Many persons visit the Unisys Sea Surface Temperature Anomaly webpage to check on the surface temperatures of the oceans. Some people prefer the webpage due to the color scaling, which has light blues extending into positive anomalies. Unfortunately, the scaling makes sea surface temperature anomalies appear cooler than they are.
Example: The sea surface temperature anomalies on the map for today, Figure 1, appear very cool in the extratropical portions of the North Atlantic (20N-65N, 100W-20E) and North Pacific (20N-65N, 100E-100W). Some parts are cool, no doubt about it, especially along the east coast of Asia and in the Bering Sea.
Figure 1
Note: Just in case you’re wondering if the recent volcanic eruption in the Aleutians caused the cooling in those portions of the North Pacific, we can look at the Unisys animation here and determine the cooling started before the May 13th eruption. Then again, we can’t say the volcanic aerosols haven’t contributed to the cooling since then.
But if we look at the weekly sea surface temperature anomaly data for the extratropical portions of the North Atlantic and North Pacific, we’ll see that the temperatures there are cool, but not abnormally so. See Figures 2 and 3.
Figure 2
###############
Figure 3
A few things do stand out, however, and we can see them easier if we look at the monthly data. Based on the linear trends, the extratropical North Pacific (Figure 4) hasn’t warmed since 1994, and in the North Atlantic (Figure 5), the extratropical sea surface temperatures have warmed very little since 1997, even with the spike in 2012.
Figure 4
###############
Figure 5
The sea surface temperatures of the tropical North Pacific (0-20N, 100E-100W), of course, haven’t warmed appreciably in over 31 years. See Figure 6. And yet we know that somewhere in the Northern Hemisphere that sea surface temperatures are warming. Unfortunately, it’s the tropical North Atlantic (0-20N, 20W-20E), Figure 7, the spawning ground of hurricanes.
Figure 6
###############
Figure 7
As a reminder, we discussed the sea surface temperatures of the hurricane breeding grounds recently here.
And as we’ve illustrated and discussed for more than four years, the ocean heat content records and satellite-era sea surface temperature data, indicate the oceans have warmed naturally. If the topic of the natural warming of the global oceans in new to you, refer to my illustrated essay “The Manmade Global Warming Challenge” [42MB].
SOURCE
The Sea Surface Temperature anomaly data used in this post is available through the NOAA NOMADS website:
http://nomad1.ncep.noaa.gov/cgi-bin/pdisp_sst.sh
or:
http://nomad3.ncep.noaa.gov/cgi-bin/pdisp_sst.sh?lite=
For those who haven’t noticed:
The Unisys page also has the “new” version with yellow to red colors for all positive anomalies: http://weather.unisys.com/surface/sfc_daily.php?plot=ssn&inv=0&t=cur
“light blues extending into positive anomalies”
This is a peeve of mine – why can’t we all use a standard
shades of yellow-to orange-to red for high anomalies,
white for no change
shades of cyan-to blue-to purple for low anomalies?
Bob
The peak summer North Pacific SST has dropped nearly 1 degree C since 2005. That is not significant?[fig 2]
Bob
The SST for the Northern Hemisphere has cooling trend for the last 10 years .http://www.woodfortrees.org/plot/hadsst2nh/last:120/plot/hadsst3nh/last:120/trend
The north Atlantic (fig 3, 5) looks like a “micro Holocene” complete with a Younger Dryas just preceeding. Fractal pattern and all that.
Not sure what baseline is being used by Unisys, but in all the various SST anomaly maps I’ve seen, the differing baselines make them look a little different in the degree of warmer or cooler.
Bob
There was a report on the BBC this morning that the Basking sharks we expect around our west coast-notably Cornwall-are absent so far this year as the waters are abnormally cool-1 or 2 degrees below normal.
Living right next to the ocean in the next county along-Devon-I can personally attest to that, In fact I haven’t swam here for around 6 years. The waters have got noticeably cooler.
tonyb
La Nina is strengthening a bit. Might enhance N Atlantic TS activity this yr if it continues.
Could the diagonal cold anomaly stripe across the North Atlantic represent a faltering North Atlantic drift / gulf stream?
Thanks again for a good presentation.
Meteo France is forecasting a ‘rotten summer’ in part due to the cold sea temperatures in the NE Atlantic. While their worst projections suggest it could equal the cold anomaly of 1816, (https://diggingintheclay.wordpress.com/2013/05/27/a-year-without-a-summer/) it probably won’t be that cold, just not summer as we’ve gotten to know it in the last few warming decades.
Verity Jones says: “While their worst projections suggest it could equal the cold anomaly of 1816…”
And without a colossal volcano.
Brian D says: “Not sure what baseline is being used by Unisys, but in all the various SST anomaly maps I’ve seen, the differing baselines make them look a little different in the degree of warmer or cooler.”
I would assume Unisys is using the standard daily SST data from NOAA (as far as I know It’s the only daily SST data available), and it uses 1971-2000. There are other factors that impact the data on the maps. The NEDIS only uses nighttime satellite data and it seems to run warmer. See the discussion here:
http://bobtisdale.wordpress.com/2009/09/02/a-note-about-sst-anomaly-maps/
Regards
phlogiston says: “Could the diagonal cold anomaly stripe across the North Atlantic represent a faltering North Atlantic drift / gulf stream?”
The cooling along the Gulf Stream has appeared regularly over the past couple of years, if memory serves. I haven’t looked into it. But maybe it’s in response to a seasonal shift in the NAO?
Bob, in figs 2 & 3 there is a pronounced annual peak in the years following 1995. As these are anomaly graphs, what causes these peaks?
beng says: “La Nina is strengthening a bit. Might enhance N Atlantic TS activity this yr if it continues.”
I’ve seen papers that present the cause of the reduction in hurricanes during El Ninos (increased wind shear) and the data they’ve presented to support it. But, as far as I know, they don’t suggest that La Ninas contribute to more–that is, it’s two phases: either El Nino (less hurricanes) or not El Nino (more hurricanes).
Bloke down the pub says: “Bob, in figs 2 & 3 there is a pronounced annual peak in the years following 1995. As these are anomaly graphs, what causes these peaks?”
Part of the additional seasonal cycle in the anomaly data after 1995 is caused by the base years NOAA uses for anomalies (1971-2000). If I were to shift the base years to the period of 1995 to 2012 for example, those seasonal variations would decrease then, but they’d then show up during the early part of the data. Is the difference in the seasonal cycles caused by the two large volcanos during the period of 1971-2000? Not sure, but they’d probably skew the normal seasonal cycles for a couple of years afterwards.
i’m with bloke down the pub fig 2 and 3 seem to still have the annual cycle. unless the cooling is seasonal
It needs to be kept in mind that we don’t have any long-term data on ocean temperatures. While the Navy did some work in the 1950s, the degree of error was one degree fahrenheit for sensors typical of the time, and there was no comprehensive measurement of oceanic temps.
The 1970s saw the first buoy network, and so we have about 40 years’ worth of temperature data. However, this is anything but comprehensive due to the number of buoys.
Bottom line: we don’t actually know what ocean temps have been. It’s like testing one car for miles per gallon and assuming that every car from their inception in the 1890s to now have the same mileage.
Ask the participants in a recent charity swim round Southwold pier how cold the North Sea is. From memory around 90 had to be rescued, and 2 taken to hospital with full blown hypothermia. It was a disaster, and a complete miscalculation by the organisers. They were lucky no one died.
There’s a NOAA SST anomaly image master homepage at
http://www.ospo.noaa.gov/Products/ocean/sst/anomaly/index.html
They take 2 images per week. The current year has all images listed, and there are links at the bottom, to previous years going back to 1996. References for future arguments about ENSO…
That is not an El Nino; *THIS* is an El Nino http://www.ospo.noaa.gov/data/sst/anomaly/1997/anomnight.12.30.1997.gif
That is not a La Nina; *THIS* is a La Nina
http://www.ospo.noaa.gov/data/sst/anomaly/2008/anomnight.1.10.2008.gif
Climate Weenie says:
May 28, 2013 at 7:31 am
“light blues extending into positive anomalies”
This is a peeve of mine – why can’t we all use a standard
shades of yellow-to orange-to red for high anomalies,
white for no change
shades of cyan-to blue-to purple for low anomalies?
..==..
Ye-es, but, goodness me, do I have a peeve: I would have one band every four or five in a severely contrasting colour:
dark blue
mid-dark blue
fawn
mid blue
quite light blue
very light blue
mid green
very light pink
palest violet
white
pale yellow
mid purple
light yellow
yellowy orange
orange
pale turquoise
deep orange
Barbie pink
light red
emerald
mid red
maroon
deep lavender
cream
navy blue
The contrasting colours allowing the eye to pick out the approximate point in the scale . . . .
Auto
Bob Tisdale,
You’re the perfect guy to pose this thought to.
I’ve been thinking that maybe the AMO/PDO and Arctic Ice extent are part of oscillator.
The water starts to warm, creates surface winds that reinforce the warming, but at some point the collection of warm (or opposite cold) water flips a “switch”, pumping the warm water to the poles to cool. In the Arctic there’s a lot of ice melting. At some point enough of the heat has been discharged, it turns off, to await the build up of heat in the ocean to trigger it again.
Where this cycle takes 70-80 years to run through.
The AMO and PDO are very clear in the data. The PDO went negative (cool) around 2005 and the AMO went positive (warm) around 1995. As Bob has shown the PDO is primarily an after effect of ENSO but using it as an index is still useful.
Bloke down the pub says:
May 28, 2013 at 10:15 am
Bob, in figs 2 & 3 there is a pronounced annual peak in the years following 1995. As these are anomaly graphs, what causes these peaks?
Decreasing aerosols.
Decreased aerosols = decreased aerosol seeded clouds = increased summer solar insolation + increased winter radiative cooling.
In the extratropical N Atlantic the increasing amplitude of the seasonal anomalies is exactly what you would expect to see from decreasing N American aerosol production.
Just noting that I think Bob’s charts show that there is significant seasonality remaining in the Reynolds Oi.V2 dataset in many regions. This only affects the anomalies versus the gross temperatures.
Some parts seems to be okay like the ENSO regions but other regions should not have the peaks and troughs that continue to recur on a continuing annual basis. It is a big problem because this is the generally accepted dataset now.
MiCro says: “You’re the perfect guy to pose this thought to. I’ve been thinking that maybe the AMO/PDO and Arctic Ice extent are part of oscillator…”
Hi MiCro: Refer to the following recent post on multidecadal sea surface temperature variations. If you’re not aware, the PDO, as represented by JISAO’s PDO index, is in no way similar to the AMO, and shouldn’t be thought of in the same way. They are calculated/determined differently. The AMO is detrended sea surface temperature data of the North Atlantic. The PDO is determined through a drawn-out statistical analysis, and the PDO index is actually inversely related to the sea surface temperatures of the North Pacific, north of 20N.
http://bobtisdale.wordpress.com/2013/05/14/multidecadal-variations-and-sea-surface-temperature-reconstructions/
With respect to sea ice, I presented the relationships between the AMO and a number of other variables in the following post:
http://bobtisdale.wordpress.com/2012/09/24/how-much-of-an-impact-does-the-atlantic-multidecadal-oscillation-have-on-arctic-sea-ice-extent/
You should be able to confirm some of your hypotheses with respect to wind patterns by looking at sea level pressure indices like the AO, NAO and NPI.
And I have a couple of questions for you to ponder: How does the additional sea ice melt alter the meridional overturning circulation (MOC) in the North Atlantic and North Pacific? Is the oscillation frequency a function of mid-depth MOC? And does the change in the wind patterns provide positive or negative feedback? If it’s positive feedback, then something else has to switch it from one mode to the other.
Regards.
Speaking of Arctic sea ice I noticed a nice circumpolar wind pattern has set up recently. It will be interesting to see how long this lasts and what impact it has on sea ice. If it works like I think it might we could see the largest increase in the summer minimum level evah. 😉
The Navy in the 60’s was collecting extensive ocean temperature data. From personal experience every ASW capable ship would conduct a BT drop once every 4 hours, unless in formation, then only one ship did. The results, including a bucket temperature, optioned with a galvanizes steel bucket where submitted, if memory serves, to Navy Numerical Weather Center, Monterey, for Pacific readings. I think there was a NNWC for the Atlantic but I was just a Pacific sailor and that was a long time ago. I have no idea how accurate our BT’s were, also in those pre GPS days our navigational accuracy undoubtably let much to be desired. When XBT’s came in around 1966 temperature accuracy might have improved, but not navigation. There were rumors the Navy was paying US Flag merchant ships to make regular BT drops also. I have no idea when that program started and it might be continuing today. I have no idea why more use of that data isn’t made. I am sure it is no longer classified. I can’t even remember if it ever was.
Some parts seems to be okay like the ENSO regions but other regions should not have the peaks and troughs that continue to recur on a continuing annual basis. It is a big problem because this is the generally accepted dataset now.
You seem to think the seasonality isn’t real. I guess it’s possible a serious error has crept into the data processing, but I think unlikely.
Such seasonality can not be a permanent feature of an anomaly dataset, which means something has changed. You start to see the change between 1994 and 1998, After 1998 the change is clear. This coincides with the end of the late 20th century warming.
Either some effect in the last quarter of the 20th century decreased/stopped or some effect started/increased, around the middle of the 1990.
Decreased clouds are the obvious candidate, because they produce the opposite effect on temperature between summer and winter.
Note, decrease clouds, especially low level clouds, cause a spurious warming in minimum temperatures, which makes land surface temperature measurements unreliable in detecting cloud caused temperature changes. SSTs have no such problem.
The east coast of Asia should not be cool in this PDO phase. Aberrations happen. If this condition persists it will be a major challenge to our conception of PDO. Not just my wild thermohaline ideas, but the Chen index based on alternating rainfall regimes between Asia and North America.
Stephen Fox says:
May 28, 2013 at 11:59 am
This disastre was down to the race beginning just before the tide turned. The north sea is ALWAYS cold at this time of year and never gets above 16°C all year.
Bob Tisdale says:
May 28, 2013 at 10:24 am
+
Philip Bradley says:
May 28, 2013 at 3:33 pm
Thanks for that, always happy to see something from a new perspective.
Bob – are these anomaly values a difference from a static number established at some point in history, or a difference from a running data series over a period of time? I’m after the baseline against which the anomaly is generated and wonder about the validity of that zero point.
Well my pet peeve is giving ANY color to the +0.5 and -0.5 bands. They are functionally “normal” for both.
The tone of this article is along the lines of “Not cold so nothing to see”.. I would assert there is a lot to see. In Theory, we are into Global Warming about 1/2 Century now. Yet most of that chart is a nice light blue or a nice light turquoise (meaning ‘near no change’). IMHO, that “blue” (even the positive anomaly blue) is important since it shows that “Nothing Happened”… the ocean is as cold as ever and NOT warmed up.
Hi Bob-san:
Thank you for the excellent presentation.
From following Atlantic SSTs, it seems that the AMO may move from its 30-yr warm cycle, that started around 1995, to a 30-year cooling cycle a bit earlier than normal.
1) What say you, Bob? Is there any evidence to suggest the AMO cool cycle may start sooner than 2025?
2) I notice that when the AMO entered its warm phase in 1995, the Arctic Sea Ice Extent anomaly started falling. Conversely, when the PDO entered its 30-yr cooling phase in 2008, Arctic Sea Ice Extents on the Pacific side starting setting records.
Isn’t the Arctic Ice Extent simply a function of AMO/PDO 30-yr warm/cool cycles?
3) I’ve heard it argued that the Antarctic Ice Extent is setting records due to decreased salinity of the oceans surrounding Antarctica from melting Antarctic glacier Ice, which decreases the freezing point of ocean water.
From looking at data, it seems that salinity around Antarctica from 1980 to the present has only dropped from about 34.70ppt to 34.68ppt.
Could such a small differential in salinity really cause that much of a change in Antarctic Ice extents?
Sorry for all the questions, but your comments on any of these questions would be appreciated.
Thank you.