The Arctic Ocean's Wandel Sea is the easternmost sector of the Last Ice Area, where thick, old sea ice is expected to endure longer than elsewhere. Nevertheless, in August 2020 the area experienced record-low sea ice concentration. Here we use satellite data and sea ice model experiments to determine what caused this record sea ice minimum. In our simulations there was a multi-year sea-ice thinning trend due to climate change. Natural climate variability expressed as wind-forced ice advection and subsequent melt added to this trend. In spring 2020, the Wandel Sea had a mixture of both thin and — unusual for recent years — thick ice, but this thick ice was not sufficiently widespread to prevent the summer sea ice concentration minimum. With continued thinning, more frequent low summer sea ice events are expected. We suggest that the Last Ice Area, an important refuge for ice-dependent species, is less resilient to warming than previously thought.

We investigate how sea ice decline in summer and warmer ocean and surface temperatures in winter affect sea ice growth in the Arctic. Sea ice volume changes are estimated from satellite observations during winter from 2002 to 2019 and partitioned into thermodynamic growth and dynamic volume change. Both components are compared to validated sea ice-ocean models forced by reanalysis data to extend observations back to 1980 and to understand the mechanisms that cause the observed trends and variability. We find that a negative feedback driven by the increasing sea ice retreat in summer yields increasing thermodynamic ice growth during winter in the Arctic marginal seas eastward from the Laptev Sea to the Beaufort Sea. However, in the Barents and Kara Seas, this feedback seems to be overpowered by the impact of increasing oceanic heat flux and air temperatures, resulting in negative trends in thermodynamic ice growth of –2 km³month^-1yr^-1 on average over 2002–2019 derived from satellite observations.

Human-induced atmospheric composition changes cause a radiative imbalance at the top of the atmosphere which is driving global warming. This Earth energy imbalance (EEI) is the most critical number defining the prospects for continued global warming and climate change. Understanding the heat gain of the Earth system — and particularly how much and where the heat is distributed — is fundamental to understanding how this affects warming ocean, atmosphere and land; rising surface temperature; sea level; and loss of grounded and floating ice, which are fundamental concerns for society. This study is a Global Climate Observing System (GCOS) concerted international effort to update the Earth heat inventory and presents an updated assessment of ocean warming estimates as well as new and updated estimates of heat gain in the atmosphere, cryosphere and land over the period 1960–2018. The study obtains a consistent long-term Earth system heat gain over the period 1971–2018, with a total heat gain of 358±37 ZJ, which is equivalent to a global heating rate of 0.47±0.1 W m^-2. Over the period 1971–2018 (2010–2018), the majority of heat gain is reported for the global ocean with 89 % (90 %), with 52 % for both periods in the upper 700 m depth, 28 % (30 %) for the 700–2000 m depth layer and 9 % (8 %) below 2000 m depth. Heat gain over land amounts to 6 % (5 %) over these periods, 4 % (3 %) is available for the melting of grounded and floating ice, and 1 % (2 %) is available for atmospheric warming. Our results also show that EEI is not only continuing, but also increasing: the EEI amounts to 0.87±0.12 W m^-2 during 2010–2018. Stabilization of climate, the goal of the universally agreed United Nations Framework Convention on Climate Change (UNFCCC) in 1992 and the Paris Agreement in 2015, requires that EEI be reduced to approximately zero to achieve Earth's system quasi-equilibrium. The amount of CO2 in the atmosphere would need to be reduced from 410 to 353 ppm to increase heat radiation to space by 0.87 W m^-2, bringing Earth back towards energy balance. This simple number, EEI, is the most fundamental metric that the scientific community and public must be aware of as the measure of how well the world is doing in the task of bringing climate change under control, and we call for an implementation of the EEI into the global stocktake based on best available science. Continued quantification and reduced uncertainties in the Earth heat inventory can be best achieved through the maintenance of the current global climate observing system, its extension into areas of gaps in the sampling, and the establishment of an international framework for concerted multidisciplinary research of the Earth heat inventory as presented in this study. This Earth heat inventory is published at the German Climate Computing Centre.

The sea ice-albedo feedback (SIAF) is the product of the ice sensitivity (IS), that is, how much the surface albedo in sea ice regions changes as the planet warms, and the radiative sensitivity (RS), that is, how much the top-of-atmosphere radiation changes as the surface albedo changes. We demonstrate that the RS calculated from radiative kernels in climate models is reproduced from calculations using the “approximate partial radiative perturbation” method that uses the climatological radiative fluxes at the top of the atmosphere and the assumption that the atmosphere is isotropic to shortwave radiation. This method facilitates the comparison of RS from satellite-based estimates of climatological radiative fluxes with RS estimates across a full suite of coupled climate models and, thus, allows model evaluation of a quantity important in characterizing the climate impact of sea ice concentration changes. The satellite-based RS is within the model range of RS that differs by a factor of 2 across climate models in both the Arctic and Southern Ocean. Observed trends in Arctic sea ice are used to estimate IS, which, in conjunction with the satellite-based RS, yields an SIAF of 0.16 ± 0.04 W m^-2 K^-1. This Arctic SIAF estimate suggests a modest amplification of future global surface temperature change by approximately 14% relative to a climate system with no SIAF. We calculate the global albedo feedback in climate models using model-specific RS and IS and find a model mean feedback parameter of 0.37 W m^-2 K^-1, which is 40% larger than the IPCC AR5 estimate based on using RS calculated from radiative kernel calculations in a single climate model.

Over the past 40 years, the Arctic sea ice minimum in September has declined. The period between 2007 and 2012 showed accelerated melt contributed to the record minima of 2007 and 2012. Here, observational and model evidence shows that the changes in summer sea ice since the 2000s reflect a continuous anthropogenically forced melting masked by interdecadal variability of Arctic atmospheric circulation. This variation is partially driven by teleconnections originating from sea surface temperature (SST) changes in the east-central tropical Pacific via a Rossby wave train propagating into the Arctic [herein referred to as the Pacific–Arctic teleconnection (PARC)], which represents the leading internal mode connecting the pole to lower latitudes. This mode has contributed to accelerated warming and Arctic sea ice loss from 2007 to 2012, followed by slower declines in recent years, resulting in the appearance of a slowdown over the past 11 years. A pacemaker model simulation, in which we specify observed SST in the tropical eastern Pacific, demonstrates a physically plausible mechanism for the PARC mode. However, the model-based PARC mechanism is considerably weaker and only partially accounts for the observed acceleration of sea ice loss from 2007 to 2012. We also explore features of large-scale circulation patterns associated with extreme melting periods in a long (1800 yr) CESM preindustrial simulation. These results further support that remote SST forcing originating from the tropical Pacific can excite significant warm episodes in the Arctic. However, further research is needed to identify the reasons for model limitations in reproducing the observed PARC mode featuring a cold Pacific–warm Arctic connection.

The Arctic Ocean's oldest and thickest sea ice lies along the ~2,000 km arc from the western Canadian Arctic Archipelago to the northern coast of Greenland. Climate models suggest that this region will be the last to lose its perennial ice cover, thus providing an important refuge for ice‐dependent species. However, remarkably little is known about the climate or characteristics of the sea ice in this remote and inhospitable region. Here, we use the Pan‐Arctic Ice Ocean Modeling and Assimilation System model to show that the ice cover in the region is very dynamic, with changes occurring at a rate twice that of the Arctic Ocean as a whole. However, there are some differences in the changing nature of the ice cover between the eastern and western regions of the Last Ice Area, which include different timing of the annual minimum in ice thickness as well as distinct ice motion patterns associated with ice thickness extrema.

PIOMAS-20C, an Arctic sea ice reconstruction for 1901–2010, is produced by forcing the Pan-Arctic Ice Ocean Modeling and Assimilation System (PIOMAS) with ERA-20C atmospheric data. ERA-20C performance over Arctic sea ice is assessed by comparisons with measurements and data from other reanalyses. ERA-20C performs similarly with respect to the annual cycle of downwelling radiation, air temperature, and wind speed compared to reanalyses with more extensive data assimilation such as ERA-Interim and MERRA. PIOMAS-20C sea ice thickness and volume are then compared with in situ and aircraft remote sensing observations for the period of ~1950–2010. Error statistics are similar to those for PIOMAS. We compare the magnitude and patterns of sea ice variability between the first half of the twentieth century (1901–40) and the more recent period (1980–2010), both marked by sea ice decline in the Arctic. The first period contains the so-called early-twentieth-century warming (ETCW; ~1920–40) during which the Atlantic sector saw a significant decline in sea ice volume, but the Pacific sector did not. The sea ice decline over the 1979–2010 period is pan-Arctic and 6 times larger than the net decline during the 1901–40 period. Sea ice volume trends reconstructed solely from surface temperature anomalies are smaller than PIOMAS-20C, suggesting that mechanisms other than warming, such as changes in ice motion and deformation, played a significant role in determining sea ice volume trends during both periods.

Low‐level wind jets (LLJs) and strong surface winds are frequently observed near the sea ice edge in the presence of strong thermal contrast between open water and sea ice. Two LLJ cases near the sea ice edge in the Beaufort Sea are examined using dropsonde observations made from Seasonal Ice Zone Reconnaissance Survey flights. Ensembles of Polar Weather Research and Forecast simulations with and without sea ice demonstrate the contribution of the surface thermal contrast to the boundary layer structure, the LLJ, and surface ice edge jets. Because the surface temperature contrast only influences the lower most hundreds of meters in the atmospheric boundary layer, its contribution to the temperature gradient and wind speed at the level of the LLJ is limited. The sea ice does strengthen the LLJ by extending the LLJ northward over sea ice and increasing the maximum LLJ wind speeds by up to 13% and as much as 29% further north at a lower altitude. However, the primary reason for the observed strong winds in these two cases are the synoptic interactions between anticyclones and approaching cyclones. The effect of the surface thermal contrast on surface winds is controlled by a separate mechanism. The cold and stable boundary layer over sea ice prevents the momentum transport from the LLJ to the surface. This leads to weaker surface winds over sea ice and confines the strong surface winds close to the sea ice edge. This mechanism contributes to the frequent occurrence of the surface "ice edge jets."

The relative contribution and physical drivers of internal variability in recent Arctic sea ice loss remain open questions, leaving up for debate whether global climate models used for climate projection lack sufficient sensitivity in the Arctic to climate forcing. Here, through analysis of large ensembles of fully coupled climate model simulations with historical radiative forcing, we present an important internal mechanism arising from low-frequency Arctic atmospheric variability in models that can cause substantial summer sea ice melting in addition to that due to anthropogenic forcing. This simulated internal variability shows a strong similarity to the observed Arctic atmospheric change in the past 37 years. Through a fingerprint pattern matching method, we estimate that this internal variability contributes to about 40–50% of observed multi-decadal decline in Arctic sea ice. Our study also suggests that global climate models may not actually underestimate sea ice sensitivities in the Arctic, but have trouble fully replicating an observed linkage between the Arctic and lower latitudes in recent decades. Further improvements in simulating the observed Arctic–global linkage are thus necessary before the Arctic’s sensitivity to global warming in models can be quantified with confidence.

During late February and early March 2018, an unusual polynya was observed off the north coast of Greenland. This period was also notable for the occurrence of a sudden stratospheric warming. Here we use satellite and in situ data, a reanalysis and an ice‐ocean model to document the evolution of the polynya and its synoptic forcing. We show that its magnitude was unprecedented and that it was associated with the transient response to the sudden stratospheric warming leading to anomalous warm southerly flow in north Greenland. Indeed, regional wind speeds and temperatures were the highest during February going back to the 1960s. There is evidence that the thinning sea ice has increased its wind‐driven mobility. However, we show that the polynya would have developed under thicker ice conditions representative of the late 1970s and that even with the predicted trend toward thinner sea ice, it will only open during enhanced southerly flow.

A melt pond (MP) distribution equation has been developed and incorporated into the Marginal Ice‐Zone Modeling and Assimilation System to simulate Arctic MPs and sea ice over 1979–2016. The equation differs from previous MP models and yet benefits from previous studies for MP parameterizations as well as a range of observations for model calibration. Model results show higher magnitude of MP volume per unit ice area and area fraction in most of the Canada Basin and the East Siberian Sea and lower magnitude in the central Arctic. This is consistent with Moderate Resolution Imaging Spectroradiometer observations, evaluated with Measurements of Earth Data for Environmental Analysis (MEDEA) data, and closely related to top ice melt per unit ice area. The model simulates a decrease in the total Arctic sea ice volume and area, owing to a strong increase in bottom and lateral ice melt. The sea ice decline leads to a strong decrease in the total MP volume and area. However, the Arctic‐averaged MP volume per unit ice area and area fraction show weak, statistically insignificant downward trends, which is linked to the fact that MP water drainage per unit ice area is increasing. It is also linked to the fact that MP volume and area decrease relatively faster than ice area. This suggests that overall the actual MP conditions on ice have changed little in the past decades as the ice cover is retreating in response to Arctic warming, thus consistent with the Moderate Resolution Imaging Spectroradiometer observations that show no clear trend in MP area fraction over 2000–2011.

The winter Arctic atmosphere is under the influence of two very different circulation systems: extratropical cyclones travel along the primary North Atlantic storm track from Iceland toward the eastern Arctic, while the western Arctic is characterized by a quasi‐stationary region of high pressure known as the Beaufort High. The winter (January through March) of 2017 featured an anomalous reversal of the normally anticyclonic surface winds and sea ice motion in the western Arctic. This reversal can be traced to a collapse of the Beaufort High as the result of the intrusion of low‐pressure systems from the North Atlantic, along the East Siberian Coast, into the Arctic Basin. Thin sea ice as the result of an extremely warm autumn (October through December) of 2016 contributed to the formation of an anomalous thermal low over the Barents Sea that, along with a northward shift of the tropospheric polar vortex, permitted this intrusion. The collapse of the Beaufort High during the winter of 2017 was associated with simultaneous 2‐sigma sea level pressure, surface wind, and sea ice circulation anomalies in the western Arctic. As the Arctic sea ice continues to thin, such reversals may become more common and impact ocean circulation, sea ice, and biology.

A budget of meteoric water (MW = river runoff, net precipitation minus evaporation, and glacial meltwater) over four regions of the Arctic Ocean is constructed using a simple box model, regional precipitation-evaporation estimates from reanalysis data sets, and estimates of import and export fluxes derived from the literature with a focus on the 2003–2008 period. The budget indicates an approximate/slightly positive balance between MW imports and exports (i.e., no change in storage); thus, the observed total freshwater increase observed during this time period likely resulted primarily from changes in non-MW freshwater components (i.e., increases in sea ice melt or Pacific water and/or a decrease in ice export). Further, our analysis indicates that the MW increase observed in the Canada Basin resulted from a spatial redistribution of MW over the Arctic Ocean. Mean residence times for MW were estimated for the Western Arctic (5–7 years), Eastern Arctic (3–4 years), and Lincoln Sea (1–2 years). The MW content over the Siberian shelves was estimated (~14,000 km³) based on a residence time of 3.5 years. The MW content over the entire Arctic Ocean was estimated to be ≥ 44,000 km³. The MW export through Fram Strait consisted mostly of water from the Eastern Arctic (3237 ± 1370 km³ yr^-1) whereas the export through the Canadian Archipelago was nearly equally derived from both the Western Arctic (1182 ± 534 km³ yr^-1) and Lincoln Sea (972 ± 391 km³ yr^-1).

The Arctic has seen rapid sea-ice decline in the past three decades, whilst warming at about twice the global average rate. Yet the relationship between Arctic warming and sea-ice loss is not well understood. Here, we present evidence that trends in summertime atmospheric circulation may have contributed as much as 60% to the September sea-ice extent decline since 1979. A tendency towards a stronger anticyclonic circulation over Greenland and the Arctic Ocean with a barotropic structure in the troposphere increased the downwelling longwave radiation above the ice by warming and moistening the lower troposphere. Model experiments, with reanalysis data constraining atmospheric circulation, replicate the observed thermodynamic response and indicate that the near-surface changes are dominated by circulation changes rather than feedbacks from the changing sea-ice cover. Internal variability dominates the Arctic summer circulation trend and may be responsible for about 30–50% of the overall decline in September sea ice since 1979.

To better simulate the seasonal evolution of sea ice in the Arctic, with particular attention to the marginal ice zone, a sea ice model of the distribution of ice thickness, floe size, and enthalpy was implemented into the Pan-arctic IceOcean Modeling and Assimilation System (PIOMAS). Theories on floe size distribution (FSD) and ice thickness distribution (ITD) were coupled in order to explicitly simulate multicategory FSD and ITD distributions simultaneously. The expanded PIOMAS was then used to estimate the seasonal evolution of the Arctic FSD in 2014 when FSD observations are available for model calibration and validation.

Results indicate that the simulated FSD, commonly described equivalently as cumulative floe number distribution (CFND), generally follows a power law across space and time and agrees with the CFND observations derived from TerraSAR-X satellite images. The simulated power-law exponents also correlate with those derived using MODIS images, with a low mean bias of 2%. In the marginal ice zone, the modeled CFND shows a large number of small floes in winter because of stronger winds acting on thin, weak first-year ice in the ice edge region. In mid-spring and summer, the CFND resembles an upper truncated power law, with the largest floes mostly broken into smaller ones; however, the number of small floes is lower than in winter because floes of small sizes or first-year ice are easily melted away. In the ice pack interior there are fewer floes in late fall and winter than in summer because many of the floes are welded together into larger floes in freezing conditions, leading to a relatively flat CFND with low power-law exponents.

The simulated mean floe size averaged over all ice-covered areas shows a clear annual cycle, large in winter and smaller in summer. However, there is no obvious annual cycle of mean floe size averaged over the marginal ice zone. The incorporation of FSD into PIOMAS results in reduced ice thickness, mainly in the marginal ice zone, which improves the simulation of ice extent and yields an earlier ice retreat.

Arctic sea ice drift forecasts of 6 h – 9 days for the summer of 2014 are generated using the Marginal Ice Zone Modeling and Assimilation System (MIZMAS); the model is driven by 6 h atmospheric forecasts from the Climate Forecast System (CFSv2). Forecast ice drift speed is compared to drifting buoys and other observational platforms. Forecast positions are compared with actual positions 24 h – 8 days since forecast. Forecast results are further compared to those from the forecasts generated using an ice velocity climatology driven by multiyear integrations of the same model. The results are presented in the context of scheduling the acquisition of high-resolution images that need to follow buoys or scientific research platforms. RMS errors for ice speed are on the order of 5 km/d for 24–48 h since forecast using the sea ice model compared with 9 km/d using climatology. Predicted buoy position RMS errors are 6.3 km for 24 h and 14 km for 72 h since forecast. Model biases in ice speed and direction can be reduced by adjusting the air drag coefficient and water turning angle, but the adjustments do not affect verification statistics. This suggests that improved atmospheric forecast forcing may further reduce the forecast errors. The model remains skillful for 8 days. Using the forecast model increases the probability of tracking a target drifting in sea ice with a 10 km x 10 km image from 60 to 95% for a 24 h forecast and from 27 to 73% for a 48 h forecast.

To better describe the state of sea ice in the marginal ice zone (MIZ) with floes of varying thicknesses and sizes, both an ice thickness distribution (ITD) and a floe size distribution (FSD) are needed. In this work, we have developed a FSD theory that is coupled to the ITD theory of Thorndike et al. (1975) in order to explicitly simulate the evolution of FSD and ITD jointly. The FSD theory includes a FSD function and a FSD conservation equation in parallel with the ITD equation. The FSD equation takes into account changes in FSD due to ice advection, thermodynamic growth, and lateral melting. It also includes changes in FSD because of mechanical redistribution of floe size due to ice ridging and, particularly, ice fragmentation induced by stochastic ocean surface waves. The floe size redistribution due to ice fragmentation is based on the assumption that wave-induced breakup is a random process such that when an ice floe is broken, floes of any smaller sizes have an equal opportunity to form, without being either favored or excluded. To focus only on the properties of mechanical floe size redistribution, the FSD theory is implemented in a simplified ITD and FSD sea ice model for idealized numerical experiments. Model results show that the simulated cumulative floe number distribution (CFND) follows a power law as observed by satellites and airborne surveys. The simulated values of the exponent of the power law, with varying levels of ice breakups, are also in the range of the observations. It is found that floe size redistribution and the resulting FSD and mean floe size do not depend on how floe size categories are partitioned over a given floe size range. The ability to explicitly simulate multicategory FSD and ITD together may help to incorporate additional model physics, such as FSD-dependent ice mechanics, surface exchange of heat, mass, and momentum, and wave-ice interactions.

Sea ice thickness is a fundamental climate state variable that provides an integrated measure of changes in the high-latitude energy balance. However, observations of mean ice thickness have been sparse in time and space, making the construction of observation-based time series difficult. Moreover, different groups use a variety of methods and processing procedures to measure ice thickness, and each observational source likely has different and poorly characterized measurement and sampling errors. Observational sources used in this study include upward-looking sonars mounted on submarines or moorings, electromagnetic sensors on helicopters or aircraft, and lidar or radar altimeters on airplanes or satellites. Here we use a curve-fitting approach to determine the large-scale spatial and temporal variability of the ice thickness as well as the mean differences between the observation systems, using over 3000 estimates of the ice thickness. The thickness estimates are measured over spatial scales of approximately 50 km or time scales of 1 month, and the primary time period analyzed is 2000–2012 when the modern mix of observations is available. Good agreement is found between five of the systems, within 0.15 m, while systematic differences of up to 0.5 m are found for three others compared to the five. The trend in annual mean ice thickness over the Arctic Basin is –0.58 ± 0.07 m decade^-1 over the period 2000–2012. Applying our method to the period 1975–2012 for the central Arctic Basin where we have sufficient data (the SCICEX box), we find that the annual mean ice thickness has decreased from 3.59 m in 1975 to 1.25 m in 2012, a 65% reduction. This is nearly double the 36% decline reported by an earlier study. These results provide additional direct observational evidence of substantial sea ice losses found in model analyses.

The authors use the Polar Weather Research and Forecasting (WRF) Model to simulate atmospheric conditions during the Seasonal Ice Zone Reconnaissance Survey (SIZRS) in the summer of 2013 over the Beaufort Sea. With the SIZRS dropsonde data, the performance of WRF simulations and two forcing datasets is evaluated: the Interim ECMWF Re-Analysis (ERA-Interim) and the Global Forecast System (GFS) analysis. General features of observed mean profiles, such as low-level temperature inversion, low-level jet (LLJ), and specific humidity inversion are reproduced by all three models. A near-surface warm bias and a low-level moist bias are found in ERA-Interim. WRF significantly improves the mean LLJ, with a lower and stronger jet and a larger turning angle than the forcing. The improvement in the mean LLJ is likely related to the lower values of the boundary layer diffusion in WRF than in ERA-Interim and GFS, which also explains the lower near-surface temperature in WRF than the forcing. The relative humidity profiles have large differences between the observations, the ERA-Interim, and the GFS. The WRF simulated relative humidity closely resembles the forcings, suggesting the need to obtain more and better-calibrated humidity data in this region. The authors find that the sea ice concentrations in the ECMWF model are sometimes significantly underestimated due to an inappropriate thresholding mechanism. This thresholding affects both ERA-Interim and the ECMWF operational model. The scale of impact of this issue on the atmospheric boundary layer in the marginal ice zone is still unknown.

Arctic sea ice thickness distributions from models participating in the World Climate Research Programme Coupled Model Intercomparison Project Phase 5 (CMIP5) are evaluated against observations from submarines, aircraft and satellites. While it is encouraging that the mean thickness distributions from the models are in general agreement with observations, the spatial patterns of sea ice thickness are poorly represented in most models. The poor spatial representation of thickness patterns is associated with a failure of models to represent details of the mean atmospheric circulation pattern that governs the transport and spatial distribution of sea ice. The climate models as a whole also tend to underestimate the rate of ice volume loss from 1979 to 2013, though the multimodel ensemble mean trend remains within the uncertainty of that from the Pan-Arctic Ice Ocean Modeling and Assimilation System. Although large uncertainties in observational products complicate model evaluations, these results raise concerns regarding the ability of CMIP5 models to realistically represent the processes driving the decline of Arctic sea ice and to project the timing of when a seasonally ice-free Arctic may become a reality.

Atmospheric reanalyses depend on a mix of observations and model forecasts. In data-sparse regions such as the Arctic, the reanalysis solution is more dependent on the model structure, assumptions, and data assimilation methods than in data-rich regions. Applications such as the forcing of ice%u2013ocean models are sensitive to the errors in reanalyses. Seven reanalysis datasets for the Arctic region are compared over the 30-yr period 1981–2010: National Centers for Environmental Prediction (NCEP)–National Center for Atmospheric Research Reanalysis 1 (NCEP-R1) and NCEP–U.S. Department of Energy Reanalysis 2 (NCEP-R2), Climate Forecast System Reanalysis (CFSR), Twentieth-Century Reanalysis (20CR), Modern-Era Retrospective Analysis for Research and Applications (MERRA), ECMWF Interim Re-Analysis (ERA-Interim), and Japanese 25-year Reanalysis Project (JRA-25). Emphasis is placed on variables not observed directly including surface fluxes and precipitation and their trends. The monthly averaged surface temperatures, radiative fluxes, precipitation, and wind speed are compared to observed values to assess how well the reanalysis data solutions capture the seasonal cycles. Three models stand out as being more consistent with independent observations: CFSR, MERRA, and ERA-Interim. A coupled ice–ocean model is forced with four of the datasets to determine how estimates of the ice thickness compare to observed values for each forcing and how the total ice volume differs among the simulations. Significant differences in the correlation of the simulated ice thickness with submarine measurements were found, with the MERRA products giving the best correlation (R = 0.82). The trend in the total ice volume in September is greatest with MERRA (–4.1 ± 10³ km³ decade^-1) and least with CFSR (–2.7 ± 10³ km³ decade^-1).

Satellite records show a decline in ice extent over more than three decades, with a record minimum in September 2012. Results from the Pan-Arctic Ice-Ocean Modelling and Assimilation system (PIOMAS) suggest that the decline in extent has been accompanied by a decline in volume, but this has not been confirmed by data. Using new data from the European Space Agency CryoSat-2 (CS-2) mission, validated with in situ data, we generate estimates of ice volume for the winters of 2010/11 and 2011/12. We compare these data with current estimates from PIOMAS and earlier (2003–8) estimates from the National Aeronautics and Space Administration ICESat mission. Between the ICESat and CryoSat-2 periods, the autumn volume declined by 4291 km³ and the winter volume by 1479 km³. This exceeds the decline in ice volume in the central Arctic from the PIOMAS model of 2644 km³ in the autumn, but is less than the 2091 km³ in winter, between the two time periods.

This model study examines the impact of an intense early August cyclone on the 2012 record low Arctic sea ice extent. The cyclone passed when Arctic sea ice was thin and the simulated Arctic ice volume had already declined ~40% from the 2007–2011 mean. The thin sea ice pack and the presence of ocean heat in the near surface temperature maximum layer created conditions that made the ice particularly vulnerable to storms. During the storm, ice volume decreased about twice as fast as usual, owing largely to a quadrupling in bottom melt caused by increased upward ocean heat transport. This increased ocean heat flux was due to enhanced mixing in the oceanic boundary layer, driven by strong winds and rapid ice movement. A comparison with a sensitivity simulation driven by reduced wind speeds during the cyclone indicates that cyclone-enhanced bottom melt strongly reduces ice extent for about two weeks, with a declining effect afterwards. The simulated Arctic sea ice extent minimum in 2012 is reduced by the cyclone, but only by 0.15 x 10⁶ km² (4.4%). Thus without the storm, 2012 would still have produced a record minimum.

Results from a numerical model simulation show significant changes in the dynamic properties of Arctic sea ice during 2007–2011 compared to the 1979–2006 mean. These changes are linked to a 33% reduction in sea ice volume, with decreasing ice concentration, mostly in the marginal seas, and decreasing ice thickness over the entire Arctic, particularly in the western Arctic. The decline in ice volume results in a 37% decrease in ice mechanical strength and 31% in internal ice interaction force, which in turn leads to an increase in ice speed (13%) and deformation rates (17%). The increasing ice speed has the tendency to drive more ice out of the Arctic. However, ice volume export is reduced because the rate of decrease in ice thickness is greater than the rate of increase in ice speed, thus retarding the decline of Arctic sea ice volume. Ice deformation increases the most in fall and least in summer. Thus the effect of changes in ice deformation on the ice cover is likely strong in fall and weak in summer. The increase in ice deformation boosts ridged ice production in parts of the central Arctic near the Canadian Archipelago and Greenland in winter and early spring, but the average ridged ice production is reduced because less ice is available for ridging in most of the marginal seas in fall. The overall decrease in ridged ice production contributes to the demise of thicker, older ice. As the ice cover becomes thinner and weaker, ice motion approaches a state of free drift in summer and beyond and is therefore more susceptible to changes in wind forcing. This is likely to make seasonal or shorter-term forecasts of sea ice edge locations more challenging.

Uncertainty in the Pan-Arctic Ice-Ocean Modeling and Assimilation System (PIOMAS) Arctic sea ice volume record is characterized. A range of observations and approaches, including in situ ice thickness measurements, ICESat retrieved ice thickness, and model sensitivity studies, yields a conservative estimate for October Arctic ice volume uncertainty of 1.35 x 10^3 km^3 and an uncertainty of the ice volume trend over the 1979-2010 period of 1.0 x 10^3 km^3 decade^-1. A conservative estimate of the trend over this period is ~2.8 x 10^3 km^3 decade^-1. PIOMAS ice thickness estimates agree well with ICESat ice thickness retrievals (<0.1 m mean difference) for the area for which submarine data are available, while difference outside this area are larger. PIOMAS spatial thickness patterns agree well with ICESat thickness estimates with pattern correlations of above 0.8. PIOMAS appears to overestimate thin ice thickness and underestimate thick ice, yielding a smaller downward trend than apparent in reconstructions from observations. PIOMAS ice volume uncertainties and trends are examined in the context of climate change attribution and the declaration of record minima. The distribution of 32 year trends in a preindustrial coupled model simulation shows no trends comparable to those seen in the PIOMAS retrospective, even when the trend uncertainty is accounted for. Attempts to label September minima as new record lows are sensitive to modeling error. However, the September 2010 ice volume anomaly did in fact exceed the previous 2007 minimum by a large enough margin to establish a statistically significant new record.

Numerical experiments are conducted to project arctic sea ice responses to varying levels of future anthropogenic warming and climate variability over 2010–2050. A summer ice-free Arctic Ocean is likely by the mid-2040s if arctic surface air temperature (SAT) increases 4 deg C by 2050 and climate variability is similar to the past relatively warm two decades. If such a SAT increase is reduced by one-half or if a future Arctic experiences a range of SAT fluctuation similar to the past five decades, a summer ice-free Arctic Ocean would be unlikely before 2050. If SAT increases 4 deg C by 2050, summer ice volume decreases to very low levels (10–37% of the 1978–2009 summer mean) as early as 2025 and remains low in the following years, while summer ice extent continues to fluctuate annually. Summer ice volume may be more sensitive to warming while summer ice extent more sensitive to climate variability. The rate of annual mean ice volume decrease relaxes approaching 2050. This is because, while increasing SAT increases summer ice melt, a thinner ice cover increases winter ice growth. A thinner ice cover also results in a reduced ice export, which helps to further slow ice volume loss. Because of enhanced winter ice growth, arctic winter ice extent remains nearly stable and therefore appears to be a less sensitive climate indicator.

The minimum of Arctic sea ice extent in the summer of 2007 was unprecedented in the historical record. A coupled ice–ocean model is used to determine the state of the ice and ocean over the past 29 yr to investigate the causes of this ice extent minimum within a historical perspective. It is found that even though the 2007 ice extent was strongly anomalous, the loss in total ice mass was not. Rather, the 2007 ice mass loss is largely consistent with a steady decrease in ice thickness that began in 1987. Since then, the simulated mean September ice thickness within the Arctic Ocean has declined from 3.7 to 2.6 m at a rate of –0.57 m decade^-1. Both the area coverage of thin ice at the beginning of the melt season and the total volume of ice lost in the summer have been steadily increasing. The combined impact of these two trends caused a large reduction in the September mean ice concentration in the Arctic Ocean. This created conditions during the summer of 2007 that allowed persistent winds to push the remaining ice from the Pacific side to the Atlantic side of the basin and more than usual into the Greenland Sea. This exposed large areas of open water, resulting in the record ice extent anomaly.

The connection between sea ice variability and cloud cover over the Arctic seas during autumn is investigated by analyzing the 40-yr ECMWF Re-Analysis (ERA-40) products and the Television and Infrared Observation Satellite (TIROS) Operational Vertical Sounder (TOVS) Polar Pathfinder satellite datasets. It is found that cloud cover variability near the sea ice margins is strongly linked to sea ice variability. Sea ice retreat is linked to a decrease in low-level cloud amount and a simultaneous increase in midlevel clouds. This pattern is apparent in both data sources. Changes in cloud cover can be explained by changes in the atmospheric temperature structure and an increase in near-surface temperatures resulting from the removal of sea ice. The subsequent decrease in static stability and deepening of the atmospheric boundary layer apparently contribute to the rise in cloud level. The radiative effect of this change is relatively small, as the direct radiative effects of cloud cover changes are compensated for by changes in the temperature and humidity profiles associated with varying ice conditions.

A model study has been conducted of the unprecedented retreat of arctic sea ice in the summer of 2007. It is found that preconditioning, anomalous winds, and ice-albedo feedback are mainly responsible for the retreat. Arctic sea ice in 2007 was preconditioned to radical changes after years of shrinking and thinning in a warm climate. During summer 2007 atmospheric changes strengthened the transpolar drift of sea ice, causing more ice to move out of the Pacific sector and the central Arctic Ocean where the reduction in ice thickness due to ice advection is up to 1.5 m more than usual. Some of the ice exited Fram Strait and some piled up in part of the Canada Basin and along the coast of northern Greenland, leaving behind an unusually large area of thin ice and open water. Thin ice and open water allow more surface solar heating because of a much reduced surface albedo, leading to amplified ice melting. The Arctic Ocean lost additional 10% of its total ice mass in which 70% is due directly to the amplified melting and 30% to the unusual ice advection, causing the unprecedented ice retreat. Arctic sea ice has entered a state of being particularly vulnerable to anomalous atmospheric forcing.

We conduct experiments with an ice-ocean model to answer the question whether and to what degree unusually clear skies during the summer of 2007 contributed to the record sea ice extent minimum in the Arctic Ocean during September of 2007. Anomalously high pressure over the Beaufort Sea during summer 2007 appears associated with a strong negative cloud anomaly. This anomaly is two standard deviations below the 1980–2007 average established from a combination of two different satellite-based records. Cloud anomalies from the MODIS sensor are compared with anomalies from the NCEP/NCAR reanalysis and are found in good agreement in spatial patterns and magnitude. However, these experiments establish that the negative cloud anomaly and increased downwelling shortwave flux from June through August did not contribute substantially to the record sea ice extent minimum. This finding eliminates one aspect of the unusual weather that may have contributed to the record minimum.

Ensemble predictions of arctic sea ice in spring and summer 2008 have been carried out using an ice-ocean model. The ensemble is constructed by using atmospheric forcing from 2001 to 2007 and the September 2007 ice and ocean conditions estimated by the model. The prediction results show that the record low ice cover and the unusually warm ocean surface waters in summer 2007 lead to a substantial reduction in ice thickness in 2008. Up to 1.2 m ice thickness reduction is predicted in a large area of the Canada Basin in both spring and summer of 2008, leading to extraordinarily thin ice in summer 2008. There is a 50% chance that both the Northern Sea Route and the Northwest Passage will be nearly ice free in September 2008. It is not likely there will be another precipitous decline in arctic sea ice extent such as seen in 2007, unless a new atmospheric forcing regime, significantly different from the recent past, occurs.

How well can the extent of arctic sea ice be predicted for lead periods of up to one year? The forecast ability of a linear empirical model is explored. It uses as predictors historical information about the ocean and ice obtained from an ice–ocean model retrospective analysis. The monthly model fields are represented by a correlation-weighted average based on the predicted ice extent. The forecast skill of the procedure is found by fitting the model over subsets of the available data and then making subsequent projections using independent predictor data. The forecast skill, relative to climatology, for predictions of the observed September ice extent for the pan-arctic region is 0.77 for six months lead (from March) and 0.75 for 11 months lead (from October). The ice concentration is the most important variable for the first two months and the ocean temperature of the model layer with a depth of 200 to 270 m is most important for longer lead times. The trend accounts for 76% of the variance of the pan-arctic ice extent, so most of the forecast skill is realized by determining model variables that best represent this trend. For detrended data there is no skill for lead times of 3 months or more. The forecast skill relative to the estimate from the previous year is lower than the climate-relative skill but it is still greater than 0.45 for most lead times. Six-month predictions are also made for each month of the year and regional three-month predictions are made for 45-degree sectors. The ice-ocean model output significantly improves the predictive skill of the forecast model.

The melt season of the Arctic sea ice cover is greatly affected by the partitioning of the incident solar radiation between reflection to the atmosphere and absorption in the ice and ocean. This partitioning exhibits a strong seasonal cycle and significant interannual variability. Data in the period 1998, 2000–2004 were analyzed in this study. Observations made during the 1997–1998 SHEBA (Surface HEat Budget of the Arctic Ocean) field experiment showed a strong seasonal dependence of the partitioning, dominated by a five-phase albedo evolution. QuikSCAT scatterometer data from the SHEBA region in 1999–2004 were used to further investigate solar partitioning in summer. The time series of scatterometer data were used to determine the onset of melt and the beginning of freezeup. This information was combined with SSM/I-derived ice concentration, TOVS-based estimates of incident solar irradiance, and SHEBA results to estimate the amount of solar energy absorbed in the ice-ocean system for these years. The average total solar energy absorbed in the ice-ocean system from April through September was 900 MJ m^-2. There was considerable interannual variability, with a range of 826 to 1044 MJ m^-2. The total amount of solar energy absorbed by the ice and ocean was strongly related to the date of melt onset, but only weakly related to the total duration of the melt season or the onset of freezeup. The timing of melt onset is significant because the incident solar energy is large and a change at this time propagates through the entire melt season, affecting the albedo every day throughout melt and freezeup.

The low-level atmospheric temperature inversion is a dominant feature of the Arctic atmosphere throughout most of the year. Meteorological stations that provide radiosonde data are sparsely distributed across the Arctic, and therefore provide little information on the spatial distribution of temperature inversions. Satellite-borne sensors provide an opportunity to fill the observational gap. In this study, a 17-yr time series, 1980–96, of clear-sky temperature inversion strength during the cold season is derived from High Resolution Infrared Radiation Sounder (HIRS) data using a two-channel statistical method. The satellite-derived clear-sky inversion strength monthly mean and trends agree well with radiosonde data. Both increasing and decreasing trends are found in the cold season for different areas. It is shown that there is a strong coupling between changes in surface temperature and changes in inversion strength, but that trends in some areas may be a result of advection aloft rather than warming or cooling at the surface.

Winter and spring changes in cloudiness are compared over the arctic seas (ocean areas north of 60°N) from the TOVS (TIROS Operational Vertical Sounder) Polar Pathfinder retrievals and two separate datasets derived from the Advanced Very High Resolution Radiometer (AVHRR). All satellite products exhibit significant decreases in cloud fraction over the arctic seas during winter (December, January, February) on the order of 5% / decade. An equally striking increase in spring (March, April, May) cloudiness is evident from the TOVS Pathfinder (TPP) and the extended AVHRR Polar Pathfinder (APP-x) projects. In the Central Arctic these positive trends can be as large as 15% / decade. Surface observations from the Russian drifting meteorological stations are consistent with satellite-observed changes during the 1980s. Negative trends in spring cloudiness reported by Comiso [2003] are in conflict with these findings. Spring changes in cloudiness are associated with changes in the atmospheric circulation. These dramatic, large-scale changes may have substantial impacts on the surface energy balance.

Products from the TIROS-N Operational Vertical Sounder (TOVS) Polar Pathfinder (Path-P) data set are compared with surface measurements and other satellite remote sensing retrievals during the Surface Heat Balance of the Arctic Ocean (SHEBA) field program (October 1997 to September 1998). The comparison provides estimates of Path-P retrieval uncertainties. Results are placed in the context of the natural variability and timescales of variability to allow potential users to judge the applicability of the data set for their purpose. Results show temperature profiles to be accurate within 3 K, total column precipitable water within 2 mm annually, and surface temperature within 3 K. Uncertainties in temperature retrieval are below "within-season" variability during all times of the year. Uncertainties in water vapor retrieval during winter and summer are slightly below observed variability in those seasons but are well below during spring. Uncertainty in retrieved cloud fraction is highly dependent on the timescale of observations. Cloud fractions from the surface and satellite are well correlated (correlation coefficient > 0.7) at timescales greater than 4 days but show weaker correlation at shorter timescales. Uncertainty in TOVS-retrieved cloud fraction is less than 20% for 5-day averages. In winter, TOVS-retrieved cloud fractions are higher than those reported in standard meteorological observations but match those derived from lidar data. This supports the notion that standard meteorological observations may underestimate cloudiness in winter. Cloud-top temperatures measured from the surface (lidar/radar) are significantly different from those estimated using TOVS and Advanced Very High Resolution Radiometer (AVHRR) radiances, which highlights the fundamental and inherent dissimilarity between these two measurement techniques.