A different pattern of sea ice retreat

Arctic sea ice extent on July 16 fell roughly between the extent for the same day in 2007 and the long-term average. The spatial pattern of summer ice loss has evolved differently from last year; this reflects the prevailing pattern of atmospheric circulation. Areas of low-concentration ice are also developing at unusually high latitudes.

Note: Analysis updates, unless otherwise noted, now show a single-day extent value for Figure 1, as opposed to the standard monthly average. While monthly average extent images are more accurate in understanding long-term changes, the daily images are helpful in monitoring sea ice conditions in near-real time.

Map of sea ice from space, showing sea ice, continents, ocean
Figure 1. Daily Arctic sea ice extent for July 16, 2008 was 8.91 million square kilometers (3.44 million square miles). The orange line shows the 1979-2000 average extent for that day.About the data.
—Credit: National Snow and Ice Data Center
See High Resolution Image

Overview of conditions

Arctic sea ice extent on July 16 stood at 8.91 million square kilometers (3.44 square miles). While extent was below the 1979 to 2000 average of 9.91 square kilometers (3.83 million square miles), it was 1.05 million square kilometers (0.41 million square miles) above the value for July 16, 2007 (see Figures 1 and 2).

Graph with months on x axis and extent on y axis
Figure 2. Daily sea ice extent; the blue line indicates 2008; the gray line indicates extent from 1979 to 2000; the dotted green line shows extent for 2007.
—Credit: National Snow and Ice Data Center
See High Resolution Image

Conditions in context

The current pattern of sea ice retreat is noticeably different than last summer, with some areas showing less ice and others showing more. For example, in mid-July 2007, a large area of the southern Beaufort Sea north of Alaska still had ice; this year, it is already ice-free (see Figure 1). in 2007, large areas along the Siberian coast had melted out by mid-July; as of July 16, 2008, the Siberian sector remained largely ice-covered. Although the Siberian area still shows ice, satellite data reveals that the ice is low concentration and thus prone to melting.

Colleagues at the University of Bremen, Germany post daily satellite images from the NASA Advanced Microwave Sounding Radiometer (AMSR) at http://www.iup.uni-bremen.de:8084/amsr/amsre.html. The images not only show the low ice concentrations in the Siberian sector discussed above; they also show that a large part of the northern Beaufort Sea is covered by ice with low concentrations of 50 to 75 percent, and is likely to melt out soon. An unusual area of low ice concentration is also developing near 85 degrees North latitude.

Graph with months on x axis and extent on y axis
Figure 3:  Above is a map of sea-level pressure, averaged for the the three-week period from June 23 through July 13, 2008. Areas of high pressure are shown in yellow and red; areas of low pressure are shown in blue and purple.Thin lines indicate isobars, which connect areas of equal pressure. Winds between high- and low-pressure areas, indicated by “H” and “L” respectively, are from the south.
—Credit: From National Snow and Ice Data Center courtesy Climate Diagnostic Center
See High Resolution Image

Winds from the south

The spatial pattern of sea ice concentration, discussed above, is influenced by atmospheric circulation patterns. Figure 3 shows the pattern of sea-level pressure averaged over the three-week period from June 23 through July 13, 2008.

Winds tend to blow parallel to the isobar lines with a strength proportional to the pressure gradient; the closer the spacing between the isobars, the stronger the wind. The direction of the wind is determined from Buys Ballot’s Law, which states that if you are standing with your back to the wind in the Northern Hemisphere, low pressure will be on your left, and high pressure will be on your right.

So, from Figure 3, we can see that the tightly spaced isobars between the high-pressure cell in the Canadian Arctic Islands and lower pressure in the East Siberian Sea points to strong and persistent winds from the south. This wind pattern helps transport sea ice poleward, away from the Alaska coast, and favors the development of open water in the area.

View of Arctic from above

Figure 4.  This image is a map of air-temperature anomalies at the 925 millibar level (about 3,000 feet above the surface), averaged for the three-week period from June 23 through July 13, 2008.  Yellow and red indicate areas with above-average temperature; blue and purple indicate areas with below-average temperature.

—Credit: From National Snow and Ice Data Center courtesy Climate Diagnostic Center
See High Resolution Image

 

Warm winds contribute to melt

In addition to pushing sea ice poleward, winds from the south tend to be warmer. Figure 4 shows a map of air-temperature anomalies averaged for the three-week period from June 23 through 13, 2008. While temperatures over most of the Arctic Ocean have been above normal, they have been particularly warm north of Alaska. There, persistent winds from the south are causing strong melt and reduced ice concentrations.

How is this different from what we saw in the record-breaking year 2007? In early July 2007, an atmospheric pattern developed that featured high pressure over the Beaufort Sea. This pattern promoted especially strong sea ice loss. The pattern that has dominated the summer of 2008, so far, seems less favorable for ice loss. However, the melt season has a long way to go. Furthermore, as discussed above, large areas of the pack ice with fairly low concentrations are likely to melt out soon.

View of Arctic from above showing ice age
Figure 5. On the left are maps of ice thickness for late winter of the last three years; note the widespread coverage of fairly thin ice (purples and blues) in February-March 2008. The graphs on the right show the statistical distribution of ice thickness; note the thinner multiyear ice in 2008. Data from ICESat.

—Credit: From the National Snow and Ice Data Center courtesy Ronald Kwok, NASA Jet Propulsion Laboratory.
High Resolution Image not available

Sea ice thickness update

Previous discussion (see April 7, 2008) presented evidence that much of the Arctic Ocean this winter and spring, including the area near the North Pole, was covered with fairly thin, first-year ice. This thin, young ice is vulnerable to melting completely in summer. The large areas of low-concentration ice discussed above reinforce this concern.

Figure 5 shows sea ice thickness for late winter of 2006, 2007, and 2008 derived from the NASA ICESat laser altimeter instrument and provided by Ronald Kwok at the Jet Propulsion Laboratory. Based on Kwok’s analysis, the first-year ice that formed since last autumn, while spatially extensive, has a mean thickness of 1.6 meters (5.2 feet), which is close to the thickness seen in 2006 and 2007. Much of this season’s first-year ice formed rather late last autumn, so we had expected to see thinner first-year ice.

So why is the first-year ice thicker than anticipated? Sparse snow cover last winter may have hastened its growth: less snow on the ice means less insulation from the frigid winter air, and faster ice growth. Much of the snowfall over the Arctic Ocean occurs in early autumn, but early last autumn much of the Arctic Ocean was still ice-free and could not collect snow. Once the ice formed, it grew quickly.

In contrast to the first-year ice, the multiyear ice in 2008 appears to be much thinner than in the past two years. One factor may be the strong basal melt, or melt at the underside of the ice, observed during summer 2007. Based on Kwok’s ice motion analysis, another factor could be an unusually high export of thick ice out of the Arctic Ocean through Nares Strait, the narrow strait between the Lincoln Sea and Baffin Bay. This export augmented the multiyear ice outflow through the Fram Strait.

For previous analysis, please see the drop-down menu under Archives in the right navigation at the top of this page.

Melt onset earlier than normal

Arctic sea ice extent for June 2008 is close to that for 2007, which went on to reach the lowest minimum since at least 1979. More notably, however, satellite data indicate that melt began significantly earlier than last year over most of the Arctic Ocean. The large area of the Arctic Ocean covered by first-year ice (described in our June analysis), coupled with the early onset of melting, may mean more rapid and more severe summer ice retreat than last year.

Map of sea ice from space, showing sea ice, continents, oceanFigure 1. Arctic sea ice extent for June 2008 was 11.44 million square kilometers (4.42 million square miles). The magenta line shows the median ice extent for June from 1979 to 2000. About the data.

—Credit: National Snow and Ice Data CenterSee High Resolution Image

 

Overview of conditions

Arctic sea ice extent averaged for June stood at 11.44 million square kilometers (4.42 million square miles), 0.72 million square kilometers (0.25 million square miles) less than the 1979 to 2000 average for the month. This is very slightly (0.05 million square kilometers; 0.02 million square miles) lower than the average extent for June 2007, but not the lowest on record, which occurred in June 2006 (see Figure 3).

While the monthly average was slightly less than June 2007, Figure 2 indicates that on a daily basis, sea ice extent appears slightly higher than 2007 for most of the month. This apparent contradiction arises because of the monthly averaging calculation and because some days may have areas of missing data. To be included as an ice-covered region in the monthly average, the average concentration for that region must exceed 15 percent. So if the concentration is 15 percent for 29 days, but less than 15 percent for 1 day, it will not be included in the average ice extent for the month. Also, since ice extent decreases during June, if there is slightly more missing data in the early part of the month the monthly average could slightly underestimate the sea ice extent.

June sea ice extents in 2008 and 2007 are essentially identical, and near the lowest values for June ever recorded by satellite for the Arctic.

Graph with months on x axis and extent on y axis

Figure 2. Daily sea ice extent; the blue line indicates 2008; the gray line indicates extent from 1979 to 2000; the dotted green line shows extent for 2007.
—Credit: National Snow and Ice Data Center

See High Resolution Image

Conditions in context

While sea ice extent averaged for June 2008 was similar to last year, there were pronounced differences in the spatial pattern of the retreat through the month. Last year, open water quickly developed along the coasts of the Chukchi and Laptev seas. This year, an unusually large polynya has opened in the Beaufort Sea, and there is significantly less sea ice in Hudson Bay and Baffin Bay.

Graph with months on x axis and extent on y axis
Figure 3. Average June ice extent for 1979 through 2008
—Credit: National Snow and Ice Data CenterSee High Resolution Image

June 2008 compared to past Junes

June sea ice extent is very similar to last year and is now the third lowest on record. It lies very close to the linear trend line for all average June sea ice extents since 1979, which indicates that the Arctic is losing an average of 41,000 square kilometers (15,800 square miles) of ice per year in June. Last year, the rapid melt leading to the record-breaking minimum extent began in July.

View of Arctic from above

Figure 4. The colors in the above image indicate date of onset of melt over the Arctic Ocean. Light gray indicates areas that have not yet begun to melt this year, or areas for which data is not available. Data from the SSM/I sensor; algorithm used to process the data came from Thorsten Markus at Goddard Space Flight Center.

—Credit: Natonal Snow and Ice Data CenterSee High Resolution Image

Early onset of melt

Preliminary satellite data shows us that surface melt began earlier than usual over the western and central Arctic Ocean and Baffin Bay (see Figure 4). Last year was fairly typical except for significant early melt in the Laptev and Barents seas. This year, sea ice in the Beaufort Sea began to melt on average 15 days earlier than normal, and 15 days earlier than last year. Surface melt in the Chukchi and East Siberian seas was 6 days earlier than normal, and 14 days earlier than in 2007. In the central Arctic Ocean, melt began around June 9th, which was 12 days earlier than normal and 9 days earlier than the year before. In Baffin Bay, surface melt began 14 days earlier than last year and was 16 days earlier than normal. Areas where melt occurred later, compared to last year, are confined to the margins of the ice cover. These preliminary results will be updated as more data becomes available.

View of Arctic from above showing ice ageFigure 5. This image shows the percent anomaly of ocean absorption of solar heat from January 1 to September 21, 2007, compared to the 1979 to 2005 average. Dark red and orange indicate areas with especially low albedo. Data from SSM/I sensor.
Figure 4 was updated on July 3, 2008, with data through July 1. A previous version, posted on July 2, used data from June 10, 2008.

—Credit: From the National Snow and Ice Data Center courtesy Don Perovich, U.S. Army Corps of Engineers Cold Regions Research and Engineering Laboratory See High Resolution Image

Why earlier melt matters

What are the implications of this year’s earlier-than-normal melt onset? As melting begins, the layer of snow on top of the ice becomes wet and then disappears, leaving bare ice and ponded water. Each of these changes reduces the reflectance of the surface—increasing absorption of solar energy, further reducing reflectance, and promoting even stronger melt. This is known as the ice-albedo feedback.

Early melt onset exposes the snow and ice to more days with low reflectance. It also increases the exposure during the critical early summer season, when solar energy is at its peak. As colleague Don Perovich of the Cold Regions Research and Engineering Laboratory notes, this combination enhances ice-albedo feedback (see Figure 5). Perovich calculated that in 2007, some areas of the Arctic absorbed eight times as much heat because of the ice-albedo feedback, contributing heavily to last year’s record-breaking melt.

The combination of ice-albedo feedback and early melt onset in 2008 sets the stage for significant ice losses this summer. Three of the most important factors in sea ice losses are melt onset, cloud conditions throughout the melt season, and atmospheric circulation throughout the melt season. With melt onset having occurred earlier than usual, cloud and atmospheric conditions over the next two months come to the forefront. To learn more about cloud conditions and atmospheric circulation, read “More on the sea ice-atmosphere connection” in our June analysis.

A community sea ice outlook

The Study of Environmental Arctic Change program has released a Sea Ice Outlook for 2008. Their May report has predictions from a number of different scientific groups (including NSIDC) of how much sea ice will be left in the Arctic at the end of the melt season. The predictions range widely above and below last year’s record minimum of 4.13 million square kilometres (1.59 million square miles).

References

Perovich, D. K., J. A. Richter-Menge, K. F. Jones, and B. Light (2008), Sunlight, water, and ice: Extreme Arctic sea ice melt during the summer of 2007, Geophys. Res. Lett., 35, L11501, doi:10.1029/2008GL034007.

Stroeve, J.C., T. Markus and W.N. Meier (2006), Recent changes in the Arctic melt season, Annals of Glaciology, 44, 367-374

For previous analysis, please see the drop-down menu under Archives in the right navigation at the top of this page.

Arctic sea ice still on track for extreme melt

Arctic sea ice extent has declined through the month of May as summer approaches. Daily ice extents in May continued to be below the long-term average and approached the low levels seen at this time last year. As discussed in our last posting, the spring ice cover is thin. One sign of thin and fairly weak ice is the formation of several polynyas in the ice pack.

A note on satellite update and intercalibration

The DMSP F13 satellite that has been central to our Arctic sea ice analysis for the past several years is nearing the end of its mission. As is standard data practice, we have transitioned to a newer sensor, in this case the DMSP F15. The DMSP F15 has the same type of sensor as the DMSP F13.

NSIDC has done preliminary intercalibration to assure consistency with the historical record. Further calibration and processing will be necessary, which may slightly affect final reported ice extent values (on average +/- 30,000 square kilometers or 11,600 square miles per preliminary number reported).

Map of sea ice from space, showing sea ice, continents, ocean

Figure 1. Arctic sea ice extent for May 2008 was 13.18 million square kilometers (5.09 million square miles).  The magenta line shows the median ice extent for May from 1979 to 2000. Data information
—Credit: National Snow and Ice Data Center

Overview of conditions

Arctic sea ice extent for May stood at 13.18 million square kilometers (5.09 million square miles), which is 0.28 million square kilometers (0.11 million square miles) greater than May 2007, but is still 0.42 million square kilometers (0.16 million square miles) less than the 1979 to 2000 average for the month.

Graph with months on x axis and extent on y axis

Figure 2. Daily sea ice extent; the blue line indicates 2008; the black line indicates extent from 1979 to 2000; the dotted line shows extent for 2007.
—Credit: National Snow and Ice Data Center

See High Resolution Image

Conditions in context

Although ice extent is slightly greater than this time last year, the average decline rate through the month of May was 8,000 square kilometers per day (3,000 square miles per day) faster than last May. Ice extent as the month closed approached last May’s value.

Average Arctic Ocean surface air temperatures in May were generally higher than normal. While anomalies were modest  (+1 to 3 degrees Celsius, +2 to 5 degrees Fahrenheit) over most of the region, temperatures over the Baffin Bay region were as much as 6 degrees C (11 degrees F) above normal. The atmospheric circulation in May was highly variable. The first half of the month saw strong winds blowing from east to west over the southern Beaufort Sea. This wind pattern probably contributed to polynya formation near Banks Island and along the northwestern coast of Alaska.

View of Arctic from above

Figure 3. Infrared energy that the atmosphere emits to the surface during spring shows generally positive trends. Units are change in long-wave energy transfer per decade between 1979 and 2005; yellow and red colors are positive trends; white indicates regions without data. Derived from NOAA polar-orbiting satellites.

—Credit: From National Snow and Ice Data Center courtesy J. Francis, Institute of Marine and Coastal Sciences, Rutgers University See High Resolution Image

More on the sea ice-atmosphere connection

The more we study the Arctic’s shrinking sea ice cover, the more we appreciate the key role of clouds and water vapor. Our colleague, Jennifer Francis of Rutgers University, has linked changes in the ice edge northwest of Alaska to variations in springtime cloudiness and in the water vapor content of the lower atmosphere. She has observed an increase in springtime cloud and water vapor over the last three decades that can be clearly linked to retreat of the ice edge.

What is the nature of this link? More clouds act like an umbrella, shading the sea ice surface from the sun’s rays, also called solar radiation. At the same time, clouds act like a warm blanket, transferring heat in the form of long-wave radiation from the atmosphere to the ice surface. More water vapor in the atmosphere contributes to the blanket-like effect. Whether the umbrella or blanket effect dominates determines how much radiation is absorbed at the surface, which in turn influences the rate of ice melt. In spring, solar radiation is still relatively weak. Because of this, the blanketing effect of increased clouds and water vapor wins.

In the summer, the situation is reversed. Clear skies allow the strong radiation of the summer sun to reach the surface and melt sea ice. Anticyclone patterns set up these clear summer conditions. We will be watching closely for the possible onset of these conditions in coming months.

View of Arctic from above showing ice age

Figure 4. This United States National Ice Center analysis shows the percentage of multi-year sea ice in yellow, green, and dark blue. Light blue with red outline indicates ice extent; land and ocean are white.

—Credit: From National Snow and Ice Data Center courtesy United States National Ice Center See High Resolution Image

Multi-year ice continues to be low

The relative lack of thick, resilient multi-year ice in the Arctic discussed in earlier postings finds further support in the latest analysis from the United States National Ice Center (NIC). NIC uses a variety of satellite imagery, expert analysis, and other information to provide information on the amount and quality of sea ice for ships operating in the Arctic. NIC scientist Todd Arbetter suggests that much of the first-year ice is likely to melt by the end of summer, saying that despite the total ice extent appearing normal, the relative amount of multi-year ice going into this summer is very low when compared to climatological averages. NIC has found that the relative fraction of multi-year ice in the central Arctic has plummeted since the mid-1990s, creating an Arctic prone to increased melt in summer. Arbetter said, “This may be a primary reason for record summertime minimums in recent years.”

However, the unusual location of some of this year’s first-year ice may help more of it survive than otherwise might be expected. This year, much of the first-year ice is farther north than normal, and those northern areas receive weaker solar radiation. So, northern first-year ice may be less vulnerable to melt than first-year ice in typical locations.

"photo" of Arctic from space showing sea ice, clouds, landmasses

Figure 5. Open water is clearly seen near Alaska and Banks Island, and in the North Water polynya, in this visible-band satellite image mosaic on May 20, 2008. MODIS Terra and Aqua satellite data.

—Credit: National Snow and Ice Data Center
See High Resolution Image

Thinner ice already showing weakness

As mentioned, the thin ice that covers much of the Arctic Ocean is showing signs of early breakup, with large polynyas off the coast of Alaska, the Canadian Archipelago, and Baffin Bay. Coastal polynyas are not unusual, at this time of year, but the polynyas we are currently seeing appear larger and more numerous than usual. This is partly because of the thinner, weaker ice cover.

Thorsten Markus at the NASA Goddard Space Flight Center has noted the size of the North Water polynya at the northern end of Baffin Bay, which typically forms in May. The polynya is much larger than normal, possibly nearing its largest area on record.

Inuit report that sea ice is starting to break up near Baffin Bay much earlier than normal this year. They have observed wide cracks in the ice already forming, according to NSIDC scientist Shari Gearheard, who lives and works in the Baffin Island hamlet of Clyde River.

Polynyas are a source of heat for the atmosphere in spring; in summer, however, they are large absorbers of solar energy. Resultant warm ocean surface waters then eat away at the ice edge, accelerating melt.

References

Francis, J.A. and E. Hunter. 2006. New insight into the disappearing Arctic sea ice. Eos, Trans. Amer. Geophys. Union 87,509-524.

Francis, J.A. and E. Hunter. 2007. Changes in the fabric of the Arctic’s greenhouse blanket. Environmental Research Letters 2, doi:10.1088/1748-9326/2/4/045011.

Markus, T. , and B.A. Burns. 1995. A method to estimate subpixel-scale coastal polynyas with satellite passive microwave data. J. Geophys. Res.100, 4473-4487.

For previous analysis, please see the drop-down menu under Archives in the right navigation at the top of this page.

Arctic sea ice forecasts point to lower-than-average season ahead

Spring has arrived in the Arctic. After peaking at 15.21 million square kilometers (5.87 million square miles) in the second week of March, Arctic sea ice extent has declined through the month of April. April extent has not fallen below the lowest April extent on record, but it is still below the long-term average.

Taken together, an assessment of the available evidence, detailed below, points to another extreme September sea ice minimum. Could the North Pole be ice free this melt season?  Given that this region is currently covered with first-year ice, that seems quite possible.

Map of sea ice from space, showing sea ice, continents, ocean

Figure 1. Arctic sea ice extent for April 2008 was 14.49 million square kilometers (5.59 million square miles).  The magenta line shows the median ice extent for March from 1979 to 2000. Data Note

—Credit: National Snow and Ice Data Center

See High Resolution Image

Overview of conditions

For the month of April, Arctic sea ice extent stood at 14.49 million square kilometers (5.59 million square miles), which is 0.61 million square kilometers (0.24 million square miles) greater than April 2007, but is still 0.51 million square kilometers (0.20 million square miles) less than the 1979 to 2000 average for April.

Graph with months on x axis and extent on y axis

Figure 2. Daily sea ice extent; the blue line indicates 2008; the black line indicates extent from 1979 to 2000; the dotted line shows extent from December 2006 through April 2007.
—Credit: National Snow and Ice Data Center

See High Resolution Image

Conditions in context

Although there is more ice than this time last year, the average decline rate through the month of April was 6,000 square kilometers per day (2,300 square miles per day) faster than last April.

Figure 3. The spatial pattern of surface air temperature anomalies for April 2008, expressed with respect to the average for 1979 to 2007, shows unusually high temperatures over the Arctic Ocean and peripheral seas.

—Credit: From National Snow and Ice Data Center courtesy Climate Diagnostic Center

See High Resolution Image

Faster decline reflects warmer Arctic

At least part of the explanation for this fairly rapid decline lies in the warm conditions that characterized April over the Arctic Ocean and peripheral seas. Anomalies over some regions exceed 5 degrees Celsius (9 degrees Fahrenheit). For the most part, this unusual warmth is consistent with shifts in atmospheric circulation that bring warm air into the region. The distinct hot spot near Novaya Zemlya, in the upper left quadrant of Figure 3, overlies an open water area where heat is being released to the atmosphere. In past years, this area tended to be ice covered in April, preventing this heat release.

Bar graph showing estimate of 2008 sea ice minimum based on known survival rates.

Figure 4. This bar plot shows estimates of sea ice extent at the 2008 September minimum based on known ice survival rates. The blue dotted line indicates the record-breaking minimum extent of 2007; the red dotted line shows the mean estimate based on all years between 1983 and 2007.

—Credit: National Snow and Ice Data Center

See High Resolution Image

Estimating September extent based on past conditions

As discussed in our April analysis, the ice cover this spring shows an unusually large proportion of young, thin first-year ice; about 30% of first-year ice typically survives the summer melt season, while 75% of the older ice survives. For a simple estimate of the likelihood of breaking last year’s September record, we can apply survival rates from past years to this year’s April ice cover.  This gives us 25 different estimates, one for each year that we have reliable ice-age data (see Figure 4).  To avoid beating the September 2007 record low, more than 50% of this year’s first-year ice would have to survive; this has only happened once in the last 25 years, in 1996. If we apply the survival rates averaged over all years to current conditions, the end-of-summer extent would be 3.59 million square kilometers (1.39 million square miles). With survival rates similar to those in 2007, the minimum for the 2008 season would be only 2.22 million square kilometers (0.86 million square miles). By comparison the record low extent, set last September, was 4.28 million square kilometers (1.65 million square miles).

Forecasting September extent with climate predictors

Sheldon Drobot at the Center for Astrodynamics Research at the University of Colorado at Boulder and colleagues have developed a sophisticated forecasting technique. The forecast considers sea ice extent, ice age, summer and winter temperatures, cloudiness, the phase of the Atlantic Oscillation, and climate trends as predictors (see the papers cited below for details; visit the Arctic Oscillation Index).  As reported last month, the Arctic Oscillation was in its positive phase through the winter season, associated with a wind pattern helping to flush thick ice out of the Arctic, leaving thinner ice.  This is one of the factors helping to set the stage for pronounced ice losses this summer. Drobot predicts a 59% chance of a new record minimum this year; read the press release. Todd Arbetter of the U.S. National Ice Center tells us that his group is working to implement a version of Drobot’s analysis scheme for operational forecasting.

Ronald Lindsay of the University of Washington’s Applied Physics Laboratory and collaborators recently published results from their own ice prediction system, based on a retrospective analysis of the modeled state of the ice and ocean system (see the paper cited below for details).  The model is successful in explaining around 75% of the year-to-year variations for the past few decades; for 2008, the model implies a very low, but not extreme, sea ice minimum. Lindsay cautions that sea ice conditions are now changing so rapidly that predictions based on relationships developed from the past 50 years of data may no longer apply.

Map of Hudson Bay showing colors based on probable conditions.Figure 5. This image shows probable ice conditions in the Hudson Bay for July 2008; the colored area is the bay and white indicates land masses. Green shows near-normal ice conditions; red shows below average; blue shows above average.

—Credit: From National Snow and Ice Data Center courtesy A.Tivy

See High Resolution Image 

Regional shipping forecasts

Marine transportation in the Arctic is expected to increase as ice extent decreases. However, the viability of shipping through the Northwest Passage in the Canadian Arctic Islands, the Northern Sea Route along the Eurasian coast and in other areas such as Hudson Bay depend on local ice conditions, which can be highly variable. Adrienne Tivy  at the University of Calgary and colleagues have investigated the variables that affect shipping in Hudson Bay. They found that the date on which shipping routes open across Hudson Bay to Churchill is most strongly correlated with sea surface temperatures between January and March and atmospheric pressure patterns in the East Atlantic in January. This year, Adrienne Tivy and colleagues predict that shipping to Churchill in a non-ice-strengthened vessel will be possible on July 16, 15 days earlier than the long-term mean of July 31. They estimate below-normal ice concentrations in the southwestern bay, but near-normal conditions elsewhere.

References

Drobot, S.D. 2007. Using remote sensing data to develop seasonal outlooks for Arctic regional sea-ice minimum extent. Remote Sensing of Environment, 111, 136-147,doi:10.1016/j.rse.2007.03.024.

Drobot, S.D., J.A. Maslanik, and C.F. Fowler. 2006. A long-range forecast of Arctic summer sea-ice minimum extent. Geophysical Research Letters, 33, L10501, doi:10.1029/2006GL026216.

Drobot, S.D., 2003: Long-range statistical forecasting of ice severity in the Beaufort/Chukchi Sea, Weather and Forecasting, 18:1161 – 1176.

Lindsay, R. W., J. Zhang, A. J. Schweiger, and M. A. Steele. 2008.  Seasonal predictions of ice extent in the Arctic Ocean.  Journal of Geophysical Research, 113, C02023, doi:10.1029/2007JC004259

Tivy, A., B. Alt, S. Howell, K. Wilson, and J. Yackel. 2007. Long-range prediction of the shipping season in Hudson Bay: A statistical approach. Weather and Forecasting, 22, 1063–1075, doi:10.1175/WAF1038.1

Zhang, J., M. Steele, R. Lindsay, A. Schweiger, and J. Morison. 2008. Ensemble 1-year predictions of Arctic sea ice for the spring and summer of 2008, Geophysical Research Letters, 35, L08502, doi:10.1029/2008GL033244.

Arctic sea ice extent at maximum below average, thin

Arctic sea ice reached its yearly maximum extent during the second week of March, 2008. Maximum extent was slightly greater compared to recent years, but was still well below average.

Despite strong growth of new ice over the winter, sea ice is still in a general state of decline. The ice that grew over the past winter is relatively thin, first-year ice that is susceptible to melting away during the summer. Although natural variability in the atmospheric circulation could prevent the ice pack from breaking last year’s summer record, a closer look at sea ice conditions indicates that the September 2008 minimum extent will almost certainly be well below average.

Please credit the National Snow and Ice Data Center for image or content use unless otherwise noted beneath each image.

Map of sea ice from space, showing sea ice, continents, ocean

Figure 1. Arctic sea ice extent for March 2008, the winter maximum, was 15.2 million square kilometers (5.9 million square miles).  The magenta line shows the median ice extent for March from 1979 to 2000.
Please note that our daily sea ice images, derived from microwave measurements, may show spurious pixels in areas where sea ice may not be present. These artifacts are generally caused by coastline effects, or less commonly by severe weather. Scientists use masks to minimize the number of “noise” pixels, based on long-term extent patterns. Noise is largely eliminated in the process of generating monthly averages, our standard measurement for analyzing interannual trends. Data derived from Sea Ice Index data set.
—Credit: National Snow and Ice Data Center

See High Resolution Image

Overview of conditions

Arctic sea ice reached its yearly maximum on March 10, 2008, at 15.21 million square kilometers (5.87 million square miles). Average sea ice extent for March 2008 was 15.2 million square kilometers (5.9 million square miles).

Graph with months on x axis and extent on y axis

Figure 2. Daily sea ice extent; the blue line indicates December 2007 through April 1, 2008; the black line indicates extent from 1979 to 2000; the dotted line shows extent from December 2006 through April 2007.
—Credit: National Snow and Ice Data Center

See High Resolution Image

Conditions in context

As Arctic sea ice extent shrank through the summer of 2007 to its record-setting minimum in September, the large open-water areas absorbed  a great deal of the sun’s energy.  Because the Arctic Ocean needed to lose this heat before sea ice could form, autumn freeze-up began rather slowly.   Once freeze-up began, it proceeded very quickly.

As Figure 2 shows, maximum sea ice extent usually occurs during the first week of March. Ice extent then begins its seasonal decline as springtime warming takes hold. In 2008, the maximum extent occurred about a week later than normal, with the extent below average.

Graph showing year on x axis and extent on y axis

Figure 3. Average March ice extent for 1979 through 2008.
—Credit: National Snow and Ice Data Center

See High Resolution Image

March 2008 compared to Marches past

March 2008 monthly maximum extent was 780,000 square kilometers (301,000 square miles) greater than the past record low, set in March 2006, but 540,000 square kilometers (208,000 square miles) less than the 1979 to 2000 mean. Including 2008, the linear trend for March indicates that the Arctic is losing an average of 44,000 square kilometers (17,000 square miles) of ice per year in March. Although March 2008 extent is greater than in recent years, the setup looks right for another dramatic ice loss this summer; for details, see below.

Two maps of sea ice age side-by-side

Figure 4. Map of estimated ice age for the third week of March for 2007 (left) and 2008 (right). Dark blue indicates first-year ice; red indicates ice that is 6 years old or more; grey is land and white indicates areas where ice age is not tracked. Colleagues James Maslanik, Chuck Fowler, and Sheldon Drobot at the University of Colorado at Boulder developed this image.

—Credit: Image from National Snow and Ice Data Center courtesy Maslanik, Fowler, Drobot

See High Resolution Image

Bar graph with year on x axis and ice cover on y axis

Figure 5. Percent of ice cover in the Arctic Basin (70 to 90 degree North Latitude) that was first-year ice from 1985 to 2008. Blue bars indicate first-year ice at the beginning of the melt season; red bars indicate first-year ice left at the end of the melt season.
—Credit: National Snow and Ice Data Center

See High Resolution Image

New ice growth over winter 2007/2008

As the winter extent numbers indicate, new ice growth was strong over the winter. Nevertheless, this new ice is probably fairly thin.  Thin ice is vulnerable to melting away during summer. Figures 4 and 5 indicate that relatively thin, first-year ice now covers 72% of the Arctic Basin, including the region around the North Pole; in 2007, that number was 59%.  Usually, only 30% of first-year ice formed during the winter survives the summer melt season; in 2007, only 13% survived. Even if more first-year ice survives than normal, the September minimum extent this year will likely be extremely low.

Why is there so much first-year ice this spring?  Partly, it is because last summer’s record-breaking ice loss created extensive open-water areas in which new ice could form.  Anomalous winds in winter can also flush thicker, older ice out of the Arctic, leaving the Arctic with a greater coverage of first-year ice.  As noted by our colleague Ignatius Rigor of the University of Washington at Seattle, this winter saw a return of the Arctic Oscillation to its positive mode, an atmospheric pattern especially effective in flushing out thick, old ice.

So what about the multi-year ice that remained after last year’s record ice loss? Jennifer Kay and colleagues at the National Center for Atmospheric Research found that last summer’s clear skies allowed for more intense melt of the multiyear ice, leaving it  thinner than normal at summer’s end.

For more on the Arctic’s transition towards younger ice, see an animation of changing sea ice age (scroll to Figure 4 of August 22, 2007 entry) by colleague Jim Maslanik and coauthors.

Two maps side-by-side of sea ice freeboard

Figure 6. Freeboard for winter 2007 (left) and 2008 (right). Purple indicates no freeboard; red indicates ice freeboard 80 centimeters (2.6 feet) thick or greater; grey is land and white indicates no data. Colleague Ronald Kwok from Jet Propulsion Laboratory supplied this image.
—Credit: Image from National Snow and Ice Data Center courtesy Kwok

See High Resolution Image

A look at sea ice thickness

Another way to study sea ice thickness is to look at freeboard, or the amount of ice and snow that protrudes above the water surface. New information on ice thickness is coming from NASA’s ICESat instrument, a spaceborne laser altimeter.  Colleague Ronald Kwok at the Jet Propulsion Laboratory uses data from ICESat to study freeboard. His findings indicate that  freeboard in the spring of 2008 is 5 to 10 centimeters (2 to 4 inches) less than in spring 2007, pointing to thinner sea ice.

Overlying snow cover complicates calculating sea ice thickness from freeboard.  For example, a region with thick snow cover over thin ice might show a similar freeboard to a region with thin snow cover over thicker ice. Ron Kwok and colleagues developed methods to estimate snow depth that will be applied to these preliminary ICESat results in the near future.

References

Ignatius Rigor. April 1, 2008. Personal contact with Julienne Stroeve.

Kay, J.E., T. L’Ecuyer, A. Gettelman, G. Stephens and C. O’Dell. The contribution of cloud and radiation anomalies to the 2007 Arctic sea ice extent minimum, Geophysical Research Letters, in press.

Kwok, R. and G. F. Cunningham. 2008. ICESat over Arctic sea ice: Estimation of snow depth and ice thickness, J. Geophys. Res., submitted.

Kwok, R., G. F. Cunningham, H. J. Zwally, and D. Yi. 2007. Ice, Cloud, and land Elevation Satellite (ICESat) over Arctic sea ice: Retrieval of freeboard, J. Geophys. Res., 112, C12013, doi:10.1029/2006JC003978.

Maslanik, J.A., C. Fowler, J. Stroeve, S. Drobot, J. Zwally, D. Yi and W. Emery. 2007.  A younger, thinner Arctic ice cover: Increased potential for rapid extensive sea ice loss, Geophysical Research Letters, 34, L24501, dio:10.1029/2007/GL032043.

Stroeve, J., M. Serreze, S. Drobot, S. Gearheard, M. Holland, J. Maslanik, W. Meier, T. Scambos. 2008.  Arctic sea ice extent plummets in 2007, EOS Trans., AGU, 89(2), 13-14.