La réduction de la couverture de glace va donc faire varier drastiquement la « couleur » de la mer. Les modèles actuels parviennent à reproduire raisonnablement bien la circulation des glaces de mer à grande échelle (Figure 9).
Evolution récente de la vitesse de dérive et du taux de déformation de la
Introduction
This leads to fracturing and fractures in the sea ice sheet that accommodate most of its deformation [Schulson, 2004; Weiss et al., 2007]. Second, we examine the temporal evolution of sea ice deformation rates during the same period (section 5).
Dataset
The contribution from the Fram Strait (noted as "Fram" on the chart in Figure I-1), which is Figure I-2. These velocities are, on average, less than in the central Arctic (the "Central Arctic" zone on the map in Figure I-1).
Analysis of the time variation of buoy speed
- Monthly averages of buoy speed
- Seasonal averages of buoy speed
Increase in the average rate of deformation of the sea ice cover over the past 29 years. Annex: Spatial and time-scaling laws induced by the multiscale fracturing of the Arctic sea ice cover.
Are the mean IABP buoy speeds représentative of the mean Arctic sea ice speeds ?
- Constructing the mean speed fields
- Monthly mean speeds estimated from the interpolated speed fields
Increase of the mean déformation rate of the sea ice cover over the last 29 years
- Relation between dispersion of buoys and sea ice déformation
- Results
Discussion
- External forcing
- Sea ice déformation, thinning and export
- Year 2007
We found that both Arctic sea ice rates and deformation rates increased significantly over the past 29 years. This decrease most likely contributes to a decrease in sea ice mechanical strength and then favors an increase in sea ice deformation rates and associated fracturing during the period. Here we consider the mean speed of the sea ice along the coast of Greenland (the "Forward" region on Figure I-1).
However, we can interpret this positive trend as (i) a response to the change in the kinematics of the Arctic basin presented above, which was later favored by a thinning of the sea ice cover in. Central Arctic", and/or (ii) as a direct effect of thinner ice on sea ice kinematics in the "Fram". Consequently, sea ice kinematics and breaking may strengthen the albedo feedback loop, the polar amplification and the accompanying decline in Arctic sea ice cover.
Conclusion
Time correlations (memory effects) are present in ice velocity records [ Thorndike , 1986 ]. Rampal et al., in prep., 2008), and can also be found in time series of ice deformation rates as approximated by the distribution of pairs of buoys. An autocorrelation analysis shows a correlation time of about 10 h for an initial separation of L=300 km [ Rampal et al ., 2008 ]. Instead, we estimate these errors from the same bootstrap method as in section 3.1, with the difference that the number of pairs of buoys Np considered in each distribution, rather than the number of samples N, should be used as the number of independent variables .
We note that this expression is similar to the error estimate given by the central limit theorem, with the important difference that the number of pairs of buoys Np is considered instead of the number of samples N.
Champ de vitesse de la banquise Arctique
Dataset: Lagrangian trajectories of buoys
The drifting buoy dataset is provided by the International Arctic Buoy Program (IABP) and is available on the Internet (http://www.iabp.apl.washington.edu). The second criterion was based on the distance of the trajectories from the nearest coast: the positions recorded at a minimum distance of 100 km from the coasts were selected. From the selected trajectories, we separated the positions recorded in the winter (i.e. from the beginning of November to the middle of May of the following year) from those recorded in the summer (from the middle of June to the middle of September ).
Errors at these positions range from 100 m to 300 m depending on the type of positioning system embarked on the buoy [Thomas, 1999]. To obtain a more regular sampling, a cubic interpolation of the raw positions was first performed, before being resampled at a time interval of 3 hours. The reference coordinate system used in this study is a Cartesian coordinate system centered at the North Pole, with the vertical axis (i.e., y-axis) following the Greenwich meridian.
Estimating the mean velocity field u(x,t) of the Arctic sea ice
- Classical approach: the turbulent diffusion theory (Taylor, 1921)
- Definition of the mean velocity u
- Methodology
- Results and conclusions
- Discussion
Here, the mean value u L,T, and thus its associated fluctuation u', depends on the choice of the averaging scales L and T. We then investigate the influence of the choice of the averaging scales on the integral time. This is characteristic of the predictable nature, in the deterministic sense, of the mean circulation.
However, by using such averaging scales, one would lose information about the details of the mean circulation. This is the confirmation of what we deduced from the analysis of the autocorrelation functions. Near zero values are observed in the center of the Beaufort Gyre for both winters.
Analyzing the fluctuating velocity field of Arctic sea ice
- Spectral analysis
- Distributions of fluctuating velocity
- Intermittency and scaling
- Diffusion regimes of sea ice
- Discussion
Power spectrum of speed fluctuations for winter (in black) and summer (in gray) on a log-log scale. Unlike the other signatures discussed in this section, this inertial motion should not be included in the random part of the sea ice motion. From the k!2 scaling of the power spectrum of the fluctuating velocity (Figure II-7), we expect the second-order structure function to scale as.
This later shows a curvature showing that the discontinuous character of the sea ice velocity can be reproduced by a multifractal model. The regime where the displacement variance grows with t was expected, since we showed in subsection 4.2 that the distribution of the fluctuating sea ice velocity is exponential. We also calculated the turbulent diffusivity of Arctic sea ice, which follows (II-10) for t>>!.
Conclusion
Champ de déformation de la banquise Arctique
Methodology
We study how the scattering of bending pairs depends on both (i) their initial separation L and (ii) the time. In fluid mechanics, the dispersion process is characterized by the mean square change in separation !r2. The choice of the standard deviation of D rather than its mean value is partly motivated by the fact that in the limit of small deformation rate (i.e. for large temporal and spatial scales), only the standard deviation allows to characterize the deformation process.
We checked this by calculating the two parameters for each calculated distribution of D (i.e. corresponding to each L-! interval). Indeed, autocorrelation analysis of time records D for an initial separation of L ~ 300 km shows a correlation time of approximately 10 hours. Furthermore, the distributions of D are clearly not Gaussian, with significantly slower decays towards large values (this aspect will be detailed in future work).
Introduction to the dispersion process: two examples of buoy dispersions
Introduction to the Dispersion Process: Two Examples of Buoy Dispersion. regions and battle starting positions. Its shape evolves over time and illustrates the deformation of the sea ice of an individual region through a combination of shear and divergence components. These simulations do not take into account the existing spatial correlations of combat trajectories.
Standard deviation !D of strain D (calculated from the dispersion of boj pairs) and strain !tot with respect to time t for the SHEBA experiment. Consequently, the value of 2.25 obtained for SHEBA is not necessarily representative of the general case. Finally, we calculate the standard deviation !D strain rate D = !rL", as well as the strain rate.
Sea ice dispersion and deformation over the whole Arctic Ocean: Results
- Time scaling
- Spatial scaling
- Time scaling
- Spatial scaling
- Seasonal dynamics of sea ice
- Perspectives for future studies
Arctic sea ice deformation appears to follow different regimes, from sub-diffusive (at L!10 km) to super-diffusive (at L! km). Based on the scaling properties summarized in Figures III-8 & III-10, winter and summer sea ice dynamics cannot be differentiated. Wacker (2007), Influence of lead in sea ice on polar nighttime atmospheric boundary layer temperature, Geophys.
Bentley (2001), Evidence for rapid thinning of sea ice in the western Arctic Ocean in the late 1980s, Geophys. Rampal (2008b), Spatial and temporal scaling laws induced by multiscale fracturing of the Arctic sea ice, in Scaling in solid mechanics, edited by F. Rampal (2008) Spatial and temporal scaling laws induced by multiscale fracturing of the Arctic sea ice sheet, in Scaling in Solids Mechanics, P.
Conclusions et perspectives
Conclusions
Cela peut également expliquer l’augmentation de la vitesse moyenne de dérive de la glace de mer mentionnée ci-dessus, qui, en raison de fractures croissantes, est de plus en plus fragmentée et sensible aux forçages atmosphériques et océaniques. iii). Le champ de vitesse de la glace de mer arctique peut être décomposé en un champ moyen et un champ fluctuant. Cependant, les distributions de vitesses fluctuantes sont exponentielles et non gaussiennes comme c'est le cas par exemple dans l'océan ou dans l'atmosphère. v) Plus généralement, le champ de vitesse des glaces de mer présente des propriétés intermittentes, exprimées par une invariance à l'échelle multifractale des augmentations de vitesse. vi).
Cette discontinuité n’est pas une trace directe de turbulence océanique et ne peut s’expliquer par la seule nature du forçage océanique et atmosphérique. La réponse mécanique de la banquise à ce forçage, et notamment son comportement élasto-fragile, joue probablement un rôle dans cette discontinuité. vii) Le champ de déformation de la glace de mer arctique peut être caractérisé à l'aide des propriétés de dispersion des bouées intégrées dans la calotte glaciaire. Le champ de déformation des glaces de mer arctiques se caractérise par une forte discontinuité et une forte hétérogénéité, y compris à grande échelle (plusieurs mois et plusieurs centaines de km).
Discussion et perspectives
- Intensification de la boucle de rétroaction de l’albédo
- Prise en compte du processus de fracturation dans les modèles
2005), Effects of second-year ice type variability on multiyear ice cover decline, Ann. 2006), Sudden decline in Arctic winter sea ice, Geophys. Teng (2008), Evolution of Arctic sea ice concentration trends and the role of atmospheric circulation forcing, Geophys. Here we consider that nearly all sea ice deformation is accommodated by multiple fracturing and faulting at various scales, i.e.
This is another way of expressing the heterogeneous character of sea ice deformation and breaking up to very large time scales. Instead, these observations suggest an elasto-brittle behavior of the sea ice sheet, with deformation essentially accommodated by transient fracturing events at various scales. The complex relationship between space and time that characterizes sea ice dynamics is now illustrated by the dependence of.
