The 2009/10 Warming Of The South Atlantic

I’ve moved to WordPress.  This post can now be found at The 2009/10 Warming Of The South Atlantic

##################
The Sea Surface Temperature (SST) anomalies for the South Atlantic are shown in Figure 1. First observation: it’s a noisy dataset. Then, other than the dip and rebound in 1991/92, likely caused by the eruption of Mount Pinatubo, and other than the mysterious dip and rebound that occurred in 1996/97, SST anomalies have been relatively flat there from the late 1980s to the late 1990s. Until 2009/2010. The normal dip after the seasonal spike did not occur early in 2009. SST anomalies rose instead.
http://i42.tinypic.com/aorurc.png
Figure 1

This post will illustrate when and where the multiple warmings occurred in the South Atlantic during 2009, and it will illustrate if they appear to be normal or abnormal occurrences when compared to past variations.

Figure 2 shows four monthly SST anomaly maps for the South Pacific and South Atlantic. Cell a, January 2009, shows a strong warming in the mid latitudes, 50S-30S, along what appears to be the Antarctic Convergence Zone. In Cell b, May 2009, the mid-latitude warming is nearing its end, but there is a short spike in SST anomalies along the equatorial Atlantic. In cells c and d, the low-latitude warming, 30S-10S, is highlighted.
http://i40.tinypic.com/vg9sn.png
Figure 2

The low-latitude warming appears to begin just before the start of the secondary rise in NINO3.4 SST anomalies and continued until recently. What I’m referring to as a secondary rise can be seen in Figure 3. NINO3.4 SST anomalies paused during July through September then rose again in October until the peak in December.

http://i41.tinypic.com/24268nm.png
Figure 3

MID-LATITUDE SOUTH ATLANTIC
The mid-latitude (50S-30S, 70W-20E) SST anomalies for the South Atlantic are shown in Figure 4. The warming in the early part of 2009 is not unusual. Larger spikes occurred in 1999/00 and 2001/03, and an earlier spike reached a similar magnitude in 1991. What is unusual in 2009 is the duration of the warming at these latitudes, having lasted a good portion of the year.
http://i39.tinypic.com/2j2gsxs.png
Figure 4

Figure 5 is a comparison of the mid-latitude SST anomalies with the SST anomalies for the entire South Atlantic basin. The mid-latitudes do account for part of the noise in the South Atlantic data, but note that the curious 1996/97 dip and rebound does not occur in the mid latitudes.
http://i43.tinypic.com/24vr9s2.png
Figure 5

Comparing the mid-latitude SST anomalies of the South Atlantic to NINO3.4 SST anomalies, Figure 6, raises more questions than it answers. Are the spikes in 1999/00 and 2001/03 lagged responses to the 1997/98 El Niño, or seeming more likely, are they lagged responses to the variations of the 1998/99/00/01 La Niña? They do not appear to be responses to the Southern Annular Mode. (Not illustrated.)
http://i42.tinypic.com/ir5myw.png
Figure 6

EQUATORIAL ATLANTIC
I’m not going to spend much time on this area since the equatorial warming in mid year of 2009 lasted a month. Recall in Cell b of Figure 2, there was a warming in May 2009 along the equator. The very short mid-year warming and the timing are not unusual as can be seen in Figure 7.
http://i41.tinypic.com/2rqzma9.png
Figure 7

Figure 8 compares the equatorial Atlantic SST anomalies to NINO3.4 SST anomalies. The equatorial Atlantic SST anomalies respond to some (the more significant) El Niño events, but not others (those that are less significant). Note the multiyear decays in the equatorial Atlantic SST anomalies after the 1986/87/88 and 1997/98 El Niño events. This is yet another SST anomaly dataset that does not respond to La Niña events.
http://i41.tinypic.com/241romg.png
Figure 8

LOW LATITUDE SOUTH ATLANTIC
The South Atlantic low latitude SST anomalies are illustrated in Figure 9. This is the area with the significant warming in the latter part of 2009. Refer back to Cells c and d of Figure 2. What stands out for me with this dataset is how flat it is for most of the term. How flat?
http://i39.tinypic.com/aw410w.png
Figure 9

If we shorten this dataset to the period from July 1983 to November 2008 (the bottoms of the troughs in those years), Figure 10, the linear trend is 0.08 deg C per CENTURY. Can’t get much flatter than that.
http://i43.tinypic.com/vhryjb.png
Figure 10

Figure 11 is a comparison of South Atlantic low latitude SST anomalies and the SST anomalies for the South Atlantic basin. Though there are periods of divergence, the variations of the low-latitude data correlate well with the basin data--the low latitudes exaggerating the basin data.
http://i43.tinypic.com/2us76he.png
Figure 11

Smoothing those two datasets and comparing them to NINO3.4 SST anomalies reveals the influence of major ENSO events on the South Atlantic, and the lack of influence of minor ENSO events.
http://i41.tinypic.com/25znfk4.png
Figure 12

PUTTING THE RECENT RISE INTO PERSPECTIVE
Let’s take a look at the longer-term HADISST SST anomaly data for the South Atlantic and the Low Latitudes of the South Atlantic. While the responses of those datasets to the 2009/10 El Niño was significant, they are dwarfed by the reactions to the 1972/73 El Niño. Refer to Figure 13.
http://i42.tinypic.com/1z4ah6q.png
Figure 13

Note how the South Atlantic data remains relatively flat from the late 1950s to the early 1970s. It shifts up approximately 0.1 deg C after the 1972/73 El Niño, then remains flat from ~1976 to 1983. It clearly shifts again ~0.2 deg C in 1982/83, then remains relative flat for the next few decades. Curious.

A QUICK LOOK AT THE WEEKLY DATA
I began writing this post a few weeks ago, and as I always do, I created the graphs first. Just to keep the post somewhat up-to-date, Figure 14 is a comparison graph of the weekly SST anomalies for the Low Latitudes of the South Atlantic and for the South Atlantic basin starting in January 2000 and ending on the week centered on April 28, 2010. Both datasets have begun their declines. The question now is: how low will they go?
http://i41.tinypic.com/6prebt.png
Figure 14

AND ANOTHER LOOK AT THE 1996/97 DIP IN SOUTH ATLANTIC SST ANOMALIES
As part of a long-term project, I’ve been creating animations of SST anomalies where I’ve used the averaging feature (12-month) of the KNMI Climate Explorer map-making software in an attempt to minimize the seasonal variations and weather noise. So far the animations have looked good. Example: Figure 15 is a .gif animation of Atlantic SST anomalies that captures the 1996/97 dip and rebound in the South Atlantic. It appears to show the dip was a response to the unofficial (very weak) La Niña of 1996/97
http://i44.tinypic.com/303h7xy.gif
Figure 15

Figure 16 shows the Atlantic SST anomalies for the Atlantic for approximately the same period but without the smoothing. It would be difficult at best to determine what caused the dip from that animation.
http://i42.tinypic.com/eqeavp.gif
Figure 16

SOURCES
The maps in Figure 2 and the OI.v2 SST anomaly data are available through the NOAA NOMADS website:
http://nomad3.ncep.noaa.gov/cgi-bin/pdisp_sst.sh?lite=

The HADISST data and the enhanced map-making features used to create the animations are available through the KNMI Climate Explorer:
http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere