Role of Sea Ice Thickness for the Export

In the previous sections, the strong influence of the atmospheric circulation on the interannual variability of Fram Strait ice export has been demonstrated. However, studies by Koeberle and Gerdes (2003) and Arfeuille et al. (2000) pointed out that sea ice thickness anomalies have a considerable impact on the export through Fram Strait as well. In both studies, the ice export anomalies have been divided into a part, related to ice thickness anomalies and a part related to ice velocity anomalies. The results indicate an almost equal importance of ice thickness anomalies for the entire export anomalies. Our model results show an increased ice thickness in Fram Strait if the cross-strait SLP-gradient is large. Assuming the same SLP-gradient, sea ice velocity is slightly smaller in Fram Strait with thick ice than with anomalously thin ice. Both, ice velocity and thickness are to large extent driven by the wind. Hence, the results of Koeberle and Gerdes (2003) and Arfeuille et al. (2000) need not to contradict to the high correlation of SLP-gradient and ice export.

To further analyze the relation between ice thickness anomalies in the Arctic and Fram Strait sea ice export, we performed a lag regression analysis (Fig. 8.8). Five a) lag -5 years b) lag years c) lag -1 year a) lag -5 years b) lag years c) lag -1 year

d) lag 0 year e) lag 2 years f) lag 4 years

■0.5 -0.15 -0.08 -0.06 -0.04 -0.02 -0,01 0.01 0,02 0.04 0,06 0.08 0.16 0.6

Fig. 8.8 Regression coefficient between annual mean ice exports through Fram Strait and ice thickness anomalies in cm per standard deviation ice export. (a) ice export lags 5 years, (b) ice export lags 3 years, (c) ice export lags 1 year, (d) lag 0, (e) ice export leads 2 years, (f) ice export leads 4 years (Based on Koenigk et al. 2006)

■0.5 -0.15 -0.08 -0.06 -0.04 -0.02 -0,01 0.01 0,02 0.04 0,06 0.08 0.16 0.6

Fig. 8.8 Regression coefficient between annual mean ice exports through Fram Strait and ice thickness anomalies in cm per standard deviation ice export. (a) ice export lags 5 years, (b) ice export lags 3 years, (c) ice export lags 1 year, (d) lag 0, (e) ice export leads 2 years, (f) ice export leads 4 years (Based on Koenigk et al. 2006)

years before high ice exports, positive ice thickness anomalies are formed at the coasts of Chukchi and East Siberian Sea (Fig. 8.8a). In agreement with results by Tremblay and Mysak (1998) and Haak et al. (2003), these anomalies are caused by a convergent ice transport due to an anomalous wind field and are associated with a negative ice export through Fram Strait. In the next 2 years, the positive ice thickness anomaly slowly propagates clockwise along the Siberian coast (Fig. 8.8b) and crosses the Arctic to reach Fram Strait leading the ice export by 1 year (Fig. 8.8c). High ice exports themselves are associated with large anomalous ice transports all across the Arctic towards Fram Strait (Fig. 8.3b) caused by the anomalous atmospheric forcing described above (Fig. 8.3a). A negative ice thickness anomaly occurs at the Siberian coast as a consequence of the divergence in ice transports. It propagates across the Arctic to Fram Strait in the next years, which leads to a decreased ice export (Fig. 8.8e and f) 4 years later. One further year later, the ice export is still reduced and ice thickness at the Siberian coast is again increased. The entire cycle takes about 9 years and matches the peak in the power spectrum of the ice export at the same time scale (Fig. 8.2). A detailed description of this process is given in Koenigk et al. (2006).

This mode has the potential for predictability of the ice export through Fram Strait. Apparently, large ice exports are characterized by previous ice volume anomalies at the Siberian coast and vice versa. Statistical analyses show the largest predictability for the ice export through Fram Strait if ice thickness is increased 2 years before in the Laptev Sea. Figure 8.9 displays the probability distribution of the annual mean ice export 2 years after 69 years with positive and 71 years with negative ice volume anomalies (exceeding the mean ± 1 standard deviation) in the Laptev Sea. After positive anomalies, a considerable shift in the mean ice export towards positive values can be seen and vice versa. The skewness of the distribution is negative after thick ice and positive after previously thin ice in the Laptev Sea. The probability for negative ice export events through Fram Strait is highly

0 3 _ — ■ after pos. ice volume anomaly (>1 stdv) ■ ■ ■ after neg. ice volume anomaly (<-1 stdv)

Was this article helpful?

0 0

Post a comment