[asa] The Mystery of Darwin Airport Station 0 Solved

From: Rich Blinne <rich.blinne@gmail.com>
Date: Mon Dec 14 2009 - 18:48:26 EST

I found some things over the weekend that might explain some of the strange
results. The 0 station record is the presumed WMO station. The other
stations are duplicates nearby and the numbers of them are in decreasing
order of length of service. After 1941 the overlapping raw data tracked very
very well. Unfortunately, being the WMO station doesn't guarantee being the
SAME station. In the case of Darwin Airport the WMO station moved in 1941.

Here's Darwin Airport note start and end dates.

http://www.bom.gov.au/climate/averages/tables/cw_014015.shtml

and Darwin PO

http://www.bom.gov.au/climate/averages/tables/cw_014016.shtml

Here are the metadata records:

http://www.bom.gov.au/climate/cdo/metadata/pdf/metadata014015.pdf

http://www.bom.gov.au/climate/cdo/metadata/pdf/metadata014016.pdf

If you look at the metadata the temperature records do not overlap and we
see the two stations are 10km away from each other. In short, Darwin Airport
Station 0 is a fiction. It does not exist.

The best way to get an integrated view in my opinion is to use the raw
temperature after 1941 since the adjustments are quite small and the
raw temperature for station 0 prior to 1941 and splice it so the end points
match. This will homogenize the Darwin PO into the Darwin Airport. This
approach is midway between the bogus cooling and the equally bogus quick
rise. Even the most alarmist GW scenarios don't move as fast as either the
raw or adjusted data does at 1941. What homogenization assumes is that
everything changes slowly and it tends to dampen any movement. So when I saw
this originally both approaches looked like there was too much change.

http://www.appinsys.com/GlobalWarming/climgraph.aspx?pltparms=GHCNT100AJanDecI188020080900111AS50194120000x

So at 1941 we shouldn't be going way up or way down. It looks like lying to
the computer confused it. *This may actually give us some useful info to
make things better. Each unique station should be unique. The use of an
artificially merged station 0 appears to cause problems with the
homogenization.*

To show that this is not an issue with homogenization in general see what
the Australian BOM came up with:

http://www.bom.gov.au/cgi-bin/climate/hqsites/site_data.cgi?variable=meanT&area=nt&station=014015&dtype=raw&period=annual&ave_yr=T

Note the Aussies were not confused by seeing the 11-year running average.
Note the split in the trend.

http://www.bom.gov.au/cgi-bin/climate/hqsites/site_data.cgi?variable=meanT&area=nt&station=014015&dtype=raw&period=annual&ave_yr=11

If you tell the homogenization the truth (it's actually two stations and not
one) you get a much more moderate result. In fact, the trend amount of 0.13
degrees C/decade is exactly what you would expect given AGW. No great
conspiracy is required here. Just the old com sci acronym GIGO (garbage
in/garbage out).

Rich Blinne
Member ASA

To unsubscribe, send a message to majordomo@calvin.edu with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Mon Dec 14 18:49:08 2009

This archive was generated by hypermail 2.1.8 : Mon Dec 14 2009 - 18:49:09 EST