The New York Times ran a recent story about how 2016 is the
hottest year on record, and on the basis of that story, from a supposedly
dependable news source, a great many people no doubt believe that 2016 was the
hottest year on record. But was it really?
As I mentioned in a previous post, measuring the “global temperature”
is really a difficult thing to do, because there is so much day-to-day and
place-to-place variability. And things are further confused because there are
several different datasets – measures from ground station, measures from
satellites, etc, etc.
But the real issue is how accurate those measurements
are. In science all measures are approximations
limited in their accuracy by all sorts of things, just like presidential voting
polls (and we know how inaccurate those can be). In real science, measurements are always accompanied
by error bounds – by an estimate of how much the measurement might be off. Funny,
the New York Times article didn’t mention those.
So the UK Meteorological Office reports a difference between
2015 and 2016 global temperatures of 0.01º C, but with a margin of error ten
times larger of 0.1º C, which is to say they estimate the real difference is somewhere
between 0.09º C COOLER and 0.11º C WARMER.
Hardly a convincing case that 2016 was warmer than 2015.
The NOAA (National Oceanic and Atmospheric Administration)
pegs the difference between 2015 and 2016 at 0.04º C, with a likely margin of
error around 0.09º C. Satellite data
shows the difference between 1998 – the year of the last El Nino warming cycle –
and 2016 as about 0.02º C, with a margin of error of 0.10º C, meaning that
really there is no convincing evidence from this dataset that the earth has
warmed since 1998, let alone since 2015.
2016 might indeed have been the hottest year on record, but these
measurements don’t provide evidence for that. All they show is that whatever
the difference really was, it was smaller than the likely errors of measurement.
So much for good science reporting!
Posts
