|
Article Excerpt
ABSTRACT. A tidal model of the Canadian Arctic Archipelago was used to map the strength of the tidal currents, tidal mixing (h/[U.sup.3]), and the vertical excursion associated with the tidal currents that drive water upslope and downslope. The hot spots in these quantities correspond to the location of many of the small polynyas in the archipelago, supporting the idea that the tidal currents make an important contribution to the dynamics of many of these recurring polynyas. The potential link with tidal mixing means that these locations may have enhanced plankton production in the summer.
Key words: Canadian Arctic Archipelago, polynyas, tidal currents, tidal mixing, tidal mixing fronts
RESUME. Un modele des marees de l'archipel Arctique canadien a servi a mapper la force des courants de maree, le melange de maree (h/[U.sup.3]) et l'excursion verticale associes aux courants de maree qui poussent l'eau en ascendant et en descendant. Les points chauds de ces quantites correspondent a l'emplacement d'un grand nombre des petites polynies de l'archipel, ce qui vient appuyer l'idee selon laquelle les courants de maree jouent un role important dans la dynamique d'un grand nombre de ces polynies recurrentes. Le lien susceptible d'exister avec le melange de maree implique que la production de plancton a ces emplacements pourrait etre rehaussee a l'ete.
Mots cles: archipel Arctique canadien, polynies, courants de maree, melange de maree, fronts de melange de maree
Traduit pour la revue Arctic par Nicole Giguere.
INTRODUCTION
A polynya is a geographically fixed region of open water (or low average sea-ice thickness) that is isolated within thicker pack ice. Polynyas are an important component of both the physical and the biological systems in ice-covered seas (Smith and Barber, 2007), and they are widely distributed across the Canadian Arctic Archipelago (Fig. 1). From the physical point of view, polynyas are areas of enhanced air-sea fluxes in winter relative to the neighbouring ice-covered regions (Smith et al., 1983, 1990). From the biological perspective, polynyas that reliably occur each year are thought to be of particular ecological significance, especially for marine mammals and seabirds (Stirling, 1980; Stirling and Cleator, 1981).
[FIGURE 1 OMITTED]
Polynyas can continuously lose heat to the atmosphere without accumulating as much ice as the surrounding areas by several means. The two traditional categories of polynyas are latent heat polynyas, in which wind and currents drive away consolidated ice; and sensible heat polynyas, in which the heat flux from warmer subsurface waters slows or eliminates the formation of ice (Smith et al., 1990). The name "latent heat" refers to the latent heat of fusion released as the water is transformed into ice, and the name "sensible heat" refers to the oceanic heat required to keep the surface temperature above freezing. Williams et al. (2007) introduced a similar two-category classification based on the mechanisms that remove or reduce the ice: mechanically forced (ice-divergence) and convectively forced (oceanic heat flux) polynyas. This classification is largely equivalent to the latent and sensible heat classification. Whichever classification is used, most polynyas are a mixture of the two categories. For example, the measured heat flux of 329 W [m.sup.-2] at the Dundas Island polynya ( Fig. 1, polynya No. 16) was shown to be 62% sensible heat and 38% latent heat (den Hartog et al., 1983).
Several secondary mechanisms can contribute to polynya maintenance: currents can sweep the newly formed (or frazil) ice out of the polynya and under the surrounding ice (den Hartog et al., 1983: Fig. 1); strong currents can increase heat conductance at the ice-water interface (Morse et al., 2006); and turbulence resulting from surface waves, strong currents, or both may inhibit the consolidation of frazil ice (Daly, 1994). Another important factor in polynya formation is shelter from drifting ice provided by coastlines, fast ice, or an ice bridge (Ingram et al., 2002; Williams et al., 2007). This shelter is important because no amount of latent or sensible heat flux can maintain a polynya against an influx of ice formed elsewhere.
The well-known North Water polynya in northern Baffin Bay (Fig. 1, polynya No. 21) is maintained primarily by latent heat or ice divergence (Ingram et al., 2002), as are the coastal leads that form when the winds and currents conspire to move the ice away from the coast. On the other hand, the Hell Gate polynya (Fig. 1, No. 17) and several others in the Canadian Arctic are thought to reflect an appreciable contribution of sensible heat (Topham et al., 1983; Smith et al., 1990) through the combination of a warm water reservoir at depth and strong tidal mixing. The importance of tidal mixing to ice dynamics is further supported by Saucier et al. (2004), who found that tidal mixing was crucial for the simulation of a stable seasonal cycle in an ice-ocean model of Hudson Bay.
The present study identifies areas in the Canadian Arctic Archipelago where tidal currents are likely to make important contributions to polynya formation and maintenance. A simple conceptual model of tidal contribution to the sensible heat mechanism is shown in Figure 2. The tidal contribution requires three distinct features: a nearby source of warm water, a mechanism for getting the water from depth into shallower water, and strong tidal mixing to get the heat nearer to the surface. The transfer of heat from depth to the surface does not imply the existence of an unstable water column. In the Canadian Arctic, the stratification is dominated by salinity rather than temperature, and warmer water at depth is common (Melting, 2002). For the three secondary mechanisms mentioned previously, the only tidal factor is the strength of the currents.
[FIGURE 2 OMITTED]
As metrics of the potential tidal current contribution to polynya dynamics, we create maps of the tidal currents (U), the tidal mixing parameter h/[U.sup.3] (Simpson and Hunter, 1974; where h is the water depth), and the vertical excursion associated with the tidal currents that drive water upslope and downslope over a tidal period. The currents and water depths were taken from a depth-averaged tidal model of the Canadian Arctic Archipelago (Hannah et al., 2008). The analysis does not address the existence of a source of warm water.
In order to focus on the potential contributions from tidal currents, this analysis ignores currents, mixing, and upwelling due to the general circulation and wind forcing. Thus latent heat polynyas such as the North Water and Cape Bathurst polynyas should not be identified in the analysis.
The maps have potential uses beyond the interpretation of physical processes. In mid-latitude systems, the tidal mixing parameter hl[U.sup.3] has proven to be a robust measure of the potential for the vertical mixing associated with strong tidal currents to overcome the stratifying influence of the summertime surface heat flux (Simpson and Hunter, 1974; Garrett et al., 1978). The tidal mixing front is the transition zone from a stratified water column to a well-mixed one, and it is often a region of enhanced biological productivity (Backus and Bourne, 1987; Horne et al., 1989), especially in the summer.
METHODS
Tidal Model Overview
The implementation and validation of the tidal prediction system for the Canadian Arctic Archipelago is detailed in Hannah et al. (2008) and Dunphy et al. (2005). The model domain covers most of the region shown in Figure 1, including Baffin Bay and Davis Strait. The exception is that there is an open boundary in Fury and Hecla Strait (polynya No. 8), and the shelf regions to the south of the strait, such as Foxe Basin, Hudson Bay, and Hudson Strait, are not inside the model domain. The tidal model is MOG-2D (Carrere and Lyard, 2003), a two-dimensional finite element formulation with variable resolution. The horizontal resolution ranges from about 2 km in some coastal areas to 45 km in the open ocean.
The tidal elevation at the open boundaries was specified for the five major contributors to tidal variation (known as tidal constituents [M.sub.2], [N.sub.2], [S.sub.2], [K.sub.1], [O.sub.1]). These constituents were estimated using the inverse modeling system that is part of the prediction system (details later in this section). The observed tidal constituents at 54 coastal locations were used in the assimilation, and the system was validated using 47 additional locations. The root-mean-square (rms) elevation error, averaged over all...
|
