Budyko (1969, p. 25-24) suggests the "possible existence of two climatic regimes in high latitudes, connected with the presence and absence of polar ice". Both regimes "are unstable, so that the ice cover can appear and disappear as the result of small changes in climate-shaping factors and even in the absence of these changes as a result of self-oscillation processes in the atmosphere-ocean-polar ice system". This proposal was confirmed by Rakipova (1962) and in other authors' studies (e.g., Saltzman et al., 1981). However, Doronin (1968) and Zakharov (1976, 1977, 1978, 1981, 1996, 1997) and Zakharov and Malinin (2000) called attention to the relationship of ice cover extent and the impact of the underlying fresh surface water layer, beneath which there is a halocline layer in which the water density rapidly increases with depth. On the one hand, this layer constrains the heat content of the water in summer, but on the other hand, it restricts heat flux to the surface from deeper layers where there are currents carrying heat from lower latitudes as a result of winter convection and vertical turbulent exchange. These processes contribute to increased ice thickness and area in winter, resulting in a decreased probability that the ice will melt the next summer. Zakharov (1976, 1977, 1978, 1981, 1996, 1997 and Zakharov and Malinin (2000) provide convincing arguments in favor of the important role of the halocline acting as a shielding layer restricting heat flux to the surface. The absence of this layer limits the ice cover extent much of the year (January-May). This is confirmed by numerous observations of a sharp decrease in the rate of sea-ice extent increase in the North European Basin in the middle of winter—long before the time when heat loss from the surface begins to decrease. All this led Zakharov to the fundamental conclusion that "the most significant cause of climatic changes in sea ice extent in the ocean is changes in the vertical water structure in the upper ocean layer, rather than changes in thermal conditions in the atmosphere'' (Zakharov 1996, p.183). Cases are noted (Malmberg, 1969) when the appearance of ice near the northern shores of Iceland was preceded by a significant decrease in salinity and temperature in the upper water layer.
The surface Arctic water mass forms as a result of mixing of freshwater (excess of precipitation over evaporation and continental runoff) with oceanic water flowing from the Atlantic and Pacific Oceans. Zakharov (1996) considers fluctuations in freshened surface Arctic water to result from disturbance of the freshwater balance of the Arctic Ocean. The incoming component of this balance is continental runoff, inflow of decreased-salinity water through the Bering Strait and precipitation while the discharge component is composed of runoff of freshwater to the Atlantic Ocean and evaporation (Serreze et al., 2006, Ivanov, 1976). Obviously, the freshwater balance should influence the volume and extent of surface Arctic low salinity water, but other factors contributing to this process are also important. These include the volume and salinity of water entering the Arctic Ocean, predominantly relatively saline Atlantic water. Studies of average salinity changes in the upper layer of the Kara Sea using a balance model (Appel and Gudkovich, 1984) showed the role of possible anomalies of Barents Sea water inflow to the Kara Sea in these changes to be comparable with the influence of annual river runoff anomalies. Continental discharge to the Kara Sea comprises more than 25% of the river runoff to the Arctic Ocean (Ivanov, 1980). Hence, to solve the problem of the origin of anomalies in surface Arctic water, it is necessary to consider not the freshwater balance anomalies, but rather the corresponding salt balance anomalies. A satisfactory relation between the continental runoff volume to the Arctic Ocean from the coast of Asia and North America (World Water Balance, 1974) and subsequent sea ice extent of the North European Basin given in Zakharov (1996) does not provide a convincing argument in favor of a decisive role for continental runoff, because, as Zakharov points out, the data he uses were derived from calculations done by indirect methods. The anomalies of iceberg discharge were not taken into account in the calculations, and the series compared are short (25 years). The continental runoff in this calculation comprises only 42% of incoming freshwater. In addition, its influence on Arctic water extent is strongly complicated by changes in the Beaufort anticyclonic gyre system (Volkov and Gudkovich, 1967; Alekseev et al., 2000; Nikiforov, 2006). A similar correlation of sea-ice extent with observed data on runoff from the largest rivers to the Arctic Basin seas, which supplies freshened Arctic water to the North European Basin, does not reliably confirm the relationship of sea-ice extent to continental runoff, as noted in Zakharov (1981).
It is important, however, to stress that justification of the role of the halocline in forming the sea ice cover of the Arctic Ocean excludes the possibility that a small increase in the incoming part of the Arctic heat balance can lead to relatively rapid disappearance of Arctic ice.
Was this article helpful?