On The Recent Unisys Sea Surface Temperature Anomaly Maps and Cooling of Northern Hemisphere Ocean Surfaces

Guest Post by Bob Tisdale

UPDATE 2: As of Monday November 3, 2014, Unisys is no longer updating their sea surface temperature anomaly maps. See the update at the end of the post.

# # #

UPDATE: A week-old news report from Unisys clarifies a few points. See the update at the end of the post. Thanks, Michele.

# # #

The recent cooling of the sea surfaces in the Northern Hemisphere has caused a lot of comments and questions around the blogosphere…especially when someone displays a sea surface temperature anomaly map from Unisys, Figure 1.

Figure 1 (Looks Ominous)

As you’ll recall, back in June of this year, Unisys had to change sea surface temperature products when NOAA discontinued the dataset they had been using. See Figure 2, which is a screen cap of the Wayback Machine archived edition of the Unisys SSTa webpage for June 25^th, while Unisys were updating their product.

Figure 2

As shown, Unisys advises they were planning to change to the NOAA/NCEP RTG-SST analysis. See the NOAA/NCEP Real-Time, Global, Sea Surface Temperature (RTG-SST) Analysis webpage. It’s probably safe to assume Unisys wound up using the RTG-SST product, when they began displaying the updated maps in July 2014.

If you were to click on the Global SST anomaly map in the left-hand menu on that NOAA webpage, Figure 3 would be displayed. It’s the same day, and it should be the same data, with assumedly the same referenced base years. (I can’t find what base years they’re using, and I can’t see Unisys going to all the trouble to create new daily climatologies for sea surface temperature anomaly maps.)

Figure 3 (That looks ominous, too.)

Obviously, the color-coding of the temperature anomaly scales for the Unisys and NOAA presentations is different. NOAA’s color scale is readable. Unisys’s is not in the map above. Though it’s impossible to tell from the color bar below the Unisys map, they have historically used “cool” blue and green colors for positive anomalies, even above +1.0 deg C. See the older map here. The “warmer” yellows don’t appear until the anomalies are about +2.5 deg C. Thus, the tendency for the Unisys maps to look “cooler”.

But there is another significant difference between the two maps. Note the graduations in the latitudes of the NOAA map. They’re linear…equally spaced. Not so for the Unisys map. Unisys appears to use a Mercator projection, which expands the latitudes as they progress from the equator to the poles. That exaggerates the “cool” anomalies at the mid-to-high latitudes of the Northern Hemisphere, making the cooling look even more significant than it truly has been.

We discussed the reasons for the elevated sea surface temperatures in the post On The Recent Record-High Global Sea Surface Temperatures – The Wheres and Whys. And as shown in Figure 4, the “blob” in the North Pacific had a major impact on the sea surface temperatures of the Northern Hemisphere. But the sea surface temperature anomalies in the Northern Hemisphere peaked in August and September, and now they’re coming back down.

Figure 4

If you were to look at the Unisys animation, Figure 5, you’d think we were heading into a new ice age.

Figure 5

But, if you look at the animations of seasonal sea surface temperature anomaly maps with more subdued color scaling, like the one from NOAA here and the one from CMC Environment Canada here, you’ll be able to catch something else. While the sea surfaces of the high latitudes of the Northern Hemisphere have cooled—no doubt about that—there appears to have been a slight warming at mid latitudes. In other words, the warming of the mid-latitudes should offset some (but not all) of the cooling at higher latitudes.

One would think there will be a noticeable drop in the sea surface temperatures of the Northern Hemisphere in October. Unfortunately, with the NOAA NOMADS system off-line, I don’t have easy access to the weekly Reynolds OI.v2 SSTa data. So we’ll have to wait for a few more days, until NOAA provides their October updates for sea surface temperature data.

As soon as October 2014 sea surface temperature data appears for the NOAA ERSST.v3b or the Reynolds OI.v2 data, I’ll bring you up to date on how substantial (or insignificant) that decline in recent weeks has been in the Northern Hemisphere.

Enjoy your weekends.

# # #

UPDATE: On the thread of the cross post at WUWT, Blogger Michele (in her comment here) provided a link to an October 22, 2014 Unisys news update titled Unisys SST Anomaly Maps. Unisys writes:

Unisys Weather has been receiving reports of incorrect maps of the daily SST contours. The data used to make the maps are the official NOAA RTG-SST and anomaly grids, being pulled directly from the NOAA NWSTG.

We have found an issue with the color scale of the mapping, and this has caused some users to misinterpret the maps. The color scale being used by WXP is wrapping, causing the same color to appear at very low and very high ends of the color table. The color table is also stretched beyond the actual values in the SST anomaly plot. We are working to correct the color table and color bar issue.

A comparison of the Unisys generated map with the NOAA RTG-SST anomaly maps indicate the contours are in line with NOAA’s maps.

# # #

UPDATE 2: Harold Ambler of TalkingAboutTheWeather explains why Unisys is no longer updating their sea surface temperature anomaly maps in his post UNISYS pulls down map showing dramatic ocean cooling. Thanks, Harold.