SciELO - Scientific Electronic Library Online

Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados


Gayana (Concepción)

versión impresa ISSN 0717-652Xversión On-line ISSN 0717-6538

Gayana (Concepc.) v.68 n.2 supl.TIProc Concepción  2004 


Gayana 68(2) supl. t.I. Proc. : 123-129, 2004 ISSN 0717-652X



Josefino C. Comiso

NASA Goddard Space Flight Center Greenbelt, MD, USA 20771 E-mail:, tel: 1-301-614-5708, FAX: 1-301-614-5644

Global warming of about 0.5 to 1 oC during the last century has been observed from meteorological stations around the world and postulated to be caused in part by increasing anthropogenic greenhouse gases in the atmosphere. The best regions for detecting the impact of such changes are the polar regions where the signals are supposed to be amplified because of ice-albedo feedbacks. Satellite data reveal that the most remarkable change in the polar regions is in the rapid decline of 9% per decade in the Arctic perennial ice cover which has been observed to be anomalously low in 2002, 2003 and 2004. The sea ice cover in the Bellingshaussen Sea has been observed to be retreating as well, at the rate of 6%/decade. The changes are more modest in the entire Northern Hemisphere with the ice declining at around 2 to 3% per decade only while that in the Southern Hemisphere has been basically constant during the last 25 years.



The study of sea ice, which is the frozen portion of the world's oceans, is important for many reasons. First, sea ice is one of the key components of the climate system. It serves as a thermal blanket that limits the exchange of heat between the ocean and the atmosphere, especially during the winter. It also has a very high albedo (compared to that of open ocean), causing much of the sun's incident radiation to be reflected back to the atmosphere. Sea ice also affects (or is impacted by) the oceans in many ways. During the freezing stages, sea ice causes the formation of cold and high salinity water that can initiate deep ocean convection or generate the bottom water that circulates around the global oceans. During spring and summer, it causes the release of low density melt water that provides a relatively stable ocean layer that is conducive to phytoplankton blooms thereby enhancing biological productivity during the period. In a sense, during growth and decay, sea ice causes the redistribution of salt in the ocean, making the surface water more salty where they form and less salty where they melt. It is also an important habitat to many forms of life from microorganisms to mammals. On a more practical term, sea ice impacts ship navigation, oil exploration, and other human activities. It is also expected that the sea ice cover will provide an early signal of a global climate change because of ice-albedo feedbacks that has been shown by models to cause amplification of the signal in the polar regions.

The sea ice cover is one of the most seasonal parameters on the Earth's surface. As climate parameters, it is important to recognize that the variability of the sea ice cover in the Arctic is quite different from that of the Antarctic. The sea ice cover in the Arctic is surrounded by land and has an extensive perennial ice that serves to stabilize the characteristics of the Arctic Ocean. On the other hand, the Antarctic ice cover has no outer land boundaries that limits the growth and advance of sea ice in the region. In winter, much of the ice growth in the Arctic occurs in the peripheral seas while in the Antarctic, the ice cover expands from the coastal regions of Antarctica until atmospheric and oceanic forces limit its northward advance, while in the spring and summer, the ice retreats rapidly because of its vulnerability to atmospheric and oceanic heat and wave action.

Recent reports of a declining extent [Johannessen et al., 1995; Cavalieri et al., 1997; Jacobs and Comiso, 1997] and thickness [Rothrock et al., 1999; Wadhams and Davis, 2000] makes it imperative that the variability of the ice cover is examined in detail. The large scale variability of the sea ice cover has been quantified previously using satellite passive microwave data [Bjorgo et al., 1997; Parkinson et al., 1999; Zwally et al., 2002]. Large changes in the ice cover have been reported in the 1990s and the more recent years, especially in the Arctic, where the perennial ice cover have been declining at the rate of 9% per decade (Comiso, 2002, Comiso and Parkinson, 2004). This paper updates previous studies and provides an analysis of the current state, variability and trends in the sea ice cover using passive microwave satellite data from 1978 to 2004.

Satellite Observations

Detailed studies of the global sea ice cover have been made possible by the advent of satellite data that have been providing comprehensive areal coverage at a relatively high temporal resolution. The use of satellite data for sea ice studies have been discussed in numerous publications [e.g., Massom, 1991; Carsey, 1992]. Satellite data can be divided into visible, infrared, and microwave data, which are sometimes used in concert to obtain as complete understanding of the properties of the ice cover as possible, and sometimes used independently for some unique applications. Passive and active microwave systems have been the more popular techniques for monitoring sea ice because of day/night almost all-weather capabilities and global coverage at a relatively high temporal resolution. For large scale variability and trend studies, passive microwave data have been the primary tool because of a relatively long, comprehensive, and consistent historical record. An apparent weakness of passive microwave data is the coarse spatial resolution (about 25 by 25 km) which makes such data difficult to use for detecting small spatial features in the ice cover, such as leads, ridges, and ice bands. To partly overcome this weakness, the percentage of ice (called ice concentration) within each satellite grid element is calculated using mixing algorithms, as discussed earlier, that takes advantage of the high contrast of the emissivity of ice and open water.

Global and Regional Trends

Global climate change is expected to be amplified in polar regions because of feedback effects associated with the high albedo of ice and snow as indicate earlier. A global warming of 0.5 K has been observed during the last century [Jones et al., 1999] and it would be of interest to know how such trend is reflected in the Arctic and the Antarctic regions. Linear regression analyses have been applied on the monthly sea ice anomalies for the period from 1979 through 2003 and results are presented in Figures 1 and 2 for the Northern and Southern Hemispheres, respectively. The trend in ice extent in the Northern Hemisphere as inferred from monthly anomaly data from January 1979 through December 2003 is ­2.2 ± 0.3 % per decade. This is less than the rate of about -3 %/decade inferred previously from slightly different data sets and for the November 1978 to December 1996 period [Bjorgo et al., 1997; Parkinson et al, 1999]. It is thus important to note that the addition of a few years may cause significant change in the trend, especially if the latest values are very different from previous ones. The issue of accuracy associated with record lengths of data is addressed in the next section.

In the Central Arctic region, the trend is ­0.9 ± 0.2 % per decade, which is a lesser negative rate than that for the entire Northern Hemisphere. The lesser rate is not unexpected since much of the anomalies are in the peripheral seas as discussed in the previous sections. The plot in Fig. 1b indicates spikes associated with changes in the spring and summer periods which are shown to be mainly positive in the 1980s and negative (except for 1992, 1994, and 1996) in the 1990s. The negative trend appears to be primarily influenced by unusually low ice extent events of 1990, 1993, and 1995.

The trends in other sectors are varied with the only positive trend occurring in the Bering Sea at 6.4 ± 2.5 % per decade. This positive trend reflects increases in the winter ice cover in the region despite decreases in the ice cover in the adjacent (Arctic) region. Also, towards the southwest of this region at the Okhotsk Sea, the ice cover is declining at about the same magnitude but opposite sign with the trend being ­4.5 ± 2.5 % per decade. In the latter, the trend would have been significantly more negative were it not for the relatively strong recovery during the last two years of data. Other regions with significant negative trends are the Kara/Barents Seas and Hudson Bay with trends of ­5.2 ± 1.0 % per decade and ­4.5 ± 0.6 % per decade respectively. The trend at the Canadian Archipelago is also significant at ­1.2 ± 0.3 % per decade but this is mainly due to a big drop in the ice cover in the region in 1998. Trends are also negative in the Greenland Sea and Baffin Bay/Labrador Sea sectors at ­5.2 ± 1.1 and ­4.5 ± 1.2 % per decade, respectively, but changes in these regions may be partly influenced by periodic patterns as mentioned previously. A negative trend in the ice cover is also inferred at the small Gulf of St. Lawrence sector (­5.2 ± 4.2 % per decade) where a periodic pattern is also observed.

While the trend of ice cover in the entire Northern Hemisphere is basically controlled by trends in the peripheral seas, the most important trend of interest is actually that associated with the Central Arctic. The Northern Hemisphere has smaller seasonal amplitude than the Southern Hemisphere because of the presence of extensive and thick multiyear ice cover in the Central Arctic. Ice that survives the summer is often referred to as the perennial ice cover, which consists mainly of multiyear ice. The extent and area of the perennial ice cover can be inferred by finding the minimum value for each year from the time series of daily data.

The trends in maximum ice values for both extent and ice area have also been evaluated and the results are ­1.3 ± 0.6, and ­1.6 ± 0.6, respectively, and are basically consistent with those of the entire hemisphere updated to 2004 which are -2.4 and -3.3, respectively (Fig. 2a). The trends in minimum ice cover, however, are ­7.3 ± 1.7 %/decade and ­9.2 ± 1.5 %/decade for ice extent and ice area, respectively (Fig. 2b). Such large negative trends (about 6 times higher than maximum trends) indicate that the perennial ice cover that has been reported to be shrinking at a relatively high rate by Comiso [2002] is now declining at an even faster rate. A trend for the multiyear ice cover of a similar magnitude (7%) has also been reported by Johannessen et al. [1999] using multiyear ice fraction data derived during winter period from passive microwave data. However, such winter data are known to have large uncertainties [Kwok et al., 1996] partly because the emissivity of multiyear ice is highly variable and not constant as assumed in the algorithm that calculates the multiyear ice fraction.

Figure 2: (a) Monthly Sea ice anomalies in (a) the Northern Hemisphere and (b) yearly Perennial Ice Cover

It is of special interest to note that the ice minimums in the 1990s are highly variable compared to those in the 1980s. As reported in Comiso [2002] such variability suggests an increasing fraction of second year ice cover in the 1990s, compared to those in the 1980s, which in effect means a thinning of the Arctic sea ice cover (that includes the second year ice cover). This is consistent with reports by Rothrock et al. [1999] and Wadhams and Davis [2000] that the average ice thickness in the 1990s is significantly less than averages collected in the previous decades. A thinning through this mechanism would not necessarily mean a thinning in the older and thicker ice floes.

The trend results in the Southern Hemisphere (Fig. 3) are not as indicative of a changing ice cover as those of the Northern Hemisphere. The result of analysis of data from the 25-year period shows an insignificant trend of 0.3 ± 0.3 % per decade for the entire Southern Hemisphere. This is slightly less than the 1.0 ± 0.4 % per decade reported by Zwally et al. [2002] for the 1979 to 1998 period. The trends are similarly insignificant at 0.0 ± 0.7, -0.4 ± 0.7, and -0.3 ± 1.0 % per decade at the Weddell Sea, Indian Ocean, and Western Pacific sectors, respectively. In two other sectors, however, the trends are quite substantial with opposite signs of 5.3 ± 0.8 % per decade and ­7.2 ± 1.1 % per decade at the Ross Sea and Bellingshausen/Amundsen Seas sectors, respectively. Since the two sectors are adjacent to each other, it appears that the opposite trends in the two sectors are partly caused by advection of ice from one sector to the other. However, the Antarctic Penninsula that is adjacent to the B/A sector has been an area of climate anomaly as described previously by King [1994] and Jacobs and Comiso [1997]. Also, the Ross Sea region has been associated with influences of ENSO [Ledley and Huang, 1997] and the continental area adjacent to it has been experiencing some cooling during the last two decades [Comiso, 2000]. The positive trend in the Ross Sea, which is the site of a major coastal polynya, suggests increased ice production and a more important role of the region in bottom water formation.

The trends of maximum ice values for each year in the Southern Hemisphere is again similar to that inferred from the monthly time series, being 0.3 ± 0.5 % per decade and 1.2 ± 0.6 % per decade for ice extent and ice area, respectively. It is, however, surprising that the trends of minimum values for each year are negative at ­3.6 ± 2.8 and ­3.8 ± 3.2% per decade for ice extent and ice area, respectively. It appears that the summer ice cover has been declining as in the Arctic but errors are larger because of much lower extent of ice in the summer. The negative trend is due in part to the retreating perennial ice cover in the Bellingshausen Sea as reported by Jacobs and Comiso (1997).


The hemispheric changes in the sea ice cover in the last 25 years have been relatively modest with slight decreases in the Northern Hemisphere and practically no change in the Southern Hemisphere. Changes on a regional basis, however, have been substantially more varying from negative to positive trends from one region to another. Unexpected changes include the positive trend in the Bering Sea region which has been identified as an area of cooling by infrared data. An expected change is the decline in the ice cover at the Bellingshausen Sea/Amundsen Sea area, the region adjacent to the Antarctic Peninsula, which has been observed as a climatically anomalous region. The most remarkable change, however, is the retreat of the Arctic perennial ice cover which is shown to be declining more rapidly than in previous years. The record low for the Arctic perennial ice cover was observed in 2002 and this was followed by almost equally low values in 2003 and 2004. The larger open water areas created in the process may further accelerate the decline because of the ice-albedo feeback. A recovery would require long periods of anomalously cold temperatures in the Arctic summer but this may not happen in the near future because of currently observed global warming that is amplified in the Arctic.


Bjørgo, E., Johannessen, O.M. & M.W. Miles, M.W. (1997) Analysis of merged SSMR-SSMI time series of Arctic and Antarctic sea ice parameters 1978-1995, Geophysical Research Letter, 24(4), 413-416. [         [ Links ]1]

Budyko, M.I. (1966) Polar ice and climate, In Proceedings of the Symposium on the Arctic Heat Budget and Atmospheric Circulation, ed. By J.O. Fletcher, pp. 3-32, RM 5233-NSF, Rand Corposration, Santa Monica, CA. [         [ Links ]2]

Carsey, F., (1992) Microwave Remote Sensing of Sea Ice, AGU Geophysical Monograph 68, Washington, D.C. [         [ Links ]3]

Cavalieri, D.J., Gloersen, P., Parkinson, C., Comiso, J.C. & H.J. Zwally, H.J. (1997) Observed Hemispheric asymmetry in global sea ice changes, Science, 278(7), 1104-1106. [         [ Links ]4]

Comiso, J. C., 2002, A rapidly declining Arctic perennial ice cover, Geophys Res. Letts., 29(20), 1956, doi:10.1029/2002GL015650. [         [ Links ]5]

Comiso, J.C. (2000) Variability and trends in Antarctic surface temperatures from in situ and satellite infrared measurements, Journal of Climate, 13(10), 1674-1696, 2000. [         [ Links ]6]

Comiso, J. C. & C. L. Parkinson, (2004) Satellite observed changes in the Arctic, Phys. Today 57(8), 38-44, 2004. [         [ Links ]7]

Comiso, J.C. (2002) Correlation and Trend Studies of the Sea Ice Cover and Surface Temperatures in the Arctic, Annals of Glaciology, 34 (in press, July 2002). [         [ Links ]8]

Jacobs, S.S. & Comiso, J. C. (1997) A climate anomaly in the Amundsen and Bellingshausen Seas, Journal of Climate, 10(4), 697-709, 1997. [         [ Links ]9]

Johannessen, O.M., Shalina, E.V. & Miles, M.W. (1999) Satellite evidence for an Arctic sea ice cover in transformation, Science, 286, 1937-1939. [         [ Links ]10]

Jones, P.D., New, M. Parker, D.E., Martin, S. & Rigor, I.G. (1999) Surface air temperature and its changes over the past150 years, Reviews in Geophysics, 37, 173-199. [         [ Links ]11]

King, J.C. (1994) Recent climate variability in the vicinity of the Antarctic peninsula, International Journal of Climatology, 14, 357-369. [         [ Links ]12]

Kwok, R., J.C. Comiso & G. Cunningham, Seasonal characteristics of the perennial ice cover of the Beaufort Sea, Journal of Geophysical Research,101(C2), 28417-28439, 1996. [         [ Links ]13]

Ledley, T.S. & Huang, Z. (1997) A possible ENSO signal in the Ross Sea, Geophysical Research Letters, 24, 3253-3256. [         [ Links ]14]

Manabe, S., Spelman, M.J. & Stouffer, R.J. (1992) Transient responses of a coupled ocean-atmosphere model to gradual changes of atmospheric CO2 part II, Seasonal response, Journal of Climate, 5, 105-126. [         [ Links ]15]

Massom, R.A., (1991) Satellite Remote Sensing of Polar Regions, Bellhaven Press, 307pp. [         [ Links ]16]

Massom, R.A., Comiso, J.C., Worby, A.P., Lytle, V. & Stock, L. (1999) Satellite and in situ observations of regional classes of sea ice cover in the East Antarctic pack in winter, Remote Sensing of the Environment., 68(1), 61-76. [         [ Links ]17]

Parkinson, C.L., Cavalieri, D.J., Gloersen, P., Zwally H.J. & Comiso, J.C. (1999) Arctic Sea ice extents, areas, and trends, 1978-1996, Journal of Geophysical Research, 104(C9), 20837-20856. [         [ Links ]18]

Rothrock, D. A., Yu, Y. & Maykut, G.A. (1999) Thinning of the Arctic sea-ice cover, Geophysical Research Letters, 26(23), 3469-3472. [         [ Links ]19]

Wadhams, P. & N. Davis, N. (2000) Further evidence of ice thinning in the Arctic Ocean, Geophysical Research Letters, 27(24), 3973-3976. [         [ Links ]20]

Zwally, H.J., Comiso, J.C., Parkinson, C.L., Cavalieri, D.J., & Gloersen, P. (2002) Variability of the Antarctic Sea Ice Cover, Journal of Geophysical Research, 107(C5), 1029-1047. [         [ Links ]21]


Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons