Sea Surface Temperature Versus Ocean Heat Content Anomalies

OVERVIEW

This post illustrates the similarities and differences between HADISST-based Sea Surface Temperature (SST) anomalies and NODC Ocean Heat Content (OHC) anomalies, 0-700 meters. All SST and OHC anomalies in this post have been standardized.

INTRODUCTION

On an early Ocean Heat Content thread, a blogger asked if I had any graphs that compared Sea Surface Temperature and Ocean Heat Content anomalies. At that time I didn’t have any. I’m glad I didn’t respond then by creating them, since the National Oceanographic Data Center (NODC) revised and corrected its Ocean Heat Content data in October 2010. The differences between the OHC data presented by the NODC with its paper Levitus et al (2009) [Manuscript] and their October 2010 corrections and revisions are illustrated in the post Update And Changes To NODC Ocean Heat Content Data.

Sea Surface Temperature data are available in Deg C, and the NODC Ocean Heat Content anomalies are in Joules. To account for those differences, I standardized all of the data for the graphs in this post by dividing the data by its standard deviation. There is another difference between the two datasets. The SST data is presented monthly, but the OHC data is presented on a quarterly basis. To account for this, the monthly SST data were converted to quarters by averaging January-February-March data, April-May-June data, etc. The data in the graphs have also been smoothed with a 4-quarter filter, centered on the 2^ndmonth, to reduce the noise.

GLOBAL COMPARISON

The global SST and OHC anomalies are compared in Figure 1. The relationship between Sea Surface Temperature and Ocean Heat Content makes its presence known. Using the more stable Ocean Heat Content curve as a reference, there are periodic rises and falls in Sea Surface Temperature imposed on an increasing trend from the 1970s to the early 2000s. The additional rises in SST are caused by El Niño events, while the dips result most often from La Niña events (and an occasional volcanic eruption). The Ocean Heat Content curve response opposes the changes caused by the ENSO events. OHC dips and rebounds when the Sea Surface Temperatures rise and fall, and vice versa. Global Ocean Heat Content drops during an El Niño event and rises during a La Niña event.

Figure 1

###################################

Notice, however, how the variation in the OHC data appears to change in the 2000s. Not only has the rise in Ocean Heat Content slowed to a minimum, the Global Ocean Heat Content response to significant La Niña and El Niño events also appears to have lessened. There were significant La Niña events during the boreal winters of 2007/08 and 2010/11 and a mid-strength El Niño in 2009/10, yet Global OHC shows little impact. Recall that an observation system called ARGO was put in place in the 2000s. ARGO measures ocean temperature and salinity to depths of 2000 meters. Prior to ARGO there were few observations at depth in the Southern Hemisphere. Have the additional observations during the ARGO era dampened the variability of Global Ocean Heat Content data? And of course, the additional OHC surge in 2003 appears out of place, as many have noted in past Ocean Heat Content posts.

TROPICAL PACIFIC

El Niño and La Niña events impact Global Ocean Heat Content but those events occur in the Tropical Pacific. The opposing relationship between Ocean Heat Content and Sea Surface Temperature is, therefore, strongest there, Figure 2. This can be seen clearly after 1965.

Figure 2

###################################

The 1976 rise in Tropical Pacific SST anomalies is evidence of the Pacific Climate Shift. But note how a rise in Ocean Heat Content anomalies precedes it. Tropical Pacific Ocean Heat Content rose significantly from 1973 to 1976. That rise in OHC is a response to the 1973/74/75/76 La Niña event. Similarly, the rise in Ocean Heat Content from 1995 to 1996 was caused by the 1995/96 La Niña. And that increase in Tropical Pacific Ocean Heat Content provided the fuel for the super El Niño of 1997/98. The curiosity in that graph occurs in the last year. Notice how the SST and OHC data both drop. It will be interesting to see where both datasets are headed.

DIFFERENCES

For those interested, I’ve subtracted the Global OHC data from the Global SST data in Figure 3, and subtracted the OHC data from the SST data for the Tropical Pacific in Figure 4. I’ve also provided NINO3.4 SST anomalies as a reference for the frequency and magnitude of ENSO events.

Figure 3

###################################

Figure 4

###################################

HEMISPHERE AND OCEAN BASIN DATA

I’m not going to describe the comparison graphs individually for the hemispheres and ocean basins, we can discuss those in comments, but there are a few things to consider: Sea Surface Temperature observations were sparse in the Southern Hemisphere until 1982, when the satellite era for SST data started. And observations at depth for the Ocean Heat Content data were very sparse in the Southern Hemisphere until 2003.

And a note about the Pacific data: Since I’ve already presented the Tropical Pacific data, I’ve included comparisons of the data from 24N-65N for the North Pacific and from 60S-24S for the South Pacific. They are not divided at the equator.

Figure 5

###################################

Figure 6

###################################

Figure 7

###################################

Figure 8

###################################

Figure 9

###################################

Figure 10

###################################

Figure 11

###################################

SOURCE

The HADISST and NODC OHC data used in this post are available through the KNMI Climate Explorer:

http://climexp.knmi.nl/selectfield_obs.cgi?someone@somewhere