Climate is the average state of the atmosphere and the underlying land or water on time scales of seasons and longer. It is typically described by the statistics of a set of atmospheric and surface variables, such as temperature, precipitation, wind, humidity, cloudiness, soil moisture, sea surface temperature, and the concentration and thickness of sea ice (1). It may be inferred that any departure from the global mean of any one of these variables would indicate a change in the global climate.
In 2001, the White House submitted a request to the National Academies’ Committee on the Science of Climate Change (CSCC), asking for “assistance in identifying the areas in the science of climate change where there are the greatest certainties and uncertainties” (1). The CSCC’s subsequent report included, among other conclusions, “Temperatures are, in fact, rising. The changes observed over the last several decades are likely mostly due to human activities, but we cannot rule out that some significant part of these changes also reflect a natural variability” (1). The committee further suggested that “predictions of global climate change will require major advances in the understanding and modeling of…the nature and causes of the natural variability of climate and its interactions with forced changes” (1).
These statements articulate the current conception of the climatic phenomenon popularized as global warming. Save for a negligible number of exceptions, the contention surrounding global warming no longer involves a dispute over whether or not mean global temperature is increasing. Statistically, the continual elevation in mean global temperature has been measured at a rate of approximately one degree Celsius every ten years (2). Climatic debate now focuses upon the cause and progression of this unprecedented increase in global temperature, particularly deliberating between natural variability of the climate and abnormal climatic forcing. In this context, climate forcings are defined as imposed perturbations of the Earth’s energy balance (1), expressed in units of watts per square meter (Wm -2), and may be categorized as either naturally or anthropogenically derived. In an effort to delineate between and evaluate the effects of natural variability and climate forcings on global climate, the extent to which these climatic forces influence the variables that compose global climate, such as sea ice, must be quantified.
Arctic sea ice is among the primary indicators of global climate change because it is incredibly sensitive to alterations in Arctic boundary conditions (3), such as temperature and precipitation, and any change in the areal extent and thickness of the ice sheet can be precisely measured. Currently, the Arctic and sub-polar sea ice extent exhibit an annual mean of approximately 11.6 million km2, expanding to a maximum areal coverage of 14.4 million km2 in the winter, and melting to a minimum of 7.7 million km2 in the summer (4). Satellite and ground-based observations indicate that the areal extent and concentration of the Arctic ice cover have been significantly reduced over recent decades (5), decreasing by approximately 3 and 4 percent per decade (6), respectively. This is analogous to a reduction in the areal extent of the ice cover equivalent to the size of Texas every ten years, and a decrease from a mean ice thickness of 10.2 feet to a mean ice thickness of 5.9 feet over the past 40 years (7). Recent research has been conducted to assess the roles of natural climatic variability and abnormal climate forcings, in the drastic decrease in the volume of Arctic sea ice, in an effort to interpret these changes within the context of global climate.
In order to facilitate a discussion of Arctic climatology, the purpose and significance of climatic models must be identified. Climate system models are mathematical, computer-based expressions of the thermodynamics, fluid motions, chemical reactions, and radiative transfer of energy within Earth’s climate (1). These models mathematically relate climatic variables; therefore, they are capable of constructing climates that reflect the effects of differing values imported into the program to represent possible quantifications of the climatic variables. In this way, climatic models are employed to determine the hypothetical impact of the components of natural variability and abnormal climatic forcings upon a specific climatic system. A popular example of a climate system model relevant to this discussion relates global temperature and the atmospheric concentration of greenhouse gases. Numerous models created for this purpose exist, and their projections illustrate the limitations of climate system models. Conclusions extrapolated from the models indicate everything from the complete independence to the absolute dependence of global temperature upon the atmospheric concentration of greenhouse gases (8), suggesting that climatic models are only as accurate as scientific understanding of the interacting variables allows (1). Climate system models of the Arctic often include expressions of sea ice extent and thickness, atmospheric temperature, surface ocean temperature, precipitation, solar radiation, and wind patterns (8). Although these models suffer from similarly ambiguous conclusions involving natural variability and abnormal climate forcings alike, an indication of the importance of Arctic sea ice, and more specifically the significance of the dynamics and thermodynamics that govern Arctic sea ice within the context of global climate is ubiquitously confirmed (9). In order to increase the accuracy and precision of Arctic climate system models, evidence of natural variability and abnormal climatic forcings in the recent ablation of Arctic sea ice must be further examined.
Proponents of natural variability in this climatic controversy cite transient climate change and, more specifically, Arctic dynamics as the predominant causes of the decrease in the volume of Arctic sea ice. Transient climate change dictates that, just like the weather, the global climate fluctuates even in the absence of abnormal forcing (1), albeit on a far greater temporal scale. Paleoclimatic records composed of data from “tree rings, marine sediment cores, pollens, and continental ice cores have been analyzed to estimate the range of natural climate variability over periods…in excess of 400,000 years” (1). A close examination of the chemical components of these elements, particularly carbon and oxygen isotopes, “indicate that the range of natural climatic variability is in excess of several degrees [Celsius] on local and regional space scales over periods as short as a decade” (1). However, the accuracy of these forms of paleoclimatic records, with respect to current climatic trends, is limited by the magnitude of any measurement’s error on a geologic timescale, for which even a remarkably small error can represent a period longer than the duration of human existence on Earth. As a result, an expression of Arctic dynamics offers a more accurate representation of the role of natural variability in the change currently exhibited by the global climate.
Unlike terrestrial ice sheets, such as Greenland and Antarctica, the Arctic ice cover is composed of a multitude of highly mobile ice floes that interact under the principal kinetic forces of wind and ocean currents. Within the context of this discussion, only dynamic variables that have exhibited specific and significant change in correlation to the decrease in Arctic sea ice will be examined. The most important of these unstable variables concerns the pattern of winds that affect the Arctic, described by the North Atlantic Oscillation (NAO).
The NAO is considered a naturally occurring atmospheric climatic forcing that has always been a component of global climate. Historically, this wind pattern was identified in observations that a mild winter in Northern Europe occurred only when southern Greenland was experiencing a severe season, and vice versa (6). The NAO is now considered an expression of the atmospheric surface variability that affects the North Atlantic sector of the Arctic region, specifically between the longitudinal margins 90oW-40oE (10). A subset of the more geographically widespread Arctic Oscillation (AO), the NAO “refers to a redistribution of atmospheric mass between the Arctic and the subtropical Atlantic, and swings from one phase to another produce large changes in the mean wind speed and direction over the Atlantic” (10).
The NAO is quantified by spatially and temporally dependent measurements of surface air temperature (SAT) and sea level pressure (SLP) (11), which are expressed as eigenvectors on a geographical grid; located, these vectors essentially describe the strength and direction of the wind patterns that partially govern the Arctic climate (12), particularly linking a major low pressure area centered over Iceland and a high pressure area centered near the Azores. Meteorological records indicate that “when both pressure features are strong, the NAO is considered to be in the ‘positive’ mode, and when both pressure features are weak, the NAO is said to be in the ‘negative’ mode” (6). Large changes in the SLP, and the winds associated with this pressure, are generally associated with boreal winter, which lasts annually from December to February or March (10), and varies between the positive and negative modes on a decadal timescale (13). However, widespread warming over Alaska and Western Canada and cooling over Eurasia since 1970 suggests that the NAO has been “stuck” in the positive mode (6). Under a positive NAO, increased surface winds, at an average speed of 5ms-1 (10), blow counterclockwise around the Icelandic low (6). As a result, the NAO has been accused of exporting sea ice through Fram Strait from the Canadian basin before it has the opportunity to grow, thus contributing to the recent decrease in the volume of Arctic sea ice (12).
The correlation between this increased atmospheric forcing and the decrease in the areal extent and thickness of Arctic sea ice due to premature exportation into the North Atlantic has been demonstrated by an analysis of geographic and temporally variable data and relevant climatic models. The National Center for Atmospheric Research (NCAR) first described the anomalous spatial movement exhibited by Arctic sea ice over the period from 1958 to 1997 as a contoured field of vectors, then projected this pattern onto the empirical orthogonal function (EOF) associated with the SLP of the NAO. The two vector fields exhibited remarkably similar contours, and as a result, NCAR now contends that “the temporal and spatial relationships between the SLP and ice anomaly fields are consistent with the notion that atmospheric circulation anomalies force the sea ice variations” (11). Climatic models relating the simultaneous temporal evolution of both the NAO and the extent and thickness of the Arctic ice cover exhibit a high autocorrelation of 0.69 between the quasi-stochastic decadal-scale variations (4) of the atmospheric forcing associated with the NAO and the cyclical spatial patterns exhibited by the Arctic sea ice, further emphasizing the dynamic importance of the NAO to Arctic sea ice (11).
It may be interesting to note that research is now being conducted by NCAR to determine the role of abnormal climate forcings, particularly anthropogenic forcing, in the abnormal behavior of the NAO (3); however, conclusions of this endeavor have yet to be published.
Abnormal climate forcings, whether naturally or anthropogenically derived, are typically quantified in terms of the thermodynamics that govern the Arctic climate. Arctic thermodynamics do not respond to the aforementioned dynamic variables, but rather depend upon the energy balance that governs the Arctic climate, often quantified as the mass balance (IMB) of the Arctic sea ice. The IMB is simply an expression of the melt and growth of the surface and bottom of a specific areal amount of ice, in addition to any precipitation that falls on that location. Although a simple thermodynamic concept, the IMB is a highly sensitive climatic indicator because it integrates the surface heat budget and the ocean heat flux, which together form the energy balance that governs the Arctic climate. The IMB reflects Arctic climatic change in that a net warming results in thinner ice, while a net cooling produces thicker ice. Therefore, a thorough quantification of the temporal evolution of the IMB, and its associated feedback cycles, is an integral component of the evaluation of the role of thermodynamics in the elevated ablation of Arctic sea ice.
Conducted between October of 1997 and October of 1998, the Surface Heat Budget of the Arctic Ocean (SHEBA) experiment was perhaps one of the most extensive analyses of the Arctic IMB to date. In this experiment, a number of climatologically oriented institutions collaborated to freeze the Canadian ice-breaker Des Grolliers into the Arctic ice pack. Approximately located at 75oN, 143oW, the SHEBA camp was deployed in the center of the Beaufort gyre (14). In comparison to a dataset produced at a similar location in 1959, the data collected by SHEBA presented a considerable, uniformly distributed thinning of the Arctic ice cover. The surface melt in 1959 was reported to be 38 cm, compared to 56 cm at SHEBA. The increase in the bottom ablation of the ice pack was even more pronounced: 11 cm in 1959, compared to 62 cm during SHEBA (14). It must be noted that the IMB is a quasi-parabolic function that does not vary directly with the net ablation of sea ice. IMB also accounts for accumulated precipitation on the surface of the ice, which can protect the ice from surface ablation, therefore introducing a stochastic element into the relationship between heat and the resultant ablation of sea ice. In combination with measurements of simultaneously accumulated precipitation, the net decrease in the thickness of Arctic sea ice reported by SHEBA was interpreted as a net warming in the Arctic IMB (14). These results have since been confirmed by data collected by a pan-Arctic network of autonomous IMB buoys deployed by the U.S. Army Corps Cold Regions Research and Engineering Laboratory (CRREL), which is funded by the National Science Foundation (NSF) and the National Oceanographic and Atmospheric Administration (NOAA).
The predominant role of Arctic thermodynamics in the accelerated ablation of sea ice has been further demonstrated by a separate examination of paleoclimatic history, in addition to climatic models relating the atmospheric concentration of carbon dioxide (CO2) and the enhanced warming of polar regions. Speculations surrounding the ultimate seasonal disappearance of the Arctic sea ice cover vary from 5 to 10 decades (7). Again, the accuracy of a paleoclimatic history with respect to current climatic trends must be questioned; however, the American Geophysical Union reports that “there is no paleoclimatic evidence for a seasonally ice free Arctic during the last 800 millennia” (15), indicating a massive shift in the energy budget of the Arctic climate. This change in Arctic thermodynamics is physically echoed throughout the region: the 170 mi2 Ward Hunt Ice Shelf, located at the northern end of Ellesmere Island, broke apart for the first time in thousands of years during the summer of 2005 (16). Preliminary climatic models relating current isostatic climatic regimes with those of the late Cenozoic era indicate little correlation between current and paleoclimatic trends; however, these early models consider too narrow a geographic region to be truly cited as evidence in a discussion of global climate change (17).
Other climatic models developed to relate the atmospheric concentration of CO2 and anomalies in sea ice extent present more tangible conclusions. These models express the concentration of CO2 as equivalent values of thermal radiation, varying between -10 Wm-2 and +10 Wm-2 (3). A negative value of thermal radiation correlates to a climate in which no elevation exists in the atmospheric concentration of greenhouse gases; zero is an expression of the current climate; and a positive value of thermal radiation represents an increase in the atmospheric concentration of greenhouse gases. These models almost unanimously confirm that “the variability of ice extent and volume increases for a warming climate,” while the variability exhibited by ice export decreases under the same conditions (3). A correlation between thermal radiation and ice extent of this magnitude confirms the integral influence of thermodynamics on the areal extent and thickness of sea ice.
Regardless of the cause of the elevated ablation of Arctic sea ice, the decrease in the areal extent of the ice pack has activated a positive feedback cycle that is accelerating global climatic change. The ice-albedo feedback cycle depends essentially upon the ratio between the amount of solar radiation that is reflected by the ice surface and the amount of solar radiation that is absorbed by the Earth. Ice displays one of the highest indices of albedo, reflecting approximately 95 percent of solar radiation back into space. By contrast, oceanic water exhibits the lowest albedo index, reflecting only 70 percent of incident solar radiation. Therefore, as the mean global temperature increases and the areal extent of the Arctic ice pack decreases, the best albedo surface is replaced by the worst reflector. Thus, the planet continues to warm, accelerating the ablation of sea ice and establishing a climatic cycle for which the American Geophysical Union holds little hope of reversal: “There seem to be few, if any, processes or feedbacks within the Arctic system that are capable of altering the trajectory towards this ‘super interglacial’ state” (15).
The report of the CSCC to the Bush Administration concluded that “the impacts of these [climatic] changes will be critically dependent on the magnitude of the warming, and the rate at which it occurs” (1). Therefore, it is of paramount importance to quantify current climatic trends in order to accurately predict future climate, as an ice-free Arctic would produce significant biological, ecological, political, and economic global impacts. The most controversial of these effects involves the popular destruction of Arctic habitat (18) and the opening of the Northwest Passage. Historically elusive, the Northwest Passage connects Europe and Asia with a shipping lane that is 4,800 miles shorter than that which passes through the Suez Canal (7), enabling passage through the Northern Sea Route, for which there has been declared the possibility for “considerable profit potential” by involved overnments (7). The United States is attempting to establish a governing policy for the potential sea route, as an ice-free Arctic would pose a security threat to the unprotected northern Canadian border, and an un-patrolled realm of naval disaster and smuggling success alike (16). Little progress has been made in this pursuit, impeded by the uncertain variable of global climate change.
Unfortunately, interest in global climate change has evolved only with the capacity to measure the components of climate. Submarines and satellites are integral tools in this endeavor; therefore, datasets describing temporal and spatial variability of such climatic components as Arctic sea ice have existed intermittently only for the past five decades. This is not a timescale long enough to accurately reflect the progression of global climate change. However, the American Geophysical Union expresses the general consensus of the climatological community: “Evidence suggests we are witnessing the early stage of an anthropogenically induced global warming, superimposed on natural cycles, reinforced by a reduction in sea ice” (15). Current efforts to expand the information available to climatologists, employed to build models to project future climate trends, revolve about projects included in the International Polar Year 2007-2008.
Acknowledgments
The author would like to acknowledge the tremendous influence of Donald K. Perovich, of the U.S. Army Corps Cold Regions Research and Engineering Laboratory (CRREL: 72 Lyme Road, Hanover, NH 03755-1290, U.S.A) upon the subject of this research.
References:
1. “Climate Change Science: An Analysis of Some Key Questions” (Committee on the Science of Climate Change; Division on Earth and Life Studies; National Research Council, 2001).
2. International Panel on Climate Change, Climate Change 2001. Available at
http://www.grida.no/climate/ipcc_tar/wg1/349.htm (21 May 2006).
3. P. Lemke, M. Harder, M. Hilmer, Climatic Change 46, 277 (2000).
4. J. Wang, M. Ikeda, S. Zhang, R. Gerdes, Clim. Dynam. 24, 115 (2005).
5. H. Eicken, W. B. Tucker, D. K. Perovich, Ann. Glaciol. 33, 194 (2001).
6. M. Sturm, D. K. Perovich, M. C. Serreze, Sci. Am. 289, 60 (2003).
7. W. Gibbs, “Research predicts summer doom for northern ice cap,” New York Times (New York), 11 July 2000, p. F2.
8. K. W. Dixon, T. L. Delworth, T. R. Knutson, M. J. Spelman, R. J. Stouffer, Global Planet. Change 37, 81 (2003).
9. D. K. Perovich, B. Elder, Ann. Glaciol. 33, 207 (2001).
10. J. W. Hurrell, Y. Kushnir, G. Ottersen, M. Visbeck, An overview of the North Atlantic Oscillation. Geophys. Monogr. 134, 1 (2003).
11. C. Deser, J. E. Walsh, M. S. Timlin, J. Climate 13, 617 (2000).
12. M. Hilmer, T. Jung, Geophys. Res. Let. 27, 989 (2000).
13. T. M. Joyce, C. Deser, M. A. Spall, J. Clim. 13, 2550 (2000).
14. D. K. Perovich et al., J. Geophys. Res. Oceans 108, 26-1 (2003).
15. J. T. Overpeck et al., Eos. 86, 309 (2005).
16. N. Singer, Outside 105, 40 (2005).
17. F. A. Butt, H. Drange, A. Elverhoi, O. H. Ottera, A. Solheim, Quaternary Sci. Rev. 21, 1643 (2002).
18. T. Ajotts, Earth Isl. J. 6, 12 (1991).