Ice velocities drafts and fluxes

The ADCPs providing ocean velocities, also provide ice drifts at bi-hourly intervals. To show the total 8 year time series, the bi-hourly ice velocity data were again divided into 2-month sections for which the along-strait ice velocity was used to derive the bi-monthly vector mean velocities and the standard deviations (Std). Since ice is not always present in the summer months, the bi-monthly percent of ice velocity values was also calculated and shown in Figs. 7 and 8 for the southern and northern sites respectively.

E 100


>0 H—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—I—

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007


Fig. 7. Eight years of ice velocity data observed in Lancaster Sound located at a site 5 nm from the southern shore (Fig. 2). Bi-monthly means are shown as solid circles, with ±1 standard deviation illustrated with the bars. Percentage of ice presence providing ice velocities are shown by open squares.

Figure 7 shows the results from the southern site of Lancaster Sound where most of the eastwards flowing Arctic surface waters occur (Prinsenberg and Hamilton 2005). Land-fast ice conditions occur normally in March and April and in some severe winters during the months of May and June. In 2003, the ice was mobile for at least part of each 2 month period and land-fast ice conditions did not occur for the total 2 month period (March-April). During mobile ice conditions, the bi-monthly mean velocities are up to 50 cm/s but are generally around 10

cm/s. In these periods, the percentage of ice present is low, allowing the ice to move freely under ocean and wind forcing. Ice velocities towards Baffin Bay are generally larger than those towards the Arctic, as they move along with the eastward bi-monthly mean ocean currents.



Fig. 8. Same as Fig. 7 but for site 3 nm from the northern shore in Lancaster Sound (Fig. 7).

At the northern site, land-fast ice conditions persisted for the months of March and April for all years except 2003 and 2004 (Fig. 8). Bi-monthly vector mean ice velocities are smaller along the northern shore, and unlike at the southern site, have no preferred direction. In summer months the percentage of ice present drops below 25%, as was the case for the southern site.

Figure 9 shows 2 years of ice draft data from the ice season of 2003-04 when the pack ice remained mobile throughout the winter, and from the ice season 2005-06 when land-fast ice conditions occurred and thus reflect a more normal ice season. With global warming however, mobile ice conditions such as those of 2003-04 may become more prevalent. During mobile ice conditions, ice ridges passing the mooring site reach up to 24 m, but were generally up to 16 m for the 2003-04 winter. In 2005-06 the maximum ice draft was 13 m. From ice charts, it can be seen that the ice arch separating land-fast ice from mobile ice occurred in western Barrow Strait in 2003-04 (95° W), 120 km west of the mooring site, while it established itself at the mooring site in the 2005-06 winter (Can. Ice Service, Once land-fast ice conditions were established at the mooring site in February 2006, the sonar monitored the same ice that grew slowly thermodynamically. For the 2005-06 winter, the land-fast ice above the sonar was not ridged and thus low monthly maxima were detected. The small variability detected in the monthly mean is probably due to mooring motion that causes the sonar to monitor ice in a small area and not just one specific location of the pack ice.

Lancaster Sound, Monthly Maximum Ice Draft

Lancaster Sound, Monthly Maximum Ice Draft

Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Lancaster Sound, Monthly Mean Ice Draft

Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Lancaster Sound, Monthly Mean Ice Draft

Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul

Fig. 9. Two years of ice draft data from the southern site of the Lancaster Sound array. Shown are monthly maxima ice draft (top panel) and monthly means and standard deviations in the bottom panel.

Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul

Fig. 9. Two years of ice draft data from the southern site of the Lancaster Sound array. Shown are monthly maxima ice draft (top panel) and monthly means and standard deviations in the bottom panel.

In the bottom panel of Fig. 9, the monthly mean draft and the standard deviation are shown for the 2003-04 and 2005-06 winters. Relative to the 2003-04 winter, the 2005-06 winter started and ended 2 months later. During mobile ice conditions the monthly Std of ice draft and the monthly mean ice draft vary similarly; meaning that the variability about the mean and the mean itself increase and decrease proportionally. During land-fast ice conditions the Std approaches zero and the mean values vary as expected under thermodynamic ice growth. Once the land-fast ice breaks up in June 2006, ridges start to appear and increase the monthly maximum.

To estimate the freshwater flux associated with the mobile pack ice requires ice drift and ice thickness data such as shown in Figs. 6-8. However, not only are these time series scarce, but also very site-specific, as ice properties are less diffusive than ocean properties.

Although satellite imagery alone cannot provide estimates of ice-volume fluxes, it has provided long time series of ice-area fluxes in the CAA. However, the flux estimates are often hampered by coarse spatial resolution. Kwok (2006) estimated the ice-area transport across the main entrances to the Canadian Archipelago using Radarsat imagery (0.2 km resolution) and found that due to the land-fast ice within the CAA, the western entrances export ice to the Arctic Ocean, and the eastern entrances export ice to Baffin Bay. On average, Nares Strait does export ice from the Arctic to Baffin Bay, but export is prevented when the ice bridge in Smith Sound is present. Agnew et al. (2006) using Advanced Microwave Scanning Radiometer AMSR-E data (6 km resolution) found that transport directions were generally similar. However transport estimates differed in magnitude, perhaps because of different resolutions of the imagery or different time periods. In deriving ice-volume fluxes from ice-area fluxes for freshwater budgets, the uncertainty is further increased by the difficulty of estimating ice thickness. Freshwater fluxes in the form of ice that have been estimated from ADCP ice drift data (Prinsenberg and Hamilton 2005; Melling et al. 2008) are in the range of 1.5-2.5 mSv, and are an order of magnitude smaller than those in the water column.

0 0

Post a comment