Understanding model spread in sea ice volume by attribution of model differences in seasonal ice growth and melt

Arctic sea ice is declining rapidly, but predictions of its future loss are made difficult by the large spread both in present-day and in future sea ice area and volume; hence, there is a need to better understand the drivers of model spread in sea ice state. Here we present a framework for understanding differences between modelled sea ice simulations based on attributing seasonal ice growth and melt differences. In the method presented, the net downward surface flux is treated as the principal driver of seasonal sea ice growth and melt. A system of simple models is used to estimate the pointwise effect of model differences in key Arctic climate variables on this surface flux, and hence on seasonal sea ice growth and melt. We compare three models with very different historical sea ice simulations: HadGEM2-ES, HadGEM3-GC3.1 and UKESM1.0. The largest driver of differences in ice growth / melt between these models is shown to be the ice area in summer (representing the surface albedo feedback) and the ice thickness distribution in winter (the thickness-growth feedback). Differences in snow and melt-pond cover during the early summer exert a smaller effect on the seasonal growth and melt, hence representing the drivers of model differences in both this and in the sea ice volume. In particular, the direct impacts on sea ice growth / melt of differing model parameterisations of snow area and of melt-ponds are shown to be small but non-negligible.

Melt Pond Retrieval Based on the LinearPolar Algorithm Using Landsat Data

The formation and distribution of melt ponds have an important influence on the Arctic climate. Therefore, it is necessary to obtain more accurate information on melt ponds on Arctic sea ice by remote sensing. The present large-scale melt pond products, especially the melt pond fraction (MPF), still require verification, and using very high resolution optical satellite remote sensing data is a good way to verify the large-scale retrieval of MPF products. Unlike most MPF algorithms using very high resolution data, the LinearPolar algorithm using Sentinel-2 data considers the albedo of melt ponds unfixed. In this paper, by selecting the best band combination, we applied this algorithm to Landsat 8 (L8) data. Moreover, Sentinel-2 data, as well as support vector machine (SVM) and iterative self-organizing data analysis technique (ISODATA) algorithms, are used as the comparison and verification data. The results show that the recognition accuracy of the LinearPolar algorithm for melt ponds is higher than that of previous algorithms. The overall accuracy and kappa coefficient results achieved by using the LinearPolar algorithm with L8 and Sentinel-2A (S2), the SVM algorithm, and the ISODATA algorithm are 95.38% and 0.88, 94.73% and 0.86, and 92.40%and 0.80, respectively, which are much higher than those of principal component analysis (PCA) and Markus algorithms. The mean MPF (10.0%) obtained from 80 cases from L8 data based on the LinearPolar algorithm is much closer to Sentinel-2 (10.9%) than the Markus (5.0%) and PCA algorithms (4.2%), with a mean MPF difference of only 0.9%, and the correlation coefficients of the two MPFs are as high as 0.95. The overall relative error of the LinearPolar algorithm is 53.5% and 46.4% lower than that of the Markus and PCA algorithms, respectively, and the root mean square error (RMSE) is 30.9% and 27.4% lower than that of the Markus and PCA algorithms, respectively. In the cases without obvious melt ponds, the relative error is reduced more than that of those with obvious melt ponds because the LinearPolar algorithm can identify 100% of dark melt ponds and relatively small melt ponds, and the latter contributes more to the reduction in the relative error of MPF retrieval. With a wider range and longer time series, the MPF from Landsat data are more efficient than those from Sentinel-2 for verifying large-scale MPF products or obtaining long-term monitoring of a fixed area.

The contribution of melt ponds to enhanced Arctic sea-ice melt during the Last Interglacial

The Hadley Centre Global Environment Model version 3 (HadGEM3) is the first coupled climate model to simulate an ice-free Arctic during the Last Interglacial (LIG), 127 000 years ago. This simulation appears to yield accurate Arctic surface temperatures during the summer season. Here, we investigate the causes and impacts of this extreme simulated ice loss. We find that the summer ice melt was predominantly driven by thermodynamic processes: atmospheric and ocean circulation changes did not significantly contribute to the ice loss. We demonstrate these thermodynamic processes were significantly impacted by melt ponds, which formed on average 8 d earlier during the LIG than during the pre-industrial control (PI) simulation. This relatively small difference significantly changed the LIG surface energy balance and impacted the albedo feedback. Compared to the PI simulation: in mid-June, of the absorbed flux at the surface over ice-covered cells (sea-ice concentration > 0.15), ponds accounted for 45 %–50 %, open water 35 %–45 %, and bare ice and snow 5 %–10 %. We show that the simulated ice loss led to large Arctic sea surface salinity and temperature changes. The sea surface temperature and salinity signals we identify here provide a means to verify, in marine observations, if and when an ice-free Arctic occurred during the LIG. Strong LIG correlations between spring melt pond and summer ice area indicate that, as Arctic ice continues to thin in future, the spring melt pond area will likely become an increasingly reliable predictor of the September sea-ice area. Finally, we note that models with explicitly modelled melt ponds seem to simulate particularly low LIG sea-ice area. These results show that models with explicit (as opposed to parameterised) melt ponds can simulate very different sea-ice behaviour under forcings other than the present day. This is of concern for future projections of sea-ice loss.

Influence of Melt Ponds on the SSMIS-Based Summer Sea Ice Concentrations in the Arctic

As a long-term, near real-time, and widely used satellite derived product, the summer performance of the Special Sensor Microwave Imager/Sounder (SSMIS)-based sea ice concentration (SIC) is commonly doubted when extensive melt ponds exist on the ice surface. In this study, three SSMIS-based SIC products were assessed using ship-based SIC and melt pond fraction (MPF) observations from 60 Arctic cruises conducted by the Ice Watch Program and the Chinese Icebreaker Xuelong I/II. The results indicate that the product using the NASA Team (SSMIS-NT) algorithm and the product released by the Ocean and Sea Ice Satellite Application Facility (SSMIS-OS) underestimated the SIC by 15% and 7–9%, respectively, which mainly occurred in the high concentration rages, such as 80–100%, while the product using the Bootstrap (SSMIS-BT) algorithm overestimated the SIC by 3–4%, usually misestimating 80% < SIC < 100% as 100%. The MPF affected the SIC biases. For the high MPF case (e.g., 50%), the estimated biases for the three products increased to 20% (SSMIS-NT), 7% (SSMIS-BT), and 20% (SSMIS-OS) due to the influence of MPF. The relationship between the SIC biases and the MPF observations established in this study was demonstrated to greatly improve the accuracy of the 2D SIC distributions, which are useful references for model assimilation, algorithm improvement, and error analysis.

Meltwater sources and sinks for multiyear Arctic sea ice in summer

On Arctic sea ice, the melt of snow and sea ice generate a summertime flux of fresh water to the upper ocean. The partitioning of this meltwater to storage in melt ponds and deposition in the ocean has consequences for the surface heat budget, the sea ice mass balance, and primary productivity. Synthesizing results from the 1997–1998 SHEBA field experiment, we calculate the sources and sinks of meltwater produced on a multiyear floe during summer melt. The total meltwater input to the system from snowmelt, ice melt, and precipitation from 1 June to 9 August was equivalent to a layer of water 80 cm thick over the ice-covered and open ocean. A total of 85 % of this meltwater was deposited in the ocean, and only 15 % of this meltwater was stored in ponds. The cumulative contributions of meltwater input to the ocean from drainage from the ice surface and bottom melting were roughly equal.

The distribution and evolution of supraglacial lakes on 79° N Glacier (north-eastern Greenland) and interannual climatic controls

The Nioghalvfjerdsfjorden glacier (also known as the 79∘ North Glacier) drains approximately 8 % of the Greenland Ice Sheet. Supraglacial lakes (SGLs), or surface melt ponds, are a persistent summertime feature and are thought to drain rapidly to the base of the glacier and influence seasonal ice velocity. However, seasonal development and spatial distribution of SGLs in the north-east of Greenland are poorly understood, leaving a substantial error in the estimate of meltwater and its impacts on ice velocity. Using results from an automated detection of melt ponds, atmospheric and surface mass balance modelling, and reanalysis products, we investigate the role of specific climatic conditions in melt onset, extent, and duration from 2016 to 2019. The summers of 2016 and 2019 were characterised by above-average air temperatures, particularly in June, as well as a number of rainfall events, which led to extensive melt ponds to elevations up to 1600 m. Conversely, 2018 was particularly cold, with a large accumulated snowpack, which limited the development of lakes to altitudes less than 800 m. There is evidence of inland expansion and increases in the total area of lakes compared to the early 2000s, as projected by future global warming scenarios.

MOSAiC drift expedition from October 2019 to July 2020: sea ice conditions from space and comparison with previous years

We combine satellite data products to provide a first and general overview of the physical sea ice conditions along the drift of the international Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition and a comparison with previous years (2005–2006 to 2018–2019). We find that the MOSAiC drift was around 20 % faster than the climatological mean drift, as a consequence of large-scale low-pressure anomalies prevailing around the Barents–Kara–Laptev sea region between January and March. In winter (October–April), satellite observations show that the sea ice in the vicinity of the Central Observatory (CO; 50 km radius) was rather thin compared to the previous years along the same trajectory. Unlike ice thickness, satellite-derived sea ice concentration, lead frequency and snow thickness during winter months were close to the long-term mean with little variability. With the onset of spring and decreasing distance to the Fram Strait, variability in ice concentration and lead activity increased. In addition, the frequency and strength of deformation events (divergence, convergence and shear) were higher during summer than during winter. Overall, we find that sea ice conditions observed within 5 km distance of the CO are representative for the wider (50 and 100 km) surroundings. An exception is the ice thickness; here we find that sea ice within 50 km radius of the CO was thinner than sea ice within a 100 km radius by a small but consistent factor (4 %) for successive monthly averages. Moreover, satellite acquisitions indicate that the formation of large melt ponds began earlier on the MOSAiC floe than on neighbouring floes.

Freshwater Sources and Sinks for Arctic Sea Ice in Summer

On Arctic sea ice, the melt of snow and sea ice generate a summertime flux of fresh water to the upper ocean. The partitioning of this freshwater to storage in melt ponds and deposition in the ocean has consequences for the surface heat budget, the sea ice mass balance, and primary productivity. Synthesizing results from the SHEBA field experiment, we calculate the sources and sinks of freshwater produced during summer melt. The total freshwater input to the system from snow melt, ice melt, and precipitation from 1 June to 9 August was equivalent to a layer of water 80 cm thick over the ice-covered and open ocean. 85 % of this freshwater was deposited in the ocean and only 15 % of this freshwater was stored in ponds. The cumulative contributions of freshwater input to the ocean from drainage from the ice surface and bottom melting were roughly equal.