For the HadSST3 temperature data set and successors, it was assumed that 30% of the ships shown in existing metadata as measuring SST by buckets actually used engine inlet:

“It is likely that many ships that are listed as using buckets actually used the ERI method (see end Section 3.2). To correct the uncertainty arising from this, 30+-10% of bucket observations were reassigned as ERI observations. For example a grid box with 100% bucket observations was reassigned to have, say, 70% bucket and 30% ERI.”

The supposedly supporting argument for overwriting of data is as follows:

“It is probable that some observations recorded as being from buckets were made by the ERI method. The Norwegian contribution to WMO Tech note 2 (Amot [1954]) states that the ERI method was preferred owing to the dangers involved in deploying a bucket. This is consistent with the rst issue of WMO Pub 47 (1955), in which 80% of Norwegian ships were using ERI measurements. US Weather Bureau instructions (Bureau [1938]) state that the \condenserintake method is the simpler and shorter means of obtaining the water temperature” and that some observers took ERI measurements \if the severity of the weather [was] such as to exclude the possibility of making a bucket observation”. The only quantitative reference to the practice is in the 1956 UK Handbook of Meteorological Instruments HMSO [1956] which states that ships that travel faster than 15 knots should use the ERI method in preference to the bucket method for safety reasons. Approximately 30% of ships travelled at this speed between 1940 and 1970.”

It is very hard to believe that this reasoning justifies overwriting of data to such a huge extent. And it is absurd to say this alteration was done “to correct the uncertainty”. Not believing and overwriting data increases uncertainty and does not correct it.

This particular alteration increases the difference between 1940s and recent SST from about 0.2 to 0.3 degrees and hence accounts for about 50% of the warming since then.

———————–

Facts and quotations assembled from climateaudit:
http://climateaudit.org/2011/07/12/hadsst3/

I tried to discuss this on RC but my post was sent to the borehole. The IPCC rational falls apart if the explanation is a data error. The bigger question then becomes why climate science and the IPCC did not consider that one possible explanation – and indeed the most likely – is human error in compiling the data.

Time and time again scientists have reported new discoveries, and time and time again these new discoveries have later been found to be caused by data problems.

That isn’t to say that new discoveries don’t happen. Rather that a lot of “science” is not new. Most likely someone has already looked at the question in some detail in the past, and if your finding doesn’t match theirs, the first thing to suspect are your methods.

Human beings subconsciously introduce bias into all experiments and it requires very careful design to avoid this. We see this when new findings correct old mistakes. The correction does not happen all at once. Rather, each new study tends to move only part way towards the correct answer.

We are likely to see the same thing in this case. RC has put forward a figure of 17% and is trying to support this by heaping scorn on Steve and others that first pointed out the problem. As such, the 17% figure likely represents the lower bound of where the correction will eventually end up, as further studies case light on the problem.

Therefore we a likely over time to see a much larger fraction of the IPCC “unexplained warming” attributed to “data error” than what RC is currently suggesting.

“The differences between sea water temperature reported in the Log of Ship’s Weather Observations and specially observed sea surface temperature were studied for 6826 pairs of observations taken in the Pacific Ocean from 3 Military Sea Transport Service ships and 9 U.S. Navy Radar Picket ships during 92 different trips. For each trip the mean difference (trip bias), using the surface temperature observations for the reference, and the standard deviation of the differences were computed. These results were combined to give similar measures for each ship (ship bias and standard deviation) and estimates of these measures for a large number of ships (fleet).

“Reported sea water (injection) temperature observations vary considerably in quality. The fleet bias, or mean difference, of injection temperature observations as compared to surface temperature observations is estimated to be 1.2±0.6F and the standard deviation of differences to be 1.6F. Among the 12 ships, the ship bias, or mean of all differences derived from observations of a given ship, ranged from −0.5F to 3.0F, which is probably due to variations of the thermometer accuracy and of the thermometer installations between ships. The standard deviation about these means was 1.3F, which is a measure of the variability of present data records, provided that the bias for each ship could be determined and a correction applied.

“The variability of differences in the observations from a single ship is attributed to the system of taking and reporting sea water temperature observations from an injection thermometer. To improve the reliability of reported sea water temperature observations, a change to an electrical resistance or thermistor thermometer, specially designed and installed to measure the sea water temperature and having a remote indicator on the ship’s bridge, is recommended. If this change were made, it is estimated from trip data that the standard deviation of differences would be reduced to less than 1.0F.”

Did anyone ever see those sorts of systematic measurement errors propagated into the reported SST trend?

In Donn’s preface to his book he mentions that “As of January 1, 1949, the new International Weather Code, both land and marine, replaced the previous code in effect since 1942. The forms of reports from both land and marine stations were also changed considerably.” I expect I will find this has been dealt with as well.

The form I am looking at is in Wm. Donn’s “Meteorology with Marine Applications” p.216 from the 2nd ed. 1951.

It’s nice to see that the Team has discovered what we at Climate Audit have known for some time – the assumption that all SST measurements switched over to engine inlets in December 1941 was absurd, ad hoc and introduced without any justification or textual analysis. It remained in climate literature despite obvious evidence that the majority of reported SST measurements in 1970 were being done by buckets.