Steven Mosher | May 25, 2012 at 12:59 pm
<blockquote>read it again and see if you can comprehend. the difference is clear</blockquote>
OK, steven, I read it again. The difference is not clear. Stop with the drive-by postings, they're nothing but a pain and do not enhance your reputation, and explain yourself.
For example, you say:
<blockquote>In CRU take a cell that is half ocean and half land.
The final value will be a simple average of the SST and the Land.</blockquote>
No it won't, that's not how the averaging works at all. There is a good description of the actual procedure used in the HadCRUT dataset, as opposed to your simplistic fantasy, located <a href="http://www.cru.uea.ac.uk/cru/data/temperature/HadCRUT3_accepted.pdf" rel="nofollow"><b>here</b></a> on page 13:
<blockquote> <b>Blending land and marine data</b>
To make a dataset with global coverage the land and marine data must be combined. For land-only grid boxes the land value is taken, and for sea-only grid boxes the marine value; but for coastal and island grid boxes the land and marine data must be blended into a combined average.
Previous versions of HadCRUT [Jones, 1994, Jones & Moberg, 2003] blended land and sea data in coastal and island grid boxes by weighting the land and sea values by the area fraction of land and sea respectively, with a constraint that the land fraction cannot be greater than 75% or less than 25%, to prevent either data-source being swamped by the other. The aim of weighting by area was to place more weight on the more reliable data source where possible. The constraints are necessary because there are some grid boxes which are almost all sea but contain one reliable land station on a small island; and some grid boxes which are almost all land but also include a small sea area which has many marine observations. Unconstrained weighting by area would essentially discard one of the measurements, which is undesirable.
The new developments described in this paper provide measurement and sampling uncertainty estimates for each grid box in both the land and marine data sets. This means that the land and marine data can be blended in the way that minimises the uncertainty of the blended mean. That is, by scaling according to their uncertainties, so that the more reliable value has a higher weighting than the less reliable.
Tblended = ε^2_sea*Tland + ε^2_land*Tsea / ε^2_sea +ε^2_land
where Tblended is the blended average temperature anomaly, Tland and Tsea are the land and marine anomalies, ε_land is the measurement and sampling error of the land data, and ε_sea is the measurement and sampling error of the marine data.
</blockquote>
As you can see, it's not "simple averaging".
w.
↧
