwestern Weddell Sea, with a focus on the role of coastal polynyas, using the Finite Element Sea ice-Ocean Model (FESOM), a primitive-equation, hydrostatic ocean model coupled with a dynamic-thermodynamic sea ice model. The horizontal resolution of the global, unstructured mesh is up to 3 km at the southwestern Weddell Sea coastline; in vertical direction the mesh features 37 depth levels (resolution increases toward the surface). The model was initialized on 01/01/1980 with data from the Polar Hydrographic Climatology and forced with NCEP/NCAR Reanalysis data. The 20-year period 1990-2009 is used for analysis. Our results indicate that in an average year, the polynya freezing rates of 9 cm d--1 (corresponding to a salt input of 2.5 kg m--2d--1) cause a seasonal variation in salinity of 0.3 psu under the Ronne polynya and result in the production of 5.10-4 km-3 dense shelf water, which leaves the continental shelf (outlined by the 700 m isobath in this study) at a long-term mean volume flux of 5.2 Sv. Some of this water contributes to the formation of Weddell Sea Deep/BottomWater, but a large fraction is diluted by mixing with ambient water and leaves the Weddell Sea at intermediate levels.

The experiment IKAPOS (Investigation of Katabatic winds and Polynyas during Summer) was performed in June 2010. It was targeted at the summertime katabatic wind system in coastal areas of northwest Greenland and at atmosphere/sea-ice/ocean exchange processes over the North Water (NOW) Polynya. The katabatic-driven near-surface wind regime plays a key role in exchange processes of energy and momentum between the atmosphere and the underlying surface over the ice sheet of Greenland. During summer, cooling of the boundary layer and hence the katabatic forcing is less than during winter, but strong winds can occur under appropriate synoptic forcing. On the other hand, the NOW Polynya represents one of the largest polynyas of the Arctic, and the air-sea interaction over the NOW has a significant impact on ocean processes, ice formation, gas exchange and biology. IKAPOS is mainly based on measurements with the research aircraft POLAR 5 of Alfred Wegener Institute (AWI, Bremerhaven). In order to determine turbulent heat and momentum fluxes, POLAR 5 was instrumented with a turbulence measurement system collecting data on a nose boom with a sampling rate of 100 Hz. Besides navigational and further basic meteorological instrumentation, the aircraft was equipped with radiation and surface temperature sensors, laser altimeters, and video and digital cameras. A total of six research flights have been performed, two of them were katabatic wind flights (over Humboldt and Steenstrup Glacier, respectively). The katabatic wind flights capture conditions of weak and strong synoptically forced katabatic wind with up to 14 m s-1 wind speed. Over the NOW, a stable, but fully turbulent boundary layer with strong winds of 15 m s-1 to 20 m s-1 was present during conditions of relatively warm synoptically induced northerly winds. Strong surface inversions were found in the lowest 100 m to 200 m. As a consequence of channeling effects at Smith Sound a well-pronounced low-level jet system was documented. These channeling effects lead to an increased wind-induced sea-ice export from the Nares Strait through Smith Sound. Three-dimensional (3D) structures of mean quantities will be presented as well as turbulent fluxes and energy budgets in the ABL over the ice sheet of Greenland and the NOW Polynya. The findings of IKAPOS are valuable for the validation of numerical models (including climate models) and will contribute to the understanding of the exchange processes over summertime Arctic polynyas and the Greenland ice sheet.

The thermohaline circulation of the world ocean is partly driven by deep water formation at high-latitudes. In the Southern Ocean, deep and bottom water formation in the marginal seas is induced by high freezing rates as generally found at coastal polynyas. Atmospheric cooling and brine-release enable the production of very cold and saline water masses. In the southwestern Weddell Sea, wide shelves allow for a strong salinification of the whole water column and the formation of High Salinity Shelf Water (HSSW). The impact of coastal polynyas on ice production and water mass formation in the southwestern Weddell Sea was studied employing the Finite Element Sea ice-Ocean Model (FESOM) of the Alfred Wegener Institute, Bremerhaven. FESOM is a coupled system of a primitive-equation, hydrostatic ocean model and a dynamic-thermodynamic sea ice model. Simulations were conducted on a global unstructured mesh with a strong focus on the southwestern Weddell Sea coastline (up to 3 km resolution). In vertical direction, the grid features 37 z-coordinate depth levels of which 6 are within the uppermost 100 m. The model runs were initialised in 1980 and forced with NCEP reanalysis data (daily resolution). The year 2008 was also simulated with higher-resolution GME and regional COSMO forcing data. For data evaluation and analysis the period 1990-2009 is used. A comparison of AMSR sea ice concentration and model results shows good accordance in spatial and temporal polynya extent. Also, calculated vertical temperature and salinity profiles agree well with CTD measurements. Our simulations feature a 20-year winter mean area of coastal polynyas of 6.7 x 10-3 km-2 (0.4% of the continental shelf area) in the southwestern Weddell Sea which is in good agreement with observations. Winter sea ice production within the coastal polynyas exceeds the ice production of the surrounding ice-covered area by a factor of 7 in the 20-year mean, so that the polynya contribution to total sea ice formation averages at about 3%. This small percentage is due to their even smaller areal percentage and the existence of leads and small polynyas in the so-called ice-covered ocean. The latter give a major contribution to sea ice production, but do not contribute to bottom water formation since they are transient elements that open, move and close dependent on the ice drift, whereas coastal polynyas are spatially fixed and open often for days, which is essential to achieve the salinification necessary for the formation of HSSW. From our simulations we derive a 20-year mean HSSW-formation of 4.2 x 10-5 km-3/season, but only 0.5 Sv thereof are exported over the shelf break, the rest stays on the shelf and is warmed and diluted during summer. The WSBW formation rate for the southwestern Weddell Sea continental shelf in our simulation is about 6.3 x 10-4 km-3/yr (2 Sv), which is on the low side but still reasonable compared to independent estimates. We conclude that in the Weddell Sea the role of coastal polynyas for sea ice production is not as big as is widely assumed, but they are indispensable for the formation of HSSW and thus for bottom water formation.

Remote sensing data show a continuous decrease of sea ice in the past 30 years. Climate models predict a further decreasing for the future. Therefore, a closer analysis of the production processes, the trend development and the regional variability is necessary. The Laptev Sea plays an important role for the Arctic sea ice budget due to a high polynya activity at the Siberian coast. The coupled ocean-sea ice model NAOSIM (North Atlantic/Arctic Ocean-Sea Ice Model) is used for the study of thermodynamic and dynamic ice production processes for the whole Arctic for the period 1990-2008. The simulation is driven by daily NCEP/NCAR data, and the horizontal resolution of the model is about 9km. Sea ice concentration from satellite data is used for the verification of model results. The model is able to reproduce the mean annual cycle and the negative trend realistically. A detailed analysis of the thermodynamic sea ice production/melt and the dynamic redistribution for different regions of the Arctic shows that the mean sea ice production of the Laptev Sea area exceeds the sea ice melt rate by 740 km3/a. That sea ice volume is transported into the central Arctic. The net ice production in the Laptev Sea is as large as the net ice production in the central Arctic north of 80° N. The Laptev Sea is found to be the largest ice producer compared to other Arctic shelf areas. In addition, the interannual variability of sea ice production in the Laptev Sea is small compared to other regions. A negative trend of sea ice in the Laptev-Sea is not found. For the entire Arctic sea ice volume decrease amounts to the average of -450 km3/a from 1990 to 2008. Studies for years with extreme sea ice anomalies show no direct connections between the sea ice production of the Laptev Sea and the sea ice volume of the entire Arctic.

Polynyas are of high research interest since these features are areas of extensive new ice formation. The calculation of accurate ice-production values requires the knowledge of polynya area and thin-ice thickness distribution. These two variables can be derived by remote sensing data. However, a cross-validation study of various remote sensing data sets indicates that the spatial resolution issue is essential for the retrieval of accurate thin-ice thickness distribution. Thus, high-resolution remote sensing data must be used. MODIS thermal-infrared data with a spatial resolution of 1 km × 1 km is appropriate for the retrieval of thin-ice thickness distribution within the polynya. The algorithm to derive thermal-infrared thin-ice thickness is improved to state-of-the-art parameterizations. The mean absolute error of thin-ice thickness is ±4.7 cm for ice below 20 cm of thickness. The thin-ice thickness maps lack full coverage due to the restriction of the algorithm to cloud-free and nighttime data. Therefore, a compositing method is applied to compute daily thin-ice thickness maps. These maps cover on average 70 % of the Laptev Sea polynya. In order to fill the remaining gaps a combined remote sensing – model approach is developed to provide a consistent time series of high-resolution thin-ice thickness maps. This data set is valuable for the retrieval of accurate ice production within polynyas.

Coastal polynyas are frequent in many polar areas even in winter and play an important role for the coupling between the components of the sea ice-ocean-atmosphere system. Locally enhanced surface exchange processes in areas of open water have important consequences for the atmosphere and ocean processes, as well as for ice formation and the associated brine release. The non-hydrostatic, mesoscale model COSMO from the German Weather Service (DWD) has been applied for the Weddell Sea Region and therefore, adjustments of the COSMO model have been made to allow for the distinctive polar conditions. Besides a new parameterization of the roughness length, we have changed the soil/ice parameterization of the Antarctic continent in the model. For a better representation of the ice shelf elevations, we incorporated the ETOPO1 ice surface dataset. Furthermore, we use a thermodynamic sea ice model, which allows for a realistic treatment of the sea ice-atmosphere interactions. Remote sensing data from the passive microwave sounder AMSR-E was used to derive a high resolution, daily sea ice coverage for the model simulations. Results for several case studies show the improvements by the model adaptions. Whereas the model simulations show very good agreement with measurements in topographically homogeneous regions, larger differences occur with measurements in topographically heterogeneous regions. With these model adaptions and a 2 step nesting, we have produced a high-resolution COSMO dataset (15 km and 5 km) for one winter season (6 months), to provide forcing data for the high resolution sea ice-ocean model FESOM. The benefit of high-resolution forcing data for the sea ice-ocean model FESOM is investigated. Due to the higher spatial and temporal resolution, a more realistic simulation of polynya dynamics is expected, compared to the common forcing with global reanalyzes, like NCEP or GME.

It is estimated that a considerable fraction of new ice formation on Arctic shelf areas takes place in the Laptev Sea polynyas. However, the different studies reveal strong discrepancies in ice production rates. For an accurate monitoring of surface heat loss and hence, ice production within polynyas it is important to know the thin ice distribution within the polynya. We use an established thin-ice algorithm with several modifications to retrieve the thin ice thickness distribution up to 50 cm based on MODIS ice surface temperatures and atmospheric data from model simulations. We verify the MODIS ice surface temperatures with a data set measured during a field campaign in the Laptev Sea. For the calculation of thin ice thicknesses we use NCEP reanalyses, GME analyses and COSMO simulations in comparison as different atmospheric forcing data. We find that from the several atmospheric variables the air temperature at 2 m has the greatest impact on the ice thickness calculation. At ice thicknesses above 20 cm the algorithm responds sensitively to errors in the atmospheric data. In regions of very thin ice the errors in the atmospheric data are masked due to larger temperature differences between surface and atmosphere. However, a reliable atmospheric data set is necessary for the calculation of accurate thin ice thicknesses.

Polynyas in the Laptev Sea are examined with respect to recurrence and inter-annual wintertime ice production. We use a polynya classification method based on passive microwave satellite data to derive daily polynya area from long-term sea-ice concentrations. This provides insight into the spatial and temporal variability of open water and thin ice regions on the Laptev Sea shelf. Using thermal infrared satellite data to derive an empirical thin ice distribution within the thickness range from 0 to 20 cm, we calculate daily average surface heat loss and the resulting wintertime ice formation within the Laptev Sea polynyas between 1979 and 2008 using NCEP reanalysis data as atmospheric forcing. Results indicate that previous studies significantly overestimate the contribution of polynyas to the ice production in the Laptev Sea. Average wintertime ice production in polynyas amounts to approximately 55 km3 +- 27% and is mostly determined by the polynya area, wind speed and associated large-scale circulation patterns. No trend in ice production could be detected in the period from 1979/80 to 2007/08.

During June 2010, the aircraft based experiment IKAPOS was performed in northwestern Greenland. The main goals were studies of the summertime katabatic wind system and of the atmospheric boundary layer (ABL) over the North Water Polynya (NOW). Katabatic winds play a key role in exchange processes of energy and momentum between the atmosphere and the underlying surface over the ice sheet of Greenland. During summer, cooling of the boundary layer and hence the katabatic forcing is much less than during winter, but strong winds can occur during appropriate synoptic forcing. On the other hand, the NOW represents one of the largest polynyas of the Arctic, and the air-sea interaction over the NOW has important consequences for ocean processes, ice formation, gas exchange and biology. The present study is based on aicraft measurements in the ABL using the research aircraft POLAR 5 of Alfred-Wegner-Institute (AWI, Bremerhaven). In order to study the turbulence structure and 3D spatial structures of mean quantities POLAR 5 was instrumented with a turbulence measurement system collecting data on a nose boom sampling at a rate of 100 Hz, additional basic meteorological equipment, radiation and surface temperature sensors, laser altimeter, and photo and video cameras. For different synoptic situations four flights over the NOW and one flight each over the Humboldt and the Steenstrup Glaciers were performed. Over the glaciers, katabatic winds with up to 16 m/s wind speed were found. Over the NOW, a stable, but fully turbulent ABL was present during conditions of strong and relatively warm synoptically induced northerly winds. Strong surface inversions were found in the lowest 100 m – 200 m agl. As a consequence of channeling effects at Smith Sound a well-pronounced low-level jet with wind speed maxima of more than 20 m/s was detected. Thus wind-induced sea-ice export from the Nares Strait is considerably increased.

Coastal polynyas are areas in the ice-covered ocean from which the sea-ice cover has been mechanically removed, primarily by winds. They are areas of enhanced exchange processes between ocean and atmosphere. The increased heat flux allows for exceptionally high freezing rates, which lead to locally increased brine-rejection. In the southwestern Weddell Sea, wide continental shelves and a weak exchange with the open ocean provide conditions that allow for substantial salinity enrichment, forming the cold and saline High Salinity Shelf Water (HSSW), which is the densest water mass in the region. HSSW is one of the ingredients of Weddell Sea Bottom Water (WSBW) and is thus essential for the formation of Antarctic Bottom Water, which covers large parts of the World Ocean’s abyss. Thus, production rates of HSSW and WSBW are of crucial importance in the ocean’s global thermohaline circulation. To study the influence of coastal polynyas on ice production and water mass formation in the southwestern Weddell Sea, we performed simulations using the Finite Element Sea ice-Ocean Model (FESOM) of the Alfred Wegener Institute, Bremerhaven. FESOM is a coupled system of a primitive-equation, hydrostatic ocean model and a dynamic-thermodynamic sea-ice model. Simulations were conducted on a global unstructured mesh, focussing on the southwestern Weddell Sea coastline with up to 3 km resolution. In vertical direction, the grid features 37 z-coordinate depth levels of which 6 are within the uppermost 100 m. The model runs were initialised in 1980 and forced with NCEP daily reanalysis data. In addition, a hindcast for the year 2008 was computed with GME 6-hourly data forcing. For the winter period 2008, the (hourly) output from the high-resolution regional atmosphere model COSMO of the University Trier was nested into the GME fields, covering the area of the western Weddell Sea. For data evaluation and analysis the period 1990-2009 is used. A comparison of model results to AMSR sea ice concentration shows good agreement in spatial and temporal polynya extent. Also, simulated vertical temperature and salinity profiles agree well with CTD measurements. The total area of coastal polynyas is very small compared to the area of the Weddell Sea continental shelf. Winter sea ice production within the coastal polynyas, however, exceeds the ice production of the surrounding ice-covered area by a factor of 8 in the 20-year mean, so that the polynya contribution to total sea ice formation is always larger than their areal fraction. When looking at ice production, it should be kept in mind that also in the so-called ice-covered ocean, leads and small polynyas exist with an areal fraction of typically 5 %, which integrates to a total area that is much larger than the total area of coastal polynyas - but consists of small and transient elements. Thus this "fractal polynya" in the offshore Weddell Sea yields a major contribution to sea ice production, but does not contribute to bottom water formation, whereas coastal polynyas are spatially coherent for days or even weeks, which is essential to achieve the necessary salinity enrichment. Only in coastal polynyas and directly adjoining areas does surface salinity exceed 34.65, which is the defining minimum salinity for HSSW. From our simulations we derive a formation rate of 4.2 x 10-5 km-3/yr (13 Sv) of HSSW as a 20-year mean, with peak formation rates of 3 x 10-5 km-3 /month (116 Sv) in the autumn months. The WSBW formation rate in our model was found to be 6.3 x 10-4 km-3/yr (2 Sv) which is on the low side although not unrealistic when compared to observation-based estimates.

The North Water polynya (NOW) forms recurrently between Ellesmere Island and northwest Greenland and is one of the largest and therefore most important polynyas in the northern hemisphere. It is considered to have a huge impact on atmospheric, biological and ocean processes. An investigation of polynya dynamics between 1996 and 2009/2010 has been performed using sea ice concentration (SIC) products from the passive microwave sensors SSM/I and AMSR-E. From SIC data we derived the total polynya area (POLA, SIC threshold of

The aircraft based experiment IKAPOS (Investigation of Katabatic winds and POlynyas during Summer) was performed in June 2010 over northwestern Greenland. The main goals were studies of the summertime katabatic wind system and of the atmospheric boundary layer (ABL) over the North Water Polynya (NOW). IKAPOS is mainly based on measurements using research aircraft POLAR 5 of Alfred-Wegner-Institute (AWI, Bremerhaven). In order to study the turbulence structure and 3D spatial structures of mean quantities, POLAR 5 was instrumented with a turbulence measurement system collecting data on a nose boom sampling at a rate of 100 Hz, additional basic meteorological equipment, radiation and surface temperature sensors, laser altimeter, and photo and video cameras. Katabatic winds play a key role in exchange processes of energy and momentum between the atmosphere and the underlying surface over the ice sheet of Greenland. During summer, cooling of the boundary layer and hence the katabatic forcing is much less than during winter, but strong winds can occur under appropriate synoptic forcing. On the other hand, the NOW represents one of the largest polynyas of the Arctic, and the air-sea interaction over the NOW has important consequences for ocean processes, ice formation, gas exchange and biology. For different synoptic situations four flights over the NOW and one flight each over the Humboldt and the Steenstrup Glaciers were performed. Over the glaciers, katabatic winds with up to 14 m/s wind speed were found. Over the NOW, a stable, but fully turbulent ABL was present during conditions of strong and relatively warm synoptically induced northerly winds. Strong surface inversions were found in the lowest 100 m – 200 m agl. As a consequence of channeling effects at Smith Sound a well-pronounced low-level jet with wind speed maxima of more than 20 m/s was detected. Thus wind-induced sea-ice export from the Nares Strait is considerably increased.

Processes of the exchange of energy and momentum at the sea ice-ocean-atmosphere interface are key processes for the polar climate system. Heat and moisture fluxes are strongly modulated by open water fractions associated with polynyas, having important consequences for the atmosphere, ocean processes, ice formation, brine release, gas exchange and biology. Our paper aims at the study of atmospheric processes forcing and maintaining polynyas in the Laptev Sea of the Siberian Arctic. This region is known as being a highly productive area for the formation of new ice throughout the winter season. We study polynya processes using passive satellite remote sensing data, high-resolution (5km) sea-ice/ocean and atmospheric models, as well as in-situ data obtained during experimental studies in that area. Passive microwave sensor data (SSM/I, AMSR) are used together with atmospheric reanalysis to characterize the long-term spatiotemporal characteristics of polynya events. A special focus lies on the detection of thin ice in polynya areas, which is studied using thermal infrared data (MODIS, AVHRR). Thin ice statistics combined with microwave data allows for estimations of ice production rates for the last decades. The NWP model COSMO is used together with the sea-ice/ocean model FESOM to study polynya dynamics. Model simulations are validated using satellite data and in-situ measurements from two campaigns in the Laptev Sea area.

Fernerkundungsdaten haben eine kontinuierliche Abnahme des Meereises in den vergangenen 30 Jahren gezeigt, Klimamodelle prognostizieren eine anhaltende Abnahme für die Zukunft. Dies erfordert eine genauere Analyse der verursachenden Prozesse, der Trendentwicklung und der regionalen Variabilität. Dabei spielt die Laptev-See in der sibirischen Arktis eine bedeutende Rolle, da es hier, bedingt durch eine große Polynja-Aktivität, zur vermehrten Eisproduktion kommt. Zur näheren Untersuchung der verursachenden thermodynamischen und dynamischen Prozesse nutzen wir eine mit täglichen NCEP/NCAR-Daten angetriebene Simulation mit dem gekoppelten Ozean-Meereismodell NAOSIM (North Atlantic/Arctic Ocean-Sea Ice Model) von 1990-2008 mit 0.08° Auflösung. Aufgrund seiner realitätsnahen Wiedergabe des mittleren Jahresgangs und des negativen Trends der Eisbedeckung ist dieses Modell für die Auswertung gut geeignet. Die getrennte Analyse der thermodynamischen Eisproduktion bzw. Eisschmelze und der dynamischen Umverteilung für die gesamte Arktis bestätigt, dass im Bereich der Laptev-See die Eisproduktion im Mittel 850km3/a größer ist als die Eisschmelze. Dieses Eis wird von der Laptev-See in die zentrale Arktis exportiert. In der gesamten Arktis nimmt das Eisvolumen im Mittel um -450km3/a von 1990-2008 ab. usammenhänge zwischen der Eisproduktion der Laptev-See und des Eisvolumens der Arktis werden mittels einer Zeitreihenananlyse untersucht. Die Entstehungsgründe für Extremjahre (Bsp.: Minimum 2007, Maximum 1996) werden aufgezeigt und ihre regionalen Folgen in der Arktis diskutiert.

The Laptev Sea polynyas play a key role for the shelf areas of the Siberian Arctic due to their impact on ice production. Changes in polynya dynamics result in modified fluxes of energy, momentum and matter in the atmosphere-ocean-sea ice system. An improved understanding and quantification of polynya effects in the Laptev Sea can be achieved by high-resolution sea ice-ocean models. Here we use the well-established Finite Element Sea Ice-Ocean Model FESOM (5 km x 5 km) (AWI Bremerhaven). It consists of a hydrostatic primitive-equation ocean model and a dynamic-thermodynamic sea ice model. In our study the model is forced by 6-hourly GME analyses (0.5° x 0.5°), daily and 6-hourly NCEP/NCAR reanalyses (2.5° x 2.5°) and hourly COSMO data (5 km x 5 km) to investigate a polynya event during the TRANSDRIFT winter experiment 2008. The input data consists of 10 m-wind, 2 m-temperature and specific humidity, total cloudiness and precipitation rate. In order to test the quality of the forcing data, comparisons with in-situ have been performed. They show shortcomings of the atmospheric analyses model data with respect to the daily course of the temperature, but very good agreement for the wind. The opening process of a main polynya event on 29 April 2008 is represented with all atmospheric forcing fields (except the daily NCEP data) in a similarly good way. However, there are differences in direction and velocity of the icedrift and in the location and development of the polynyas. Small-scale structures are best represented by applying the high-resolution COSMO data. The maximum sensible heat flux is 220 W/m2, the maximum latent heat flux is 120 W/m2, the maximum advective ice thickness reduction is 5 cm/h and the maximum thermal ice thickness production is 5 mm/h.

Sea-ice concentrations in the Laptev Sea simulated by the coupled North Atlantic – Arctic Ocean – Sea-Ice Model (NAOSIM) and Finite Element Sea-Ice Ocean Model (FESOM)are verified using sea-ice concentrations from AMSR-E satellite data and a polynya classification method for winter 2007/08. Simulated sea-ice fields from different model runs are compared with emphasis on the impact of an integrated fast-ice mask. Sea-ice models are not able to simulate polynyas realistically when used in their operational versions. Without fast ice, our investigations indicate that the simulation of large leads and smoothed sea-ice concentration fields compensates the absence of the polynyas. After implementation of a fast-ice mask the polynya location is realistically simulated, but the total open water area is largely overestimated. The study shows that further model improvements are necessary in order to achieve the important step from the simulation of large-scale features in the Arctic towards a more detailed simulation of smaller-scaled features (here polynyas) in an Arctic shelf sea.

Previous versions of the numerical weather prediction model COSMO (Consortium for Small-Scale Modeling) have used a constant sea ice surface temperature, but observations show a high degree of variability on sub-daily time-scales. To account for this, we have implemented a thermodynamic sea ice module in COSMO and performed simulations at a resolution of 15 km and 5 km for the Laptev Sea area in April 2008. Temporal and spatial variability of surface and 2m air temperature are verified by four automatic weather stations deployed along the edge of the West New Siberian Polynya during the TRANSDRIFT XIII-2 expedition and by surface temperature charts derived from MODIS satellite data. A remarkable agreement between the new model results and these observations demonstrates that the implemented sea ice module can be applied for short range simulations. Our COSMO simulations provide a high resolution and high quality atmospheric data set for the Laptev Sea for the period 14 to 30 April 2008. Based on this data set, we derive a mean total sea ice production rate of 0.53 km3/day for all Laptev Sea polynyas. Our results indicate that ice production in Laptev Sea polynyas has been overestimated in previous studies.

Die Laptev-See Polynjen spielen eine Schlüsselrolle für den arktischen Meereishaushalt aufgrund der starken Eisproduktion und des windgetriebenen Eisexportes von der Laptev-See in die zentrale Arktis. Für dieWinter 2007 bis 2009 haben wir mit dem Ozean-Meereismodell FESOM des Alfred-wegener Instituts mehrwöchige Simulationen für die Laptev-See mit einer horizontalen Auflösung von 5 km durchgeführt. Vergleiche mit dem aus AMSRE-Daten täglich abgeleitetem Eisbedeckungsgrad zeigen für den April 2008 unter Vorgabe einer Festeismaske grundsätzlich eine Übereinstimmung zwischen Simulation und Beobachtung bzgl. des Öffnen und Schließen der Polynjen. Im Dezember 2007 und im März 2009 hingegen sind die Polynjaflächen in den Simulationen wesentlich größer als die der AMSR-Daten. Aus MODIS Satellitendaten abgeleitete Dünneisdicken zeigen, dass die Größe der simulierten Polynjen realitätsnah ist, die Polynjen aber aufgrund einer dünnen Eisbedeckung nicht von den AMSR-E-Daten erfasst werden. Allerdings sind die Polynjen in den bisherigen FESOM-Simulationen nicht homogen mit dünnem Eis bedeckt, sondern teilweise mit dickem Eises und teilweise eisfrei. Dafür haben wir die Parametrisierung des Schließens von Meereis so modifiziert, dass bei starker Eisbildungsrate (> 10 cm/d) nicht nur Konvergenz sondern auch Eisbildung den Anteil offenen Wassers reduziert. Simulationen mit der geänderten Parametrisierung zeigen, dass damit Dünneisdicken in den Laptev-See Polynjen realitätsnah simuliert werden.

Synoptic weather systems in the Arctic are studied using a track algorithm based on a feature-tracking method. The tracking is performed with the 850 hPa relative vorticity field derived from the NCEP/NCAR reanalysis dataset for the winter seasons 1978-2007. A climatology of synoptic systems is calculated for the whole Arctic. A low cyclone track density is found for eastern Siberia and the Pacific part of the Arctic. Cyclonic and anticyclonic track density decreases from the first half to the second half of the 30-year investigation period in large parts of the Arctic. A detailed study is performed for the Laptev Sea area of the Siberian Arctic, whis is an active area for sea ice production in flaw polynyas. The reaction of the Laptev polynya system to tracks of cyclones and anticyclones is investigated by selecting favourate track directions. The Anabar-Lena Polynya is affected by cyclones moving eastwards across the Laptev Sea. These cyclones cause an opening of the polynya on the day before the cyclone passage and a closing on the day after. The West-New-Siberian Polynya (WNS) is affected by cyclone tracks moving northwards along the western flank of the Laptev Sea. The cyclones mainly have an opening impact one and two days before the cyclone passage. For the WNS polynya, anticyclones passing from the east have the largest effect.

Die Polynjen der Laptev-See nehmen eine Schlüsselrolle in den Schelfgebieten der sibirischen Arktis ein. Hier wird ein beachtlicher Teil des Meereisvolumens im arktischen Ozean gebildet. Zur Simulation der Dynamik der Polynjen und Quantifizierung der Eisproduktion verwenden wir das numerische Meereis-Ozean-Modell FESOM (Finite Element Sea Ice-Ocean-Model) (AWI Bremerhaven). In den bisherigen Simulationen wurde das FESOM mit täglichen NCEP (National Centers for Environmental Prediction) Daten angetrieben. Für den 1. April bis 9. Mai 2008 untersuchen wir den Einfluss von folgenden verschiedenen Antriebsdaten: Tägliche und 6-stündliche NCEP/DOE (Department of Energy) Reanalysen 2 (1.875 ̊ x 1.875 ̊), 6-stündliche NCEP/NCAR (National Centers for Atmospheric Research) Reanalysen 1 (2.5 ̊ x 2.5 ̊), 6-stündliche Analysen des GME (Globalmodell des Deutschen Wetterdienstes) (0.5 ̊ x 0.5 ̊) und hoch aufgelöste stündliche COSMO (Consortium for Small-Scale Modelling) Daten (5 km x 5 km). Vergleiche mit In-situ-Messungen des TRANSDRIFT XIII-2 Experiments 2008 zeigen, dass der Wind von allen atmosphärischen Antriebsdaten realistisch über dem Festeis wiedergegeben wird. Mit Ausnahme der FESOM-Simulationen mit täglichen NCEP-Daten werden die Öffnungs- und Schließvorgänge der Polynjen in guter Übereinstimmung mit AMSR-E (Advanced Microwave Scanning Radiometer - Earth Observing System) Produkten simuliert. Allerdings bestehen beträchtliche Diskrepanzen zwischen den Meereisproduktionsraten der unterschiedlichen Simulationen. Um die Eisproduktion in Polynjen zu quantifizieren sind stündliche, hoch aufgelöste atmosphärische Daten notwendig.

Wie wirken sich Prozesse der Wechselwirkung zwischen Atmosphäre, Meereis und Ozean entlang der Küstenränder der Laptev-See aus? Was passiert, wenn Meereis durch Wind wegtransportiert wird? Von der Expeditionsbasis in der sibirischen Hafenstadt Tiksi, starteten Wissenschaftler mit dem Hubschrauber in die Eiscamps auf dem Meereis. Vom Festeis am Rand der Polynja wurden zahlreiche meteorologische und ozeanographische Messungen durchgeführt sowie hydrologische und biologische Proben aus verschiedenen Wassertiefen der Laptev-See genommen. Kurzzeit-Meeres- bodenobservationen nahmen während der gesamten Dauer der Expedition kontinuierlich ozeano- graphische Daten auf.

Data processing methods; Energy balance closure; Intra-annual temporal dynamics of ET; Tables S1-S2; Figures S1-S2

Data processing methods; Energy balance closure; Intra-annual temporal dynamics of ET; Tables S1-S2; Figures S1-S2

Coastal polynyas in the southwestern Weddell Sea were simulated using the Finite Element Sea Ice-Ocean Model forced with NCEP/NCAR Reanalysis data. The period 1990-2009 was used in analysis. Also, shorter model runs forced with GME and COSMO model data were branched off and compared. Depending on the region, the 20-year mean winter heat flux to the atmosphere is 310-510 W/m^2, whereof 50-60 W/m^2 are supplied by cooling the ocean and the remainder induces a sea ice production of 7-13 cm/(d m^2). The coastal polynyas (0.6% of the area) contribute 11% to the southwestern Weddell Sea sea ice production. In the mean 2 Sv High Salinity Shelf Water are exported from the continental shelves. Different forcing data sets can cause substantial disparities in the regional results.