Bob Tisdale is in a tizz over record hot seas

Update - I've added a bit on trends and anomalies, highlighting Bob's deception - just in case a stray arrives from WUWT. Sou 15 Dec 14.


Bob Tisdale has got himself into a real tizz over the record hot seas. He's posted several articles about how the seas are only getting hotter because they are getting hotter. Six just in the last week at WUWT, would you believe! It's not that they are getting hotter from global warming. They just happen to be getting hotter. Bob seems to think it's got nothing to do with global warming. It's just warming.

Bob's article had the rather long title: "The Nonsensical “Just What AGW predicts” and Other Claims By Alarmists about “Record-High” Global Sea Surface Temperatures in 2014"

Be warned - this article is a bit long-ish. While it's too, too easy to ridicule Bob's stupid greenhouse effect denial, that's not what made this article long. What happened was that I got sidetracked down various other sea surface temperature pathways.

It's not greenhouse gases, it's magic...

Bob doesn't deny the fact that the sea surfaces are getting hotter and hotter. He doesn't deny that they've never been this hot ever before in the instrumental era. What is getting him so worked up is that he is worried that people might think it's a sign of global warming. Worse still, they might think global warming is being caused by greenhouse gases. Worse still, someone at WUWT might think the warming might be the end result of all the extra greenhouse gases we've thrown up in the planet-wide rubbish dump in the sky.

Someone had the cheek to write a comment in one of Bob's previous articles that got Bob all hot and bothered (and no doubt didn't add anything to global warming, only to Bob's personal warming). He wrote:

A well-known alarmist chose to leave a comment on my blog here, in which he stated:

 Global SSTs at record highs… Just what AGW predicts.

I not only responded to that comment on that thread...

There was nothing wrong with the snippet he quoted, so I went to his previous article to see what all the fuss was about. It was curious for two reasons. First of all it showed that occasionally a "warmist" will visit Bob Tisdale's own personal blog. Hard to credit but there you go. In this case it was David Appell, who has an alarming blog - he writes about climate! Secondly, Bob didn't like the comment but he wasn't able to dispute it. What he did instead was write something about climate models not mimicking all the variations in surface temperature at precisely the time they vary. That's not how Bob wrote it. What he wrote was:

December 13, 2014 at 3:50 am
David Appell, you’re wrong once again. First off, I let your comment through this time, even though I banned you for wasting my time a while back…and still you insist on wasting my time.

Second, climate models don’t simulate Earth’s oceans. They simulate some planet’s oceans, but they’re not Earth’s. Even with the warm sea surface temperatures this year, the models still double the warming rate of global ocean surfaces over the past 33 years:

...and proceeded to put up a a couple of charts of sea surface temperature, including this one:

Source: Bob Tisdale (in case you couldn't tell!)

You can click on the chart to enlarge it. One of the first things you'll notice is that Bob modeled monthly data, where over shortish periods it's much harder to see the long term trend. Another thing you'll notice is that the chart has this written up the top:

Data and model outputs shifted so trend lines zero at start of time series.

I've no idea why Bob thinks it necessary to have the trend lines zero at the start. What he should have done was align the charts to the same baseline anomaly, not to a zero point trend line. Then he might have got something that looked like this - if he'd used different observational data:

Data sources: KNMI Climate Explorer and Hadley Centre

For the period he chose - or close enough - the trends are as follows for the data I used:

• HadSST v.3.1.1.0 the trend is 0.132 a decade.
• CMIP5 tos RCP6.0 the trend is 0.166 a decade.

So the CMIP 5 trend is certainly higher, but nothing like the difference that Bob got, where he said that the observed trend was only 0.87 a decade. He said he took tos for the CMIP5 data, which I did too. I chose the CMIP5 mean for tos at RCP6.0 from KNMI Climate Explorer - which I think is the same data as Bob Tisdale chose. However the above chart shows HadSST instead of Bob's choice, which was Reynolds' OI SST v2.

Another thing you'll notice is that the Hadley Centre's sea surface temperature is now approaching the modeled temperature. That's a tad surprising given the models of the time didn't allow for the lower solar irradiance, low level volcanic activity or aerosols over the past few years.

Difference in observations of sea surface temperature

The main reason for the difference between the chart above of HadSST and CMIP5 and Bob Tisdale's chart is the data he used for observations. Bob used the NOAA data - Reynolds Optimum Interpolation Sea Surface Temperature Analysis or Reynolds' OI SST v2, which is described here. 

I've plotted that data, together with HadSSTv3.1.1.0 and with ERSST (used by NASA). I've aligned them all to a 30 year baseline from 1982 to 2011. 1982 was the first full year I could get in the Reynolds' data from Climate Explorer. All the data was from KNMI Climate Explorer except for HadSST, which came from the Hadley Centre. I've also extended the CMIP5 data out to 2018 - just to show the variation over time. It may make watching the deniers interesting over the next two years, because the CMIP5 mean tapers off a little. If sea surface temperature keeps going up the way it has been lately, it will rise above the model mean sooner rather than later.

Data sources: KNMI Climate Explorer and Hadley Centre

One thing you'll probably notice is that although all follow similar patterns over time, the Reynolds and ERSST are pretty close to each other whereas HadSST starts off cooler and ends up warmer.

Different baselines

Another thing is that choosing a different baseline period for the data makes a difference when doing comparisons. In the above chart, HadSST is almost touching CMIP5. The difference between the anomalies for the two are shown below:

1982-2011 baseline less the 1961-1990 baseline:

• HadSST = 0.215°C; 
• CMIP5 = 0.231°C
• Difference = 0.016°C 

The reason is the way I adjusted the anomalies, which was simple and probably not the most accurate. What I was aiming for was to use the same baseline average for each one over time.

For each data set I worked out the average anomaly for each year. I took the average of the annual anomaly/temperature over the 30 year period as the zero line. Then I deducted the annual average temperature from the baseline average for each year. The difference between one average 30 year period and another can be different for different data sets.

If you can't follow that just ask. If you can tell me how to improve the calculation (without having to average each month separately), please do so.

Why did Bob Tisdale use Reynolds' OI SST v2?

Another interesting thing you'll have noticed is the trend for Reynolds' is the lowest of the lot. Is that pure coincidence or is that why Bob Tisdale chose it? I'll leave that to you to speculate. Here are the trends - note the different period for CMIP5, which lowers the trend a tad (compared to the chart above):

• HadSST = 0.132°C/decade (1982-2014).
• Reynolds = 0.088°C/decade (1982-2014).
• ERSST = 0.098°C/decade (1982-2014).
• CMIP5= 0.161°C/decade (1982-2018)

Update: Trend vs anomaly

I may not have made it sufficiently clear. For the period that Bob focused on - 1982 to now, not only did he pick the data with the least trend, but he shoved the CMIP5 model mean for sea surface temperature way up higher above the Reynolds data. He didn't align the charts properly. So he made it look as if there was a much bigger difference between the model mean and observations than there actually is. The way he did this was by aligning the bottom of the trend lines rather than aligning the charts themselves on a common baseline of the same thirty year average. Because the trend line for the CMIP5 mean is steeper, that meant he had to shove it way up higher on the chart and move the observations down lower. So his curves on that chart are in no way comparable.

I know I said that already, and the chart comparisons show what he did. Still, I thought I'd better mention it again just in case anyone missed it. It's easy to get distracted by short-ish trends and lose sight of the fact that the seas are heating up in absolute terms. And any recent gap between the model mean and the observations is closing fast. There have been gaps before. Give me a second and I'll put up a longer term chart so you can see what I mean. Bear in mind the poor coverage prior to 1970, so uncertainty will be higher way back when. Anyway - here it is - click to enlarge as always:

Data sources: KNMI Climate Explorer and Hadley Centre

I'm wondering if Bob Tisdale even knows what temperature anomalies are, given what he's done. Or if he does know. In other words was he trying to fool his readers or is he just plain ignorant?

[Added by Sou 15 December 2014 01:16 pm AEDT]

Sources of data

Why the difference in sea surface temperatures, you may be asking, so I did some investigation. Here is what I found out.

Reynolds' OISST

This is the description of what is referred to as the Reynolds' data or the NOAA Optimum Interpolation (OI) Sea Surface Temperature (SST) V2. It has a grid of 1.0 degree latitude x 1.0 degree longitude global grid (180x360) and covers virtually all the globe, from 89.5N - 89.5S, 0.5E - 359.5E. It doesn't say where the data come from except for mentioning satallite data (see below for more info). 

The analysis uses in situ and satellite SST's plus SST's simulated by sea-ice cover....The OI analysis is done over all ocean areas and the Great Lakes. There is no analysis over land. The land values are filled by a Cressman interpolation to produce a complete grid for possible interpolation to other grids. The ocean and land areas are defined by a land sea mask (the lsmask.nc file). Land values for sst do not necessarily coincide with land values from ice analysis.

There is a paper discussing these data and correction made to version 1, which resulted in version 2.  It's gives a lot of insight into how sea surface temperatures are worked out and where the data come from. Here are some excerpts, explaining the difficulties with satellite and other data and the inconsistencies with other data sets:

The in situ SST data are determined from observations from ships and buoys (both moored and drifting). Most ship observations in our period of interest were made from insulated buckets, hull contact sensors, and engine intakes at depths of one to several meters. ...

...In late 1981, Advanced Very High Resolution Radiometer (AVHRR) satellite retrievals improved the data coverage over that of in situ observations alone. The satellite retrievals allowed better resolution of smallscale features such as Gulf Stream eddies. Because the AVHRR cannot see the surface in cloud-covered regions, the biggest challenge in retrieving SST is to eliminate cloud contamination. The cloud-clearing algorithms are different during the day and the night because the AVHRR visible channels are useful in detecting clouds but can only be used during the day. Once clouds have been eliminated, the SST retrieval algorithm is designed to minimize the effects of atmospheric water vapor. The algorithms are ‘‘tuned’’ by regression against quality-controlled buoy data using the multichannel SST technique of McClain et al. (1985)...

Tuned algorithms - I bet that was what got Bob so enamoured. It wasn't the fact that it showed the lowest trend for the period after all. It was tuned algorithms that won him over :)

ERSST

Here is the NOAA description of ERSST.  ERSST is used by NASA's GISS in it's land-ocean surface temperature record, GISTemp. It is also used by NCDC/NOAA for its global surface temperature anomalies. It has a coarser grid of 2.0 degree latitude x 2.0 degree longitude global grid (89x180). It covers almost the same amount as the OISST, from 88.0N - 88.0S, 0.0E - 358.0E.

The extended reconstructed sea surface temperature (ERSST) was constructed using the most recently available International Comprehensive Ocean-Atmosphere Data Set (ICOADS) SST data and improved statistical methods that allow stable reconstruction using sparse data. This monthly analysis begins January 1854, but because of sparse data the analyzed signal is heavily damped before 1880. Afterwards the strength of the signal is more consistent over time. 

HadSST

And here is how HadSSTv3.1.1.0 is described at the Hadley Centre website. It has the coarsest grid of the three, being 5° latitude by 5° longitude grid. It uses the same data from ICOADS as does ERSST, up until 2006. After that it uses GTS (Global Telecommunication System) observations. (GTS data is transmitted from buoys and ships as far as I can make out.)

The SST data are taken from version 2.5 of the International Comprehensive Ocean-Atmosphere Data Set, ICOADS (external web page), from 1850 to 2006 and on GTS observations from 2007 onwards. HadSST3 is produced (a slightly more detailed description) by taking in-situ measurements of SST from ships and buoys, rejecting measurements that fail quality checks, converting the measurements to anomalies by subtracting climatological values from the measurements, and calculating a robust average of the resulting anomalies on a 5° by 5° degree monthly grid. After gridding the anomalies, bias adjustments are applied to reduce the effects of spurious trends caused by changes in SST measuring practices. The uncertainties due to under-sampling and measurement error have been calculated for the gridded monthly data as have the uncertainties on the bias adjustments following the procedures described in the paper.

For a detailed description of the dataset and its production process, please read the papers (part 1 and part 2) describing the data set. We recommended you read both papers before using the data.

ICOADS

ICOADS is the International Comprehensive Ocean-Atmosphere Data Set. It:

...offers surface marine data spanning the past three centuries, and simple gridded monthly summary products for 2° latitude x 2° longitude boxes back to 1800 (and 1°x1° boxes since 1960)—these data and products are freely distributed worldwide. As it contains observations from many different observing systems encompassing the evolution of measurement technology over hundreds of years, ICOADS is probably the most complete and heterogeneous collection of surface marine data in existence. 

The ICOADS page has some interesting information, such as where the measurements come from - over time. They list ships, moored buoys, drifting buoys and C-MAN (Coastal-Marine Automated Network stations ). And the chart below shows how much of the ocean has been sampled over time and for what. The area sampled grew hugely between 1940 and 1970:

Figure 3. Percentage global ocean and coastal area (1800-2007) sampled in Release 2.5 based on area-weighted 2° boxes (smoothed) for sea surface temperature (S), requiring at least five observations per month in each box, and determined from the "enhanced" (4.5σ trimming) product that includes ship and buoy records. Other curves compare the S coverage, at five observations per month, with that for sea level pressure (P), air temperature (A), wind speed (W), total cloudiness (C), and relative humidity (R). Also plotted is the evaporation parameter (G), which is computed from S, P, A, W, and R, and thus illustrates the extent to which surface fluxes can be computed from the individual observations. Source: NOAA

I didn't expect to wander around so much in this article. It was interesting to me to see that there are differences between the data sets, even though some of them overlap in regard to sources (eg ICOADS).

Models vs observations

In the comments a few days back, Harry Twinotter remarked that deniers have changed the messaging and are redefining the meaning of "pause". They are no longer claiming "it's not warming" - well the main disinformers have stopped saying that.  Many of them have stopped talking about a "pause" altogether, (leaving aside Christopher Monckton, and he doesn't go beyond the RSS satellite lower troposphere data). With the record global temperatures of the past few months, disinformers have shifted to "it's not warming as much as the models".

I wonder how long they'll be able to keep that one up for? They'll certainly have a hard time if more model runs come out with the correct forcing for solar and volcanic activity and aerosols.

Haunted by the heat!

Bob likes to make sure nobody misses the points he is making, even when they are plain dumb. He wrote:

THE WARMING IN THE NORTH PACIFIC IS NOT THE “MISSING HEAT” COMING BACK TO HAUNT US
We’ve already seen comments around the blogosphere to the effect of “high sea surface temperatures in the North Pacific indicate the increase in ocean heat is coming back to haunt us”.

He doesn't say where he saw those comments. I can't say I have but then I could have missed it. Bob explains what he wants you to believe:

That nonsense is contradicted by data. We only have reasonably complete and reasonably realistic measurements of subsurface ocean temperatures for the past decade or so, all thanks to the ARGO program. The vertically averaged subsurface temperature data from the NODC, Figure 2, for the extratropical North Pacific (20N-65N, 100E-90W), for the depths of 0-700 meters and 0-2000 meters, both show a negative trend, and that means the extratropical North Pacific from the surface to depths of about 1.25 miles have cooled slightly, not warmed, since the ARGO floats have been in place.

What he seems to be saying is that the seas haven't warmed from the bottom up all in the same spot. He trips around various oceans and touches on different papers and finally pounces on Durack14 (which I wrote about here), which had this as the first sentence in the abstract:

The global ocean stores more than 90% of the heat associated with observed greenhouse-gas-attributed global warming.

Ooh! Bob really didn't like the mention of greenhouse gases. He's a greenhouse effect denier from way back. He wrote:

The statement “observed greenhouse‐gas‐attributed global warming” is nothing more than a clever regurgitation of climate science fallacy. Warming of the oceans to depth has been observed, but climate models are used to attribute the warming to greenhouse gases.  

It's physics, not models

Of course Bob's got it back to front and upside down. There is no fallacy - there is physics. Warming has been observed but it's physics that attributes the warming to the greenhouse effect, not climate models. The same physics is the basis of climate models. Physics drives the models not the other way around. And the physics is based on observations. It's not pie in the sky stuff. Bob's a plain vanilla denier who thinks "scientists don't know nuffin'". Every time he puts his fingers to the keyboard he demonstrates the Dunning-Kruger effect.

From the WUWT comments

Lil Fella from OZ

December 14, 2014 at 2:23 am
As usual AGW alarmists are dreaming!

Mike H. (not the one we know, unless he's Poe-ing)

December 14, 2014 at 2:57 am
You’re correct we did get UKIP and very well they did too in the elections for MEP’s. Sadly their message on useless wind and solar energy projects is lost in the noise made by the media (The BBC mainly) about their policies on immigration. UKIP need to start banging the drum about the AGW scam a lot louder. 

Brute  has noticed how WUWT has banned almost every normal person from his blog:

December 14, 2014 at 2:48 am
There are fewer and fewer vociferous warmists every year.

mairon62  doesn't read science:

December 14, 2014 at 2:49 am
When will the alarmists show some respect for their own learning-curve? They cling to their silly mono-variable (co2) explanations to describe a complex world that involves multiple variables interacting in dynamic and perhaps chaotic relationships. It reminds me of Wiley Coyote’s “95% confidence” as he madly saws through the tree branch he’s standing on, only to discover that he’s standing on the wrong side of his cut when he’s in FREE FALL. How can we learn from our mistakes if we can never admit that we make them? “Look out below!” to all the arrogant, name-calling, pseudo-scientists who have made the learning process their enemy. Thanks to Bob for yet another informative post and a great example of what to do with new information.

KNR makes some grand statements out of ignorance and only toward the end thinks to "hear" the question:

December 14, 2014 at 3:01 am
Next year should there be no ‘record ‘ the alarmist can revert back to ‘one year means nothing ‘ , the heads you lose tails you I win approch, although scientifically worthless, sure does make life a lot easier for those that wish to be be always ‘right’ no matter what the reality.

Meanwhile the problem remains that the amount of data with have on ocean temperatures is tiny compared to the vast range of the ocean , so once again we revert to claiming that one star can tell us about all the billions of others stars ‘because ‘

Hear is a question, in an ideal world how many actual valid measurements of temperature would be required both land and sea to produce an actually average temperature that has real honest value , and how many do we actual have compared to that ? 

Does this paper help answer KNR's question - maybe coupled with this?

Source: Argo UCSC

Oh, a WUWT thread about hotter seas wouldn't be complete without speculation about undersea volcanoes (somehow managing to skip warming the lower ocean and jump straight to heating the top). This time it's Gerry, England

December 14, 2014 at 4:04 am
The oceans are the difficult bit since they cover 7/8 of the planet and yet how many measuring points are there? But I would suggest that even if it were possible to create a reliable network, you would need to know what is going on below the surface at the ocean floor to put your readings in context. If an area sees warming, could it be from subsea volcanic action? Note that the melting of the antarctic peninsula ice has been attributed to warm water below not warm air above and that the heat source is volcanic.

I knew it wouldn't take long before I could refer to Frank D's volcanic calculations :)

The mods must have been napping because Rob Painting wrote - and it's still there:

December 14, 2014 at 3:36 am
The multi-model mean eliminates natural variability, so of course the multi-model mean doesn’t match the recent period dominated by the negative phase of the Interdecadal Pacific Oscillation (IPO) – the largest source of natural variability in the climate system over multi-decadal timeframes. It’s a strawman argument to make such a comparison.

The record sea surface temperatures are suggestive of the IPO moving into its warm (positive) phase. The weaker winds mean the export of warm surface water out of the tropics is slowing down, and the mixing of heat down into the subtropical gyres is weakening. The outcome of all the changes in ocean circulation should see sea surface temperatures increasing dramatically over the coming years. The sluggish transport of warm surface water out of the tropics will be very problematic for coal reefs. We will see a swarm of bleaching events during the positive phase of the IPO.

I noticed in the press there is a lot of concern at the moment about coral bleaching - also see the video here.

Rob Painting explains more about climate models, which deniers like Bob Tisdale will never 'get" - though I know he'd have read the same here and elsewhere.

December 14, 2014 at 4:10 am
It’s really not that difficult to understand. Run a whole bunch of model simulations and natural variability occurs at different times because the models are individual possible realisations of Earth’s climate. Average them all together, which is what the multi-model mean is, and you remove all the wiggles representing these natural climate variations from the individual simulations.

Bob Tisdale isn't shy about showing off his ignorance, nor does he probably realise he's substantiating what Rob Painting wrote:

December 14, 2014 at 7:50 am (excerpts)
...Rob, I presented model-data comparisons starting in 1981. Are you aware the IPO was positive for most of that time. But that’s neither here nor there.

Yes, Bob - right up until the sea surface temperature drifted below that which was modeled. He continues:

Second, you’re wrong about the IPO about being the largest source of natural variability over any timeframe. ENSO dominates the Pacific and the globe. Are you aware there are multidecadal variations in the strength, frequency and duration of El Niño and La Niña events, so that El Niños dominate during some periods and La Niñas in others? It’s easy to see when you smooth HADISST-based NINO3.4 SST anomalies with a 121-month filter.

Bob's misquoting - Rob wrote over "multi-decadal time frames" not *any* time frame. Bob knows very little about anything except ENSO and how to plot sea surface temperature charts.  Surprisingly he doesn't know anything much about the PDO. (I noticed this when he wrote an article recently without mentioning the PDO once, when he should have.) When Bob talks of "multidecadal variations" in ENSO - they demonstrate the impact of the PDO, with El Nino's not being as strong when the PDO is in its cool phase, while La Nina's can dominate. For example, from NASA (my emphasis):

The PDO can intensify the impacts of La Niña or diminish the impacts of El Niño. In its "cool, negative phase," warm water, which causes higher-than-normal sea-surface heights (because warmer water expands and takes up more space), forms a horseshoe pattern that connects the north, west and south Pacific with cool water in the middle. In its "warm, positive phase," these warm and cool regions are reversed, and warm water forms in the middle of the horseshoe.

Such phase shifts of the PDO result in widespread changes in Pacific Ocean temperatures and have significant global climate implications. During the 1950s and 1960s, the PDO was strongly negative, or cool, and global temperatures seemed to level off. During most of the 1980s, 1990s and 2000s, the Pacific was locked in a strong positive, or warm, PDO phase and there were many El Niños. We are currently in the early stages of a cool PDO phase that began around 2006. Cool, negative phases tend to dampen the effects of El Niños.

whiten wrote a very long comment trying to back up Bob's argument that global warming doesn't mean greenhouse warming - it could be something else that's causing it, but whiten doesn't say what that might be. Not even a guess. I haven't pasted the lot, but you get the drift. Anything but CO2, though "anything" isn't specified.

December 14, 2014 at 4:55 am (extract)
The bigger problem with the AGW mentality is that it confuses the GW (what we call global warming) with AGW.

In that mentality Gw is by default an AGW.

While NO ONE REALLY CONTENDS OR DISPUTES THE ~0.8C warming of the last ~150 years(actually what we call a GW) they keep claiming and interpreting any sign or effect of it as a certain AGW.

As far as I can tell, the problem of the oceans heat content in the past, previous to the hiatus, has actually been ignored by the AGW croud simply because it’s patern of variation did confuse the AGW.
The signal in that variation does not seem to have lagged behind and following the atmospheric warming as it should have been expected in the case of a GW caused by CO2 emissions or in the case of an AGW.

It’s pattern actually makes the oceans heat content to be considered more significant as a probable cause of the GW of the last 150 years. It seems like the GW has followed the oceans warming pattern actually, especially in the first part of the warming period.
Is ridiculous to think, claim and pretend that the 0.8C warming will not show any signs in the pattern of the oceans heat or warming.

Is even more ridiculous under the cicumstances to blindly and without any care at all consider the latest observed warming of oceans surface as a clear sign of AGW.

Luke tries to tell Bob that the oceans can't keep warming forever without some forcing agent. Blaming it on "weather" won't wash:

December 14, 2014 at 7:17 am
Bob, your concluding statement is nonsensical. You state “For nearly six years, we’ve discussed how weather events can lead to long-term warming of the oceans—at the surface and at depth.” If the coupled atmosphere/ocean system is not gaining heat, how can you have long-term warming of the oceans without a large decline in atmospheric temperatures? The answer, of course, is that the overall heat content is increasing tremendously- the latest estimate is about 250 trillion Joules per second.

Bob Tisdale is either thick as two planks or being disingenuous - I opt for the former. Feel free to pick the latter. Bob ignored the bit about the "coupled atmosphere/ocean system", which is the point that Luke was making. In other words, if there wasn't greenhouse warming then the overall system wouldn't be getting hotter. If the system is being forced then it can't just be "weather".

December 14, 2014 at 8:12 am
Luke, my statement is not nonsensical. I did not say the coupled atmosphere/ocean system is not gaining heat. Read what I wrote. It’s real easy to understand. Here it is again:

For nearly six years, we’ve discussed how weather events can lead to long-term warming of the oceans—at the surface and at depth.

You even quoted it, so I’m really surprised you misunderstood it. Then again, you may have been trying to misrepresent what I wrote. Now, that would not surprise me.
Have good day, Luke. 

---

Paul J. Durack, Peter J. Gleckler, Felix W. Landerer and Karl E. Taylor "Quantifying Underestimates of Long-term Upper-Ocean Warming." Nature Climate Change 5th October 2014. DOI: 10.1038/nclimate2389 (pdf here)

Reynolds, Richard W., Nick A. Rayner, Thomas M. Smith, Diane C. Stokes, and Wanqiu Wang. "An improved in situ and satellite SST analysis for climate." Journal of climate 15, no. 13 (2002): 1609-1625. doi: http://dx.doi.org/10.1175/1520-0442(2002)015<1609:AIISAS>2.0.CO;2 (open access)