Claims that global warming has slowed down over the past decade were partly based on faulty data. Instead, the rate of global warming was underestimated because of a new way of measuring sea-surface temperatures, suggests a new study….
[Lead author John] Kennedy says the underestimation of the change in sea-surface temperature could account for up to 0.03 °C of the apparent slowdown in global temperatures. The correction could mean that 2010 will be the warmest year on record, surpassing 1998 and 2005.
That’s the New Scientist reporting on a new re-analysis of the global sea-surface temperature (SST) dataset by the Met Office.
Of course, everybody but the anti-science disinformers have known for a long time that the Hadley/CRU (Climatic Research Unit) temperature data UNDERestimates “” not OVERestimates “” the recent global temperature rise. Their data excludes “the place on Earth that has been warming fastest” (see “Why are Hadley and CRU withholding vital climate data from the public?” and “What exactly is polar amplification and why does it matter?“). NASA’s James Hansen has made this point for years.
Last December, the Met Office had concluded that “The global temperature rise calculated by the Met Office’s HadCRUT record is at the lower end of likely warming.” In a 2009 analysis, “New evidence confirms land warming record,” the Met Office explained that they had been lowballing land-based temperatures, “because HadCRUT is sampling regions that have exhibited less change, on average, than the entire globe over this particular period.”
There is little doubt that these two problems are responsible for a large fraction of the apparent slow down of the rate of warming in the Hadley/CRU dataset — and that the rest of the slowdown is due to well understood factors such as the deepest solar minimum in a century, which we are just coming out of. The Met Office said as much last week.
I emailed Kennedy to get the study, which has been accepted for publication in Remote Sensing of Environment but isn’t online yet. I also wanted to ask him if the second paragraph cited above in the New Scientist article was accurate — it is — and whether the Met Office would be incorporating these corrections for its final 2010 temperature calculation. Kennedy elaborated:
Regarding the accuracy of the New Scientist paragraph. I think it is correct in so far as 0.03 is an upper limit on the size of the effect and that such a change in the global average temperature of 2010 in the HadCRUT3 data set (0.52 from January to October) would mean that the average was higher than the average for 1998 as a whole (0.52) and 2005 as a whole (0.47). Note that the uncertainties quoted on annual average temperatures are typically around 0.1K, so the effect on an individual year is likely smaller than other effects.
I would add that there are still considerable uncertainties in the sea-surface temperature measurements (as described in the paper) and the difference between ship and buoy data is only one factor that needs to be taken into account. This is why we are not yet applying any corrections to the SST data based on the RSE paper. We are still working on a more comprehensive analysis that takes these other factors into account. This will not be ready in the next couple of months.
So the Met Office may not make its correction in time for it announcement sometime in January as to the ranking of 2010 among hottest years. So the deniers may have something to hold onto for a while. Ultimately, this correction will force them to reorder all of the recent years from this:
The simple explanation of this correction is here:
Over the past decade, sea-surface temperature has mostly been measured by thermometers on buoys, whereas previously it was measured aboard ships. Ship measurements tend to be too high because the water warms up as it is taken on board.
So although the newer buoy measurements are more accurate, the switch in method has erroneously shown sea-surface temperatures appearing to level off.
“Compared with ships, buoys show cooler temperatures,” says Vicky Pope at the Met Office. “You have to be careful of false signals.”
Here is a slightly more detailed explanation, from ReportingClimateScience’s piece, “Met Office to revise global warming data upwards“:
Satellite data has reported a bigger increase in sea surface temperatures than in situ data from buoys and ships, according to Met Office scientist John Kennedy. “We suspect that there has been this difference for quite a while. And when we make a correction for the data from buoys we find that the trend from in situ data is much closer to the trend observed by the satellites,” explained Kennedy. “This is what makes us think, all other things being equal, that the increase in the number of buoys leads to a cooling bias in the global average sea surface temperature.”
In particular, Met Office scientists found that sea surface temperature observations from the Along-Track Scanning Radiometer (ATSR) satellite instrument and observations from the same area made by in situ platforms were different. Kennedy and colleagues from the Met Office have submitted a paper on this issue that has been accepted for publication in the journal Remote Sensing of Environment (RSE) which is called “Using AATSR data to assess the quality of in situ sea-surface temperature observations for climate studies”.
The RSE paper states that “The trend in global-average SST between 1991 and 2007 calculated from in situ data was compared to its counterpart calculated from the ATSR instruments. The in situ record warms more slowly than the ATSR record, probably due to a decrease in the fraction of relatively warm-biased ship observations contributing to the global-average SST over the period and a corresponding increase in the fraction of relatively unbiased drifting buoy observations.”
The problem seems to be related to the way that some ships measure the temperature of the water which leads to the average temperature measured by ships being higher than the average sea surface temperature measured by a thermometer on a buoy. “On average the difference ranges between 0.13C and 0.18C,” Kennedy told Reporting Climate Science .Com. The scale of this difference across the globe and over the years is sufficient to add a warming of 0.03C per decade to the HadCRUT surface temperature record.
If you are a real SST junkie, here is a May 2010 paper on the subject, “Effects of instrumentation changes on sea surface temperature measured in situ.”
So the global warming deniers, disinformers, and confusionists had it half right — the Hadley/CRU data was flawed, but just not in the direction they have been saying.
Finally, it always bears repeating that, as we learned in two key 2009 papers, the planet is warming from GHGs just where climate science said it would “” the oceans, which is where more than 90% of the warming was projected to end up (see “Skeptical Science explains how we know global warming is happening.“). The key findings in the second study are summed up in these two figures
Total Earth Heat Content from 1950 to 2003 (Murphy 2009).
Time series of global mean heat storage (0-2000 m), measured in 108 Jm-2.
The second study makes clear that upper ocean heat content, perhaps not surprisingly, is simply far more variable than deeper ocean heat content, and thus an imperfect indicator of the long-term warming trend.