Analyzing The Interaction Between Mean Monthly SSTs and Winds off the Coast of West Africa

1
Analyzing The Interaction Between Mean Monthly
SSTs and Winds off the Coast of West Africa

• Holly A. Anderson
• Linden Wolf
• Physics of Air-Sea Interaction
• Dr. Mark Bourassa
• December 6, 2007

2
Objectives

• Investigate the impact the curl of the wind
  stress has on SST in terms of upwelling and
  downwelling through Ekman pumping.
• Conduct a qualitative, climatological look into
  Ekman pumping and the related changes in SST in
  the sub-equatorial region off west Africa.

3
Relevance

• Upwelling of cooler, nutrient-rich oceanic water
  leads to increased biomass in the euphotic zone.
• Since increased biomass means increased living
  organisms, upwelling is a vital process for the
  success of the fishing industry.
• Upwelling and downwelling also has significant
  effects on the oceanic energy budget and oceanic
  circulation through its transportive effects.

4
Wind-Induced Vertical Motion

• The Coriolis force deflects wind to the left in
  the Southern Hemisphere.
• Therefore, cyclonic (anticyclonic) motion will
  result in mass divergence (convergence), and
  upwelling (downwelling) will occur. We would
  expect cooling (warming) of SSTs to occur in this
  situation.

Upwelling and therefore cooling of SSTs would
occur.
5
Ekman Pumping

• Vertical velocity in the Ekman layer is
  proportional to the curl of the wind stress.
• If Ekman transport causes divergence, there must
  be upwelling to replace the negative outflow.
• In the Southern Hemisphere, negative (positive)
  curl will correspond to positive (negative)
  vertical velocities, or upwelling (downwelling)
  of cooler (warmer) SSTs.

6
Data

• QuikScat/SeaWinds Scatterometer Psuedostress
  Fields
• Global
• 0.5 x 0.5 degree grid
• Monthly
• Converted into equivalent neutral wind speed
• 10m reference height
• QuikSCAT/NCEP Blended Curl of the Wind Stress
• Global
• 0.5 x 0.5 degree grid
• 6 hour time steps
• 10m reference height
• NCDC Extended Reconstructed Sea Surface
  Temperatures
• Global
• 2 x 2 degree grid
• Monthly

7
Methodology

• QuikScat/SeaWinds Psuedostress Fields
• QSCAT/NCEP Blended Curl of the Wind Stress
• NCDC Extended Reconstructed SST
• These three fields were averaged over August
  1999-June 2006 to provide a pseudo-climatology
  of winds, stresses, and SST over an area west of
  the continent of Africa for the 7 years that all
  three sources of data were available.
• The mean monthly fields were then plotted on a
  domain from 10 N and 40 S and from 25 W and 35 E.
  This area is off the west coast of central and
  south Africa.
• For this project, it is assumed that there is no
  surface current. That is, the change in SST is
  related to the change in the curl of the wind
  stress alone.
• Density is also assumed to be constant through
  seasons, as it relates to the Ekman equation.

8
(No Transcript)
9
Equatorial Cold Tongue

• The equatorial cold tongue (ECT) arises in the
  Southern Hemisphere winter when a zonal westward
  wind over the equator leads to mass vertical
  transport of water through upwelling.
• Winds north of the equator are deflected
  northward and winds south of the equator are
  deflected southward. Upwelling of cooler water is
  needed to replace the resulting outflow.
• The ECT obtains maximum amplitude during the
  Southern Hemisphere winter and surrounding months
  June, July, August, and September.

10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
Area 1 10W, 2S to 8W, 2N
Area 2 4W, 27S to 0, 23S
19
Area 1 10W, 2S to 8W, 2N
20

• SSTs and the curl of the wind stress both have a
  cyclical trend.
• The SSTs closely follow that of the curl of the
  wind stress.
• This is exactly as expected. In Area 1, we see
  that negative values of the curl of the wind
  stress are related to cooling SSTs (upwelling).
  Likewise, positive values of the curl of the wind
  stress are related to warming SSTs (downwelling).

21
(No Transcript)
22
(No Transcript)
23
SST Change Prediction in Area 1

• Changes in the magnitude of the curl of the wind
  stress do appear to be related to the magnitude
  change in SSTs, although not perfectly.
• The low correlation may be due to sampling
  errors, as well as density differences during
  different parts of the year.

24
Area 2 4W, 27S to 0, 23S
25

• In Area 2, we again see cyclical trends in SSTs.
  Unfortunately, we see little correlation between
  positive and negative values of the curl of the
  wind stress related to warming (downwelling) and
  cooling (upwelling) of SSTs, respectively.

26
(No Transcript)
27
(No Transcript)
28
SST Change Prediction in Area 2

• Whereas Area 1 was relatively well correlated,
  Area 2 is disappointing. The magnitude change in
  the curl of the wind stress does not appear to be
  well correlated to the magnitude change in SST
  whatsoever.
• What could affect this?
• This could be due to poor sampling of
  mid-latitude cyclones in the area, or due to eddy
  circulations in the area that spin off of South
  Africa.
• Also, it could be due to oceanic currents.

29
Ocean Currents off South Africa

• Area 2 is right above the Benguela Current,
  known for upwelling cold water.

• http//www-das.uwyo.edu/geerts/cwx/notes/chap11/s
  africa.html

30
Conclusions

• Calculating a monthly mean psuedo-climatology
  for the oceanic region west of Africa shows the
  presence of an equatorial cold tongue during
  winter months, as well as the monthly influence
  of Ekman pumping due to changes in the curl of
  the wind stress on SSTs.
• The prediction of SST change for more northern
  latitudes in this domain is decent.
• The relation between the magnitude change of both
  curl of the wind stress and SST is less obvious
  for latitudes farther south. This may be due to
  underlying oceanic currents in the South Atlantic.