## Watt about the Sea Surface Temperature?

I don’t want to discuss this in too much detail as Sou over at HotWhopper has already done quite a thorough job. Basically, however, there is a recent Watts Up With That (WUWT) post called by land and by sea. It includes the figure below and in the text says

In addition, the land temperature rises much more rapidly than the SST.

Land and ocean temperature anomalies (credit : Willis Eschenbach – WUWT).

The implication seems to be that they seem to be the same in the 60s and 70s, but for some reason the land temperature has risen faster, in recent times, than the ocean temperature. Isn’t it fairly obvious that this is precisely what one would expect.

Down to a depth of 700m, the oceans have a mass of about 2.5 x 1020kg. Sea water has a specific heat capacity of 4000 J kg-1 K-1 and so to increase the ocean temperature by 1 K would take require an energy input of 1024 J. The atmosphere has a mass of 5 x 1018 kg. Estimating the land mass involved is more difficult, but unless the energy can penetrate deeply it is less than 1018 kg. Let’s say the total mass of land and atmosphere is 1019 kg. The specific heat capacity is about 1000 J kg-1 K-1, so it would take 1022 J to increase the temperature of the land and atmosphere by 1 K. I appreciate that this is simple, but it is showing that to change the ocean temperature by some amount would require 100 times as much energy as changing the land and atmosphere temperature by the same amount.

Additionally, observations suggest that 90% of the excess energy in the last few decades has gone into the ocean. This would suggest that the change in ocean temperature should be about 10 times smaller than the change in land and atmosphere temperature. This is not quite what is seen in the figure above, but that’s because I’ve ignored any gradients in the energy distribution and so one might not expect the ocean temperature (or atmospheric temperature) to increase by the same amount at all levels.

The simple point, however, is that the ocean temperature rising more slowly than the land temperature is not a surprise. It would be much more of a surprise if they did rise at the same rate. Why are the temperature anomalies similar in the 60s and 70s? Well, these are anomalies, not absolute temperatures. It’s the difference between some long-term mean and the current average temperature. The long-term mean is typically based on the period 1950-1980 (or 1960-1990) and so one would expect the anomaly to be close to zero during that period. That’s the only reason they’re the same in the 60s and 70s. As shown over at HotWhopper, if you extend it to earlier times, they start to diverge again, as expected.

I’ve been trying to not be too unpleasantly critical of what is said on WUWT in the hope that it might be possible to discuss climate change in a manner that isn’t so antagonistic. It’s posts like this, where they seem to show a complete lack of understanding of basic science, that makes me wonder if it is indeed possible to engage with those who typically post on WUWT.

In the comments below, caerbannog666 explains where to get a python script (credit to Kevin C at Skeptical Science) that will produce temperature anomalies from the raw data. I downloaded this about 15 minutes ago, followed the instructions and indeed have produced a figure (below) of the temperature anomaly data (solid line). I haven’t been through the code in detail, but it does indeed appear to do exactly as described. The dashed line is the NASA temperature anomalies and indeed appears remarkably similar to that determined very quickly with a python script that is about a page long.

Temperature anomalies determined using the python script described in the comments below (solid line) together with the NASA results (dashed line).

This entry was posted in Anthony Watts, Climate change, Global warming, Watts Up With That, Willis Eschenback and tagged , , , , , , , , , . Bookmark the permalink.

### 10 Responses to Watt about the Sea Surface Temperature?

1. Sou says:

One thing worth pointing out is that sea surface temperatures aren’t the same thing as ocean temperatures. Ocean temperature usually refers to the top 700 m I believe, or if specified (deeper ocean) down to 2000 m. The sea surface temperature is just the temperature of the very top ‘skin’ of the ocean. I don’t know how close is the relationship between ocean temperature and sea surface temperature. Something new to research. (I can attest that the top few cm in a body of water can be much warmer than the water below.)

When Australia had it’s broiling January heat wave, the sea surface temperature was at record highs. Most of what I’ve read blames the late monsoon as the primary cause of the heat wave. Why was the monsoon delayed? I haven’t looked into that. Some reports did mention the record SSTs as well.

2. Yes, that is a very good point. I was hinting at that a little in the post, but I have to acknowledge that I am no expert at sea surface temperatures or ocean temperatures, so was simply trying to look at this using some fairly basic physics. As you say, anyone who has spent any time in the ocean should recognise that there can be quite a change in temperature in the first few metres.

The way I’ve always understood it is that the ocean being a large heat sink means that it is harder to heat the sea surface up to the same temperature as the atmosphere when it is hot, but it can remain warmer than the atmosphere when it gets cold. Hence the variation in sea surface temperatures is typically smaller than the variation in land/atmospheric temperatures.

3. I see now what you mean. I’ve used ocean temperature in the post, when the figure is really for sea surface temperature. Should probably have made the distinction clearer. That’s the problem with writing these posts quickly before starting work 🙂

4. This is for any WUWT visitors (or anyone else) who still needs to be convinced that the global land temperature results produced by NASA/NOAA/CRU really are reliable and robust.

Last year, “Kevin C” posted a neat little global-temperature calculation script to skepticalscience.com. It reads in GHCN V3 data (raw or adjusted) and performs straightforward area-weighted averages of the station data anomalies. It produces results that are amazingly similar to the official NASA/GISS “meteorological stations” results.

With the script, you can compare raw vs adjusted data results and rural vs. urban results to see for yourselves what non-issues UHI and “data adjustments” really are.

The script is written in python — if you snip out the comments and print statements, the entire thing will fit on a single 8.5×11 printed page. That’s right — a one-page program will replicate the official NASA results amazingly closely, even with *raw* temperature data. Study it and you will see that the algorithm is quite straightforward — there is absolutely no data “manipulation” or “adjusting” involved. (How could there be? The whole program is just 1 page long!)

I hacked at the code ever so slightly to make it easier for folks to perform an “apples vs. apples” comparison of the script’s results with the official NASA met-stations results, and then added a bunch of comments to explain to new users exactly how to set it up and use it (including installing python, downloading/unpacking the GHCN data, etc.)

Folks who are still skeptical of the robustness of the global temperature data are invited to download the script from https://docs.google.com/file/d/0B0pXYsr8qYS6SXkzMW1nY2Zodlk/edit?usp=sharing and try it out themselves.

After you download the script, go to http://www.python.org and install python 2.7 (python 3.X not recommended — too new). Windows users should right-click on the ghcn-simple.py file that you downloaded from the above link and select “open with idle” (as in “Eric Idle” — the python software project was named in honor of Monty Python). Mac users — you are on your own here, but I’m sure that it’s just as easy.

You can then read the comments in the ghcn-simple.py file to find out to get the temperature data and process it to produce your own global-temperature results.

5. Thanks, I think I’ve seen a similar comment you made somewhere else. I’m tempted to have a look at this as I’ve been meaning to learn python as it seems superior to what I currently use. This may be a good way to learn.

6. Yeah, I just pitched this package over at blog.hotwhopper.com — I’ve found it to be an effective tool (at least when demonstrated in person) to counter the typical Wattsian misinformation about the global temperature record. So any time I hear about Watts and Co spreading their (ahem) fertilizer about the NASA/NOAA global temperature results, I trot out this little python script.

It’s simple, transparent, and easy for computer-savvy folks to figure out how to use.

It ties in nicely with the “Don’t just tell ’em, *show* ’em!” approach that I’ve found to be fairly effective with climate “fence sitters”.

But as much as I’d like to claim that I wrote that python script, I can’t… because I didn’t. 😦

7. George Montgomery says:

At a fundamental level, the folk at WUWT don’t know the difference between temperature and heat.

8. Exactly, I’m glad someone else has noticed this. I was amazed to watch even Don Easterbrook (who is an emeritus professor of Geology), in a Senate Energy, Environment & Telecommunications Committee hearing, basing his assessment of whether or not global warming is happening on temperature anomaly data only.

9. For a bit of an eye-opener, make a plot of the python script’s results vs. the equivalent NASA results (NASA’s “meteorological stations” index). The NASA met-stations index numbers can be downloaded from here: http://data.giss.nasa.gov/gistemp/graphs_v3/Fig.A.txt

The script results will line up with the numbers in the “Annual_Mean” column amazingly closely.

It’s a nice demonstration of the robustness of the NASA global-temperature work. You really can get 98 percent of the answer with just a fraction of 1 percent of the effort. It also shows that if NASA & Hansen were really manipulating the temperature data to produce their global-warming results, they sure didn’t get much additional warming out of all of their “data manipulation” efforts!

Anyway, a plot of the script results vs. the official NASA results might be a nice result to put up in a new post.

10. Done. I’ve added the NASA data to the figure, and it is indeed remarkably similar.