U.S. Dept. of Commerce / NOAA / OAR / PMEL / Publications

Copyright ©1991 by the American Geophysical Union. Further electronic distribution is not allowed.

The vertical turbulent entrainment across the base of the mixed layer can deepen
the mixed layer and contributes a term *Q*
to equation (1):

(8)

where *T* is the temperature jump
across the mixed layer base. Estimation of *Q*
requires a measure of *W*
and *T*. Following *McPhaden*
[1982], *W* = *W*
> 0 where

(9*a*)

The time rate of change of mixed layer depth is *h*/*t*
and *W* is the vertical
velocity in the thermocline just below the base of the mixed layer. Requiring
*W* > 0 is necessary to satisfy the condition that entrainment can only
cool the mixed layer. Assuming negligible heat flux across isotherms and ignoring
horizontal advective terms in the thermocline, the entrainment velocity can
be approximated by:

(9*b*)

The depth of the 20°C isotherm (*h*)
is used to represent thermocline motions, since this isotherm is usually just
below the mixed layer in the low-pass-filtered time series (Figure
5).

The appropriate temperature difference *T*
at the base of the mixed layer is difficult to specify a priori. However, we
estimated this temperature by computing the linear regression between -*C**W*
and *Q*. The correlation
coefficient was 0.64, and the regression coefficient (1.7°C) was used for *T*
in equation (8). The mean temperature gradient
below the mixed layer was about 0.1°C m,
so this average value of *T* represents
upwelling from about 20 m deeper than *h*.

Our estimates of *W* and *T*
were used to compute *Q*
(Figure 6*c*). This time series is well
correlated with the mixed layer heating (correlation coefficient of 0.64 as
noted above) and has a standard deviation of 22 W m.
Both positive and negative values of *Q*
are included in Figure 6*c* and the correlation,
to take into account the fact that the local winds are almost always upwelling
favorable. Thus one can consider the fluctuations of *Q*
to be superimposed on a mean cooling, with positive values of *Q*
representing a reduction in this cooling.

Variability in *Q* was not
significantly correlated with the local zonal wind speed. The largest fluctuations
in *Q* were between early
July and mid-September 1986 and between November 1986 and January 1987. In both
cases the change in *Q* was
almost 85 W m and was in phase with the
fluctuations of mixed layer heating. Local wind variations during this period
were small.

Go back to previous section or forward to next section