up previous
Up: Interaction of the North Previous: Bibliography

Figures


  
Figure 1: (a): Regression map of SLP (contoured every mb, dashed for negative values) onto the time series of the first principal component (PC1) of SLP anomalies over the North Atlantic sector. The SLP data are from NCEP-NCAR from 1958 to 1998. (b): The NAO index of Hurrell [1995] defined as the difference in normalized pressure between Lisbon and Stykkisholmur, for the winter months, December-March (thin line). The thick curve represents the PC1 time series.


  
Figure 2: (a) Regression of the surface wind stress curl (every $2\; 10^{-8} Pa m^{-1}$, dashed for negative) onto the NAO index of Hurrell [1995]. The thick black line indicates the position of the climatological zero windstress curl line. (b) regression of surface turbulent heat flux (latent+ sensible, contoured every $10\;Wm^{-2}$, dashed for negative, i.e., out of the ocean) onto the PC1 in Fig. 1b, thick curve. (c) Same as Fig. 2b, but for SST anomaly (every $0.1 \;K$, dashed for negative). All surface data are from the NCEP-NCAR reanalysis.


  
Figure 3: Sverdrup circulation computed from the NCEP-NCAR reanalysis surface windstress (contoured every Sverdrup), and annual mean position of the zero wind-curl line (thick curves). The two boxes indicate the regions used to construct the dipole SST index $\Delta T \, = \, T_N - T_S$, where TN and TS are the temperatures in the northern and the southern box, respectively.


  
Figure 4: Schematic diagram of the anomaly in thermohaline circulation induced by the dipole in ocean thermal anomalies created by anomalies in air-sea heat fluxes associated with high NAO. The overturning circulation sketched in the meridional section on the right represents a zonal average picture.


  
Figure 5a: model

Figure 5b: observations

Panel a: Predicted $\Delta T$ power spectra for the [FH77] model (red curve), the passive gyre model (blue curve), the delayed oscillator model (green curve), and the combination of passive gyre and feedback (black curve). The frequency is in cycle per td with the assumption td= 10 yrs. The power has been scaled to observed spectra (panel b). Panel b: Observed $\Delta T$ power spectra for the difference TN-TSin late winter, and for the SLP anomalies in the GIL region.



  
Figure 6: Upper panel: Late winter (February - April) time series of $\Delta T$. Raw yearly data (blue curve) are superimposed onto 6yr running mean (green curve). Lower panels: Composite maps of SST anomalies (in K) based on years during which $ \vert \Delta T \vert \, > \, 0.7$K (denoted by stars and circles in upper panel). The 'high - low' map is obtained by subtracting the high from the low index composite. The 'N yr later' map is obtained by subtracting the high index + N yr composite from the low index + N yr composite. Black curves indicate the 95% confidence levels estimated from a Student's t-test, assuming 6 (warm independent $\Delta T$ years) + 6 (cold independent $\Delta T$ years) - 2 = 10 degrees of freedom.


  
Figure 7: Composite of large-scale SLP anomalies (contours, levels in mb) for 'high - low' $\Delta T$ yrs, correspondign to Fig. 6. Gray shadings indicate the COADS mean SLP in late winter.


up previous
Up: Interaction of the North Previous: Bibliography
Patrick Heimbach
2000-10-24