From the onset of melt in early June, corresponding declines in Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) freeboard and surface albedo can be seen over the entire Arctic sea ice cover. In the 2019 summer, area-averaged freeboard decreased from 34 cm prior to melt to a minimum of 12 cm in August while the area-averaged albedo decreased from ~0.7 to 0.38 for the same period. Calculations using ICESat-2 freeboards and modeled ice thickness from Pan-Arctic Ice Ocean Modeling and Assimilation System (PIOMAS) give area-averaged snow depths ranging from 17 cm prior to melt to 3 cm in August over seasonal ice and from 34 to 4 cm over multiyear ice. Mean rates of snow ablation (including evaporation) in mid-June were as high as 2 cm/day, comparable to field records from other years. Increases in freeboard after mid-August in the high latitude (>80°N) multiyear ice cover, north of the Greenland coast, are likely due to earlier freeze-up and snow accumulation in these regions with shorter melt seasons.

Key Points

We provide a first look at the evolution of the summer ice cover in ICESat-2 sea ice products
Observations show a clear lowering of the freeboard associated with snow melt over the entire Arctic Ocean ice cover
Area-averaged snow depths ranged from 17 cm prior to melt to 3 cm in August over seasonal ice and from 34 to 4 cm over multiyear ice

Plain Language Summary

As the air temperature warms with the onset of summer, the Arctic Ocean ice cover undergoes its annual transformation from a highly reflective snow-covered surface in winter to a darker, more absorbent surface during summer composed of bare ice, melt ponds, and open water. Several weeks after the onset of melt, much of the snow layer disappears as the snow is converted into meltwater. Over the entire ice cover, this loss of the snow layer can be seen in the lowering of surface heights measured by the lidar on NASA's Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2). In this paper, we describe the changes in surface height and the corresponding changes in snow depth from the satellite observations and compare them with existing measurements from field programs, which are limited due to challenging logistics during the summer. We show, for the first time, that we can observe corresponding changes in surface height and reflectivity as well as the different rates of melt in the decay of the snow cover. These results will allow for an improved understanding of the impact of the snow layer on the fate of the ice cover at the end of the summer.

1 Introduction

As the air temperature warms with the onset of the Arctic summer, the sea ice cover begins its annual transformation from a high-albedo snow-covered surface (>0.8) during winter to one that is dominated by lower-albedo surfaces of bare ice, melt ponds, and open leads (Perovich & Polashenski, 2012). Several weeks after melt onset, much of the snow layer disappears as meltwater accumulates in ponds on the surface. Melt pond coverage varies initially as formation and drainage occur at different rates that depend on surface topography, available freshwater, and ice permeability (Eicken et al., 2002; Polashenski et al., 2012; Webster et al., 2015), but on average, it increases into late summer. Recovery in large-scale albedo does not occur until new ice forms and snow accumulates with freeze-up in the fall.

Observations of albedo variability during summer have received significant attention due to its impact on the radiative balance at the surface (Maykut & Untersteiner, 1971; Sturm et al., 2002), but documentation of the associated rate of ablation of snow and ice is more limited due to difficulties in accessibility and sampling of the summer ice cover (Perovich et al., 2003; Radionov et al., 1997). The frequent coverage offered by spaceborne altimetry of summer sea ice could potentially be useful for providing large-scale characterization of surface height changes during the progression of melt. Radar altimetry, however, is confounded by the changes in penetration into the snow layer with the wetting of the snow when temperatures approach freezing and, more specifically for CryoSat-2, the presence of strong off-nadir reflections from quasi-specular surfaces like melt ponds and leads. To date, we are not aware of studies of summer sea ice utilizing altimetric measurements from CryoSat-2 in published literature. Lidar altimetry (e.g., Ice, Cloud, and Land Elevation Satellite [ICESat], ICESat-2 [IS-2]) offers another opportunity with the higher-resolution footprints that allow the separation of surface types and does not suffer from issues associated with radar altimetry. But the campaign-mode acquisitions of the ICESat mission (2003–2009) was not able to provide coverage of the summer season for a detailed examination of the capabilities of lidar altimeters until the launch of IS-2 in September 2018. The Advanced Topographic Laser Altimeter System (ATLAS) onboard IS-2 consists of six photon-counting beams for surface profiling with a 10-kHz pulse rate (interpulse distance ~0.7 m) and footprints of ~14 m, an improvement over the ICESat footprints of 50–70 m and spacings of 167 m. This lidar design provides multiple surface profiles for improved spatial coverage and higher spatial resolution than preceding and current spaceborne altimeters. For sea ice, the capability to obtain high-resolution samples (tens of meters) with vertical precision of less than a few centimeters is critical for obtaining local sea level in narrow openings within the ice cover for freeboard determination and for subsequent conversion to thickness estimates. Early assessments of the capabilities of IS-2 during the winter in Kwok et al. (2019) and Kwok et al. (2019) suggest that the observed height precision may be useful for observing changes in the summer ice cover.

At this writing, the IS-2 project has released to the community a suite of IS-2 data products, which spans the period between 14 October 2018 and 6 September 2019, which includes the summer of 2019. This data set contains two sea ice products: one containing surface heights (ATL07) and the other total freeboards (snow+ice, ATL10). Here, the aim is to provide a first assessment of the summer sea ice products in terms of the changes in total freeboard and what can be inferred from these observations during the melt season. The next section describes the data sets used in this analysis. Section 3 discusses the observed changes and their correspondence to changes in surface albedo over the Arctic Ocean. Section 3.3 provides estimates of the snow depth changes using ice thickness from a coupled ice-ocean simulation. The last section concludes this paper.

2 Data Description

Three data sets are used in our analysis: IS-2 freeboard, surface albedo from EUMETSAT's Satellite Application Facility on Climate Monitoring (CM-SAF), and model ice thickness from Pan-Arctic Ice Ocean Modeling and Assimilation System (PIOMAS).

2.1 IS-2 Along-Track Freeboard

IS-2 employs three beam pairs to profile the surface; the pairs are separated by ~3.3-km cross-track with intrapair spacings of 90 m. Each pair consists of a strong and a weak beam, where the pulse energies of the strong beams are 4 times that of the weak. As mentioned earlier, each beam profiles the surface with a lidar footprint of ~14 m and a pulse rate of 10 kHz (or interpulse spacing of ~0.7 m). A more detailed description of the IS-2 mission and the onboard lidar system can be found in Markus et al. (2017) and Neumann et al. (2019). The sea ice products (ATL07 and ATL10) contain along-track heights and total freeboards (snow+ice) for individual beams. Here, we use Release 002 of ATL10 (Kwok et al., 2019), which covers the summer of 2019. More details about the products can be found in Kwok, Markus, Kurtz, et al. (2019) and Kwok, Cunningham, Hancock, et al. (2019). The total freeboards in ATL10 are calculated in 10-km segments where sea level estimates are available; the along-track spatial resolution of individual estimates varies between ~27 and 200 m and is dependent on surface reflectance.

2.2 Surface Albedo

Weekly mean surface albedo fields of Arctic sea ice (on a 15-km map grid), from May to August 2019, are from the Cloud, Albedo and Surface Radiation data set (CLARA-SAL) produced by CM-SAF. The retrievals are based on aggregation of radiance data acquired by the Advanced Very High Resolution Radiometer (AVHRR) instrument onboard polar-orbiting NOAA and Metop satellites (Karlsson et al., 2012). Raw radiances are corrected for atmospheric, radiometric, and topographic effects and expanded into hemispherical spectral albedos before shortwave albedo is obtained using a narrow-to-broadband conversion. Expected uncertainty in the albedo estimates is around 5% (or 0.05), and retrievals of temporal mean area-averaged albedo derived from only a few observations are vulnerable to errors in cloud masking. The reader is referred to Karlsson et al. (2012) for a more detailed description of this product.

2.3 Sea Ice Thickness

Daily thickness fields are from PIOMAS (Schweiger et al., 2011; Zhang & Rothrock, 2003). For the PIOMAS simulations used here (Version 2.1), sea ice concentrations and sea surface temperatures are assimilated into the model to improve ice thickness estimates and representation of ice-free areas. Winds, surface air temperature, and cloud cover to compute solar and long-wave radiation are prescribed by atmospheric fields from the NCEP/NCAR atmospheric reanalysis. The pan-Arctic ocean model is forced with input from a global ocean model at its open boundaries located at 45°N.

3 Data Analysis

We first describe the IS-2 summer freeboards and the biweekly fields that were used in our analysis. Second, the observed evolution of the freeboards over the summer (between May and September) is described. Last, we estimate the snow depth using modeled ice thickness from PIOMAS and examine their changes over this summer.

3.1 ATL10 Freeboard During 2019 Summer

In winter, the total freeboard in ATL10 represents the height of the snow surface from the local sea surface. The retrieved freeboard is the sum of the height of the ice freeboard and thickness of the snow layer. In summer, the profiled surface types also include those of wet snow, melt ponds, and bare ice. The surface finding procedure in ATL10 (Kwok et al., 2019) is designed not to respond to the bottom of melt ponds, and therefore the likelihood of biases due to this effect is small. In the sea ice surface finder, when two statistically distinct height distributions are detected in the photon heights, the surface finder preferentially outputs the height of the top surface. It is also unlikely that ponds would be classified as sea surface references used in the calculation of freeboard because (1) a mixture of ponds and ice in individual freeboard estimates including ponds that are typically smaller than the freeboard resolution (i.e., >25 m) would appear as rough surfaces in the photon clouds and would not be designated as sea surface samples, which are expected to be smooth, and (2) even if ponds were mistakenly classified as the sea surface, the procedure that picks the sea surface always selects the lowest height samples; during the summer when the ice concentration is low and there is an abundance of sea surface, the misclassification rate is expected to be small. In the absence of coincident field data, the best assessment of the satellite freeboards, as with most remote sensing data sets, is by testing whether the retrievals are consistent with the expected evolution of the snow cover during summer—the subject of this paper.

The gridded biweekly freeboard composites (25-km spacing) in Figure 1 are constructed with per orbit freeboards from the three strong beams in the ATL10 products. The displayed fields have been smoothed using a 25-km Gaussian kernel. With an ~3.3-km separation between the three beams, there can be up to 500 independent freeboard samples for a given 25-km segment of the orbit, or 1,500 samples from all three beams for computing a composite average. Instead of the monthly composites shown in Kwok, Markus, Kurtz, et al. (2019), the biweekly sampling is more suitable for visualizing and quantifying the faster changes in freeboard associated with snow melt in the summer while still providing relatively good spatial coverage of the entire Arctic Ocean. We also note again that, in the IS-2 data set used here, there is a 1-month gap in coverage (26 June to 26 July, indicated in the figures) due to a spacecraft anomaly, and consequently, there are two missing biweekly fields.

Details are in the caption following the image — **Figure 1**
Open in figure viewer PowerPoint

Biweekly composites of total freeboard from ICESat-2 (first column), sea ice thickness from PIOMAS (second column), and derived snow depth (third column) and its distribution (fourth column—numerical values show mean and standard deviation in meters) between 1 May and 6 September. Note the gap in IS-2 data collection between 27 June and 25 July due to a spacecraft anomaly.

3.2 Evolution of IS-2 Freeboard and Albedo During Summer (May–September)

The eight biweekly composites of total freeboard from IS-2 ATL10 products (first column, Figure 1) show the evolution of Arctic Ocean sea ice freeboards (ATL10) over the four summer months (1 May through 6 September). Here, we limit our analysis to the area within Arctic Basin that is bounded by the gateways into the Pacific (Bering Strait), the Canadian Arctic Archipelago (CAA), and the Greenland (Fram Strait) and Barents Seas.

Marked changes are seen in total freeboard, between the first half (1–15) and second half (16–26) of June (Figure 1). The freeboard has lowered everywhere; the higher freeboards north of the coast of Greenland and the CAA (in red) have all but disappeared. This can be compared with the month of May where there were relatively minor freeboard changes (visually) in the biweekly composites. The total freeboard continues to decrease, albeit at a slower rate, through the second half of July (noting again the gap in coverage) and the first half of August. In the second half of August, the total freeboards at the higher polar latitudes (>80°N) begin to increase and continue to do so until the first week in September (the end of the current IS-2 data release).

The large signal in freeboard changes, during initial melt, suggests that the decreases are largely associated with snow melt and to a lesser degree the decrease in ice thickness (more on ice thickness in the next section), and these changes in snow coverage can be associated with the changes in albedo as surface melt evolves (Figure 2). Figure 2a shows the spatial evolution of surface albedo of Arctic sea ice between May and early August. The correspondence between decreases in area-averaged surface albedo over first-year (FYI) and multiyear (MYI) sea ice over that of total freeboard is shown in Figure 2b. We note that the albedo time sequence terminates around mid-August because, after the summer solstice, the data hole in surface albedo grows as the retrievals becomes unreliable with reducing solar illumination.

As the area-averaged albedo over the entire ice cover (including open water and mixtures of ice types) dropped from ~0.7 to a minimum of 0.38 in the time sequence (Figure 2b), the total freeboard decreased from 34 cm prior to melt to a minimum of 12 cm in August. Here, we examine the changes in freeboard within the context of the two generalized time sequences of surface albedo evolution (in seven stages) over the two dominant Arctic ice types (undeformed FYI and MYI) formulated from 4 years of field experiments (Perovich et al., 2002; Perovich & Polashenski, 2012). Each distinct stage, delineated in Figure 2c, is discussed below, recognizing that our temporal sampling frequency will not be able to resolve some of the finer and sharper changes that delineate the seven stages of albedo and freeboard evolution observed in the field data.

Prior to melt onset in May (Stage 1: Cold snow), while the snow is still cold, the area-averaged surface albedo in the satellite data is around 0.71 (slightly lower over FYI). The area-averaged satellite surface albedos are expected to be different from that of pure snow in field data (e.g., lower over MYI) due to the mixture of surface types in limited spatial resolution cells. Beginning with the onset of snow melt (near the end of May), there is a steep decrease in albedo and an associated lowering of freeboard (Figure 2b) (Stage 2: Melting snow). In Figures 2b and 2c, we show the rough correspondence between freeboard changes and the evolution of MYI albedo (blue) and seasonal ice albedo (red) as discussed in Perovich and Polashenski (2012), recognizing that the timing of the different stages could vary regionally.

Pond formation begins after a sufficient amount of meltwater collects in surface depressions (Stage 3: Pond formation), and the surface albedo and freeboard decrease as a result. At this stage, there is a clear lag between the development of surface albedo in FYI and MYI that is due to the northward progression of melt during the summer. Also, the seasonal snow cover on FYI is not as deep (Radionov et al., 1997; Webster et al., 2014), so it melts away faster and because of smoother topography encourages more extensive pond formation and coverage (Polashenski et al., 2012; Webster et al., 2015).

In the next stage of the sequence (Stage 4: Pond drainage), the albedo increases locally where there is pond drainage, and the ice becomes exposed at the surface. We do not observe this increase in our low-resolution area-averaged data set. Following this stage, ponds continue to develop and the exposed ice surface continues to melt, and the freeboard continues to decrease as the snow and ice surface ablation progresses. During pond evolution (Stage 5), the thinner seasonal ice still displays lower albedos than MYI, since the albedo of seasonal ponds is typically lower than multiyear ponds (Perovich & Polashenski, 2012) and because ponds typically are more extensive on seasonal than MYI during this stage (Grenfell & Perovich, 2004). At this stage, the decrease in total freeboard has increasing contributions from ice surface and basal melt.

Over the seasonal ice zone (Stage 6: Open water) where the ice disappears, the albedo reaches that of open water (0.07). In the satellite fields, the area-averaged on-ice albedo reaches a minimum of 0.37 and 0.43 on FYI and MYI, respectively. Eventually, after fall freeze-up (Stage 7: Freeze-up), once the seasonal ice begins to form and snow coverage increases, the surface albedo starts to increase over both ice types. The increase in freeboard can be seen in the IS-2 total freeboards as the snow begins to accumulate. In Figure 2b, there are gaps in satellite albedo as the retrievals become unreliable as the sun sets in the higher polar latitudes. Overall, the correspondence between the freeboard changes and the expected evolution of albedos is quite remarkable.

3.3 Estimates of Snow Depth Using PIOMAS Sea Ice Thickness

Of particular geophysical interest is the steep decreases in snow depth as the snow layer is transformed into meltwater starting with the initiation of ice surface melt in late spring. Here, we use a simple model to obtain estimates of snow depth from total freeboard discussed in the previous section. Snow depth, h_fs, can be calculated if the contribution of ice freeboard, h_fi, can be removed from the total ice freeboard, h_f (i.e., h_fs = h_f − h_fi). At this time, however, direct observations of ice freeboard during the melt season are not readily available from remote sensing or any other means. Instead, we solve for the snow depth using the sea ice thickness, h_i, from PIOMAS (a coupled ice-ocean model) with the following equations:

$urn:x-wiley:00948276:media:grl60695:grl60695-math-0001$ (1)

The numerical coefficients on the right side of the equation are based on the densities of snow (ρ_s), ice (ρ_i = 917 kg/m³) and seawater (ρ_w = 1,024 kg/m³). Here, we assume the bulk density of snow to be ~350 kg/m³, which is approximately the average of the June and July densities in Warren et al. (1999). In August, we use a bulk density 530 kg/m³ for wet and melting snow found in Radionov et al. (1997). In the solution of this equation, higher snow densities result in higher snow depths for a given h_f and h_i. By August, the snow depth is so low that the step change in snow density is not noticeable in the results.

Figure 1 shows the biweekly fields of IS-2 freeboard, PIOMAS ice thickness used in the calculations (Columns 1 and 2), and the derived snow depth and their distributions (Columns 3 and 4). The time sequence of area-averaged snow depth and ice freeboards for FYI and MYI are shown in Figure 3. Below, we first discuss the sensitivity and shortcomings of the snow depth calculations to modeled ice thickness before we describe the retrieved results from the summer of 2019.

3.3.1 Sensitivity to Ice Thickness

How credible are snow depths calculated with modeled ice thickness? It can be seen, in equation 1, that the changes in snow depth (h_fs) are relatively insensitive to changes in ice thickness (h_i) compared to changes in total freeboard (h_f), that is, changes in snow depth is only dependent on 15% of changes in ice thickness (at 350 kg/m³). Since the onset of ice surface melt is delayed until the high-albedo snow layer disappears (Petrich et al., 2012), changes in snow depth can be estimated credibly especially during Stage 1 through most of Stage 4 (when snow is melting and the albedo is decreasing rapidly) when the snow depth changes dominate the changes in total freeboard. And, during initial melt, the ice freeboard (and thickness) also varies slowly compared to changes in total freeboard (Figure 3), and this rate is only moderately higher after mid-June.

Still, an estimate of snow depth depends on the uncertainty of the modeled ice thickness even though only fractionally. We constrain the upper bound in area-averaged ice thickness so that the calculated area-averaged ice freeboard (over FYI or MYI) cannot exceed the total freeboard (at its lowest point in the summer), that is, there are no negative snow depths. This works out to be ~1.36 m in thickness, or 0.25 m above the model mean. The lower bound is constrained such that the lowest area-averaged thickness (at lowest freeboard in the summer) equals 1 m, which gives a value of 0.11 m below the model mean. We use the higher of these bounds in thickness to define a symmetric measure of uncertainty in the modeled thickness. Relative to the modeled ice thickness, this uncertainty works out to be ±0.25 m. In the results (Figure 3), the impact of the uncertainty on the derived parameters (snow depth and ice freeboard) is calculated based on this uncertainty in ice thickness and is indicated by the shaded area around the mean estimates.

3.3.2 Evolution of Snow Depth Estimates

The snow depth composites/distributions and the time sequence of area-average freeboard calculated using IS-2 freeboard and PIOMAS ice thickness are shown in Figure 1 (Column 3 and 4) and Figure 3.

In May, the composite fields (Figure 1) show a small build-up of the snow cover from 0.21 to 0.24 m. The decline in snow depth, between May and mid-August, is clearly seen in the spatial fields. Dramatic differences between the first and second half of June highlights rapid snow melt almost everywhere (a change of almost 10 cm); the tail in the snow depth distribution has disappeared. The average snow depth at the end of July is only 5 cm. After mid-August, the increase in snow depth is seen concentrated in the region north of the Fram Strait and the CAA likely owing to storms that originate in the Greenland Sea. By the first week of September, there is an average snow depth of ~15 cm on the ice cover poleward of 80°N.

The time sequences in Figure 3 show the evolution of total freeboard and the calculated snow depth and ice freeboard over the FYI, MYI, and the entire ice cover. Estimates suggest area-averaged snow depths ranged from 17.0 ± 3.0 cm prior to melt to 3.0 ± 1.8 cm in August over seasonal ice (comparable to that observed by Radionov et al., 1997 and Webster et al., 2014) and from 34.0 ± 3.5 cm to 4.1 ± 1.9 cm over MYI. This is comparable to the ~40 cm (May) and 6 cm (August) north of the Greenland Coast in Warren et al. (1999). June saw the largest decline in total freeboard and snow depth in the MYI cover, where there is almost 20 cm of loss. The uncertainties are from the ±0.25 m in ice thickness discussed above. For both ice types, the snow depth estimates largely parallel the changes in total freeboard because of the relatively slower changes in ice freeboard.

The highest rate of snow melt of 2 cm/day is seen over the MYI cover in June. This can be compared to reported melt rates of 1.9 cm/day on drifting station NP-22 in 1979 (Radionov et al., 1997) and surface melt rates that averaged 1 cm/day, with a peak value of 4 cm/day at SHEBA (Perovich et al., 2003). Of course, our estimates are probably lower compared to the field data due to the biweekly and area-averaged nature of our data set, that is, faster changes are masked. After the middle of June, the melt rates are more moderate and more like those of changes in ice freeboard due to melt, as well as concomitant melt pond formation and drainage. The minima in snow depths are 2 and 4 cm on FYI and MYI, respectively. The maximum change in our area-averaged total freeboard over MYI (which includes snow and ice surface melt) is 42 cm compared to the average of ~50 ± 25 cm recorded at manned ice camps and by drifting autonomous buoys for 1957–2014 (Perovich & Richter-Menge, 2015).

After mid-August, while the ice freeboard is still decreasing, the increases in total freeboard suggest that it is due entirely to snow accumulation on the surface on the high latitude ice cover.

4 Conclusions

In this paper, we provide a first examination of the summer freeboards in the IS-2 sea ice products. The results demonstrate that the IS-2 multibeam photon-counting lidar altimetry and the sea ice products are capable of providing good spatial coverage of the Arctic Ocean throughout the melt season and thus offer an unprecedented basin-scale view of the evolution of the summer ice cover. Further, the correspondence of the decline in total freeboard and development of surface albedo is described, and estimates of snow depths are derived using the total freeboard and ice thickness from a coupled ice ocean model (PIOMAS).

From the onset of melt in early June, declines in IS-2 freeboard associated with drops in surface albedo can be seen as the Arctic undergoes a spectacular transformation from a high-albedo, snow-covered surface to a low-albedo surface dominated by melt ponds and open leads. The large changes in total freeboard are primarily due to snow melt. While surface albedo decreased from ~0.7 to 0.38, area-averaged freeboard decreased from 34 cm prior to melt to a minimum of 12 cm in August for the same period. Between late June and mid-August, the freeboards and albedo continue to decrease as melt ponds develop. After fall freeze-up, as the surface over the MYI freezes and seasonal ice begins to form, the increase in surface albedo and freeboard can be seen in the IS-2 total freeboards as the snow begins to accumulate.

Estimates of snow depth, calculated using modeled ice thickness from PIOMAS, suggest area-averaged snow depths ranged from 17 cm prior to melt to 3 cm in August over seasonal ice and from 34 cm to 4 cm over MYI. These values are comparable with those in Radionov et al. (1997), Warren et al. (1999), and Webster et al. (2014). In the biweekly data, the highest rate of snow melt of 2 cm/day is seen over the MYI cover in June and can be compared to melt rates of a few centimeters per day from field observations (Perovich et al., 2003; Radionov et al., 1997). Increases in freeboard after mid-August in the higher latitude (>80^oN) MY ice cover north of the Greenland coast is likely due to earlier freeze-up in regions with shorter melt seasons.

The present examination of the summer IS-2 data provides a first look at the IS-2 retrievals during the melt season. These are early results and provide a guide to the use of the sea ice products from IS-2. These early results do not provide an exhaustive or comprehensive examination of the sea ice products. There are many aspects of data quality, some of which will only be revealed when assessed with data acquired by dedicated airborne campaigns (e.g., NASA's Operation IceBridge, AWI's IceBird), other satellite altimetry missions (e.g., CryoSat-2, Sentinel 3A/B), upcoming field programs (e.g., MOSAiC), and when a longer IS-2 time series becomes available.

Acknowledgments

M.A.W. acknowledges support by the National Aeronautics and Space Administration's New Investigator Program in Earth Science. R.K., G.F.C., and S.K. performed this work at the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.

Open Research

Data Availability Statement

The ICESat-2 ATL10 data set used herein is available at the National Snow and Ice Data Center: https://nsidc.org/data/ATL10/. PIOMAS sea ice thicknesses are from http://psc.apl.uw.edu/research/projects/arctic-sea-ice-volume-anomaly/, and the surface albedo products are from https://www.cmsaf.eu/EN/Products/AvailableProducts/OperationalProducts/Operational_Products_node.html;jsessionid=490CB74D6E885425310B7B1A0A830ABE.live11292.

References

Eicken, H., Krouse, H. R., Kadko, D., & Perovich, D. K. (2002). Tracer studies of pathways and rates of meltwater transport through Arctic summer sea ice. Journal of Geophysical Research, 107(C10). https://doi.org/10.1029/2000jc000583
10.1029/2000jc000583
Google Scholar
Grenfell, T. C., & Perovich, D. K. (2004). Seasonal and spatial evolution of albedo in a snow-ice-land-ocean environment. Journal of Geophysical Research, 109, C01001. https://doi.org/10.1029/2003JC001866
10.1029/2003JC001866
Web of Science®Google Scholar
Karlsson, K.-G., Riihelä, A., Müller, R., Meirink, J. F., Sedlar, J., Stengel, M., Lockhoff, M., Trentmann, J., Kaspar, F., Hollmann, R., & Wolters, E. (2012). CLARA-A1: CM SAF CLouds, Albedo and Radiation dataset from AVHRR data—Edition 1—Monthly means/daily means/pentad means/monthly histograms.
Google Scholar
Kwok, R., Cunningham, G. F., Hancock, D. W., Ivanoff, A., & Wimert, J. T.. (2019). Ice, Cloud, and Land Elevation Satellite-2 Project: Algorithm Theoretical Basis Document (ATBD) for sea ice products. https://icesat-2.gsfc.nasa.gov/science/data_products
Google Scholar
Kwok, R., Cunningham, G., Markus, T., Hancock, D., Morison, J. H., Palm, S. P., Farrell S. L., Ivanoff A., Wimert J., and the ICESat-2 Science Team (2019). ATLAS/ICESat-2 L3A Sea Ice Height, Version 1. Boulder, Colorado USA. NSIDC: National Snow and Ice Data Center. https://doi.org/10.5067/ATLAS/ATL07.001
10.5067/ATLAS/ATL07.001
Google Scholar
Kwok, R., Kacimi, S., Markus, T., Kurtz, N. T., Studinger, M., Sonntag, J. G., Manizade, S. S., Boisvert, L. N., & Harbeck, J. P. (2019). ICESat-2 surface height and sea ice freeboard assessed with ATM lidar acquisitions from operation IceBridge. Geophysical Research Letters, 46, 11,228–11,236. https://doi.org/10.1029/2019gl084976
10.1029/2019gl084976
Web of Science®Google Scholar
Kwok, R., Markus, T., Kurtz, N. T., Petty, A. A., Neumann, T. A., Farrell, S. L., Cunningham, G. F., Hancock, D. W., Ivanoff, A., & Wimert, J. T. (2019). Surface height and sea ice freeboard of the Arctic Ocean from ICESat-2: Characteristics and early results. Journal of Geophysical Research: Oceans, 124, 6942–6959. https://doi.org/10.1029/2019jc015486
10.1029/2019jc015486
ADSWeb of Science®Google Scholar
Markus, T., Neumann, T., Martino, A., Abdalati, W., Brunt, K., Csatho, B., Farrell, S., Fricker, H., Gardner, A., Harding, D., Jasinski, M., Kwok, R., Magruder, L., Lubin, D., Luthcke, S., Morison, J., Nelson, R., Neuenschwander, A., Palm, S., Popescu, S., Shum, C. K., Schutz, B. E., Smith, B., Yang, Y., & Zwally, J. (2017). The Ice, Cloud, and land Elevation Satellite-2 (ICESat-2): Science requirements, concept, and implementation. Remote Sensing of Environment, 190, 260–273. https://doi.org/10.1016/j.rse.2016.12.029
10.1016/j.rse.2016.12.029
ADSWeb of Science®Google Scholar
Maykut, G. A., & Untersteiner, N. (1971). Some results from a time-dependent thermodynamic model of sea ice. Journal of Geophysical Research, 76(6), 1550. https://doi.org/10.1029/JC076i006p01550
10.1029/JC076i006p01550
ADSWeb of Science®Google Scholar
Neumann, T. A., Martino, A. J., Markus, T., Bae, S., Bock, M. R., Brenner, A. C., Brunt, K. M., Cavanaugh, J., Fernandes, S. T., Hancock, D. W., Harbeck, K., Lee, J., Kurtz, N. T., Luers, P. J., Luthcke, S. B., Magruder, L., Pennington, T. A., Ramos-Izquierdo, L., Rebold, T., Skoog, J., & Thomas, T. C. (2019). The Ice, Cloud, and Land Elevation Satellite–2 Mission: A global geolocated photon product. Remote Sensing of Environment, 233, 111325. https://doi.org/10.1016/j.rse.2019.111325
10.1016/j.rse.2019.111325
ADSWeb of Science®Google Scholar
Perovich, D. K., Grenfell, T. C., Light, B., & Hobbs, P. V. (2002). Seasonal evolution of the albedo of multiyear Arctic sea ice. Journal of Geophysical Research, 107(C10), 8044 Retrieved from <Go to ISI>://WOS:000180568000020. https://doi.org/10.1029/2000JC000438
10.1029/2000JC000438
ADSWeb of Science®Google Scholar
Perovich, D. K., Grenfell, T. C., Richter-Menge, J. A., Light, B., Tucker, W. B., & Eicken, H. (2003). Thin and thinner: Sea ice mass balance measurements during SHEBA. Journal of Geophysical Research, 108(C3), 8050. Retrieved from <Go to ISI>://000182224300006. https://doi.org/10.1029/2001JC001079
10.1029/2001JC001079
ADSWeb of Science®Google Scholar
Perovich, D. K., & Polashenski, C. (2012). Albedo evolution of seasonal Arctic sea ice. Geophysical Research Letters, 39, L08501. https://doi.org/10.1029/2012gl051432
10.1029/2012gl051432
ADSWeb of Science®Google Scholar
Perovich, D. K., & Richter-Menge, J. A. (2015). Regional variability in sea ice melt in a changing Arctic. Philosophical Transactions. Series A, Mathematical, Physical, and Engineering Sciences, 373(2045), 20140165. https://doi.org/10.1098/rsta.2014.0165
10.1098/rsta.2014.0165
ADSPubMedWeb of Science®Google Scholar
Petrich, C., Eicken, H., Polashenski, C. M., Sturm, M., Harbeck, J. P., Perovich, D. K., & Finnegan, D. C. (2012). Snow dunes: A controlling factor of melt pond distribution on Arctic sea ice. Journal of Geophysical Research, 117, C09029. https://doi.org/10.1029/2012jc008192
10.1029/2012jc008192
Web of Science®Google Scholar
Polashenski, C., Perovich, D., & Courville, Z. (2012). The mechanisms of sea ice melt pond formation and evolution. Journal of Geophysical Research, 117. https://doi.org/10.1029/2011jc007231
10.1029/2011jc007231
Google Scholar
Radionov, V., N. N. Bryazgin, & E. I. Alexandrov. (1997). The snow cover of the Arctic Basin (APL-UW TR 9701). Retrieved from Applied Physics Laboratory, University of Washington.
Google Scholar
Schweiger, A., Lindsay, R., Zhang, J. L., Steele, M., Stern, H., & Kwok, R. (2011). Uncertainty in modeled Arctic sea ice volume. Journal of Geophysical Research, 116, C00D06. https://doi.org/10.1029/2011jc007084
10.1029/2011jc007084
Web of Science®Google Scholar
Sturm, M., Perovich, D. K., & Holmgren, J. (2002). Thermal conductivity and heat transfer through the snow on the ice of the Beaufort Sea. Journal of Geophysical Research, 107(C10), 8043. Retrieved from <Go to ISI>://WOS:000180574200001. https://doi.org/10.1029/2000JC000409
10.1029/2000JC000409
ADSWeb of Science®Google Scholar
Warren, S. G., Rigor, I. G., Untersteiner, N., Radionov, V. F., Bryazgin, N. N., Aleksandrov, Y. I., & Colony, R. (1999). Snow depth on Arctic sea ice. Journal of Climate, 12(6), 1814–1829. https://doi.org/10.1175/1520-0442(1999)012<1814:Sdoasi>2.0.Co;2
10.1175/1520-0442(1999)012<1814:Sdoasi>2.0.Co;2
ADSWeb of Science®Google Scholar
Webster, M. A., Rigor, I. G., Nghiem, S. V., Kurtz, N. T., Farrell, S. L., Perovich, D. K., & Sturm, M. (2014). Interdecadal changes in snow depth on Arctic sea ice. Journal of Geophysical Research: Oceans, 119, 5395–5406. https://doi.org/10.1002/2014JC009985
10.1002/2014JC009985
ADSWeb of Science®Google Scholar
Webster, M. A., Rigor, I. G., Perovich, D. K., Richter-Menge, J. A., Polashenski, C. M., & Light, B. (2015). Seasonal evolution of melt ponds on Arctic sea ice. Journal of Geophysical Research: Oceans, 120, 5968–5982. Retrieved from <Go to ISI>://WOS:000363470300005. https://doi.org/10.1002/2015JC011030
10.1002/2015JC011030
ADSWeb of Science®Google Scholar
Zhang, J. L., & Rothrock, D. A. (2003). Modeling global sea ice with a thickness and enthalpy distribution model in generalized curvilinear coordinates. Monthly Weather Review, 131(5), 845–861. https://doi.org/10.1175/1520-0493(2003)131<0845:Mgsiwa>2.0.Co;2
10.1175/1520-0493(2003)131<0845:Mgsiwa>2.0.Co;2
ADSWeb of Science®Google Scholar

Citing Literature

Volume47, Issue12

28 June 2020

e2020GL088209

This article also appears in:

The Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) On-Orbit Performance, Data Discoveries and Early Science

Decay of the Snow Cover Over Arctic Sea Ice From ICESat-2 Acquisitions During Summer Melt in 2019

Abstract

Key Points

Plain Language Summary

1 Introduction