At 17:45 14/04/2009, you wrote:
Quoting
P.Jones@uea.ac.uk:
Phil,
I will do that, but there seem to be two problems:
1) why would it all happen in 1997-98? its hard to believe that many
new drifters were deployed, starting just that year.
2) there are examples of abrupt shifts in other parts of the time
series - why should this be especially suspect?
thanks for any additional help on this, tom
Tom,
The issue Ray alludes to is that in addition to the issue
of many more drifters providing measurements over the last
5-10 years, the measurements are coming in from places where
we didn't have much ship data in the past.
For much of the SH
between 40 and 60S the normals are mostly made up as there is
very little ship data there.
Whatever causes the divergence in your plot it is down to
the ocean.
You could try doing an additional plot. Download from
the CRU web site the series for SH land. It doesn't matter if
is from CRUTEM3 or CRUTEM3v (the former would be better). If that
still has the divergence, then it is the oceans causing the
problem. What you're seeing is too rapid to be real.
Cheers
Phil