Considered questions and comments arising from free to ALL "shape issue" pdf.

[Derek]

I have taken CPDN's definition of cell size, ie,
" The resolution is 2.5° in latitude by 3.75° in longitude. "
and applied it to a "shape" that is a perfect sphere along a line of latitude,
and an even eliptoid alone a line of longitude.
(This does not account for earth tide, and is not absolutely correct, but I think it is "near enough")

From the above I have produced the below table in excel (attached to this post)
Ooops cell height wrong.. redoing table.
This is the corrected table.


Which if I "zoom in" shows the cell sizes as modelled according to CPDN's definition.
Ooops cell height wrong.. redoing table.
This is the corrected table.


Not exactly "squares" are they, but that is probably the main effect (and reason why it is used) of
the primitive cylinder projection they use to show the gridding system of climate models.

As an initial "impression" the below is probably "more correct", as to the composite shape modelled alone a single line of longitude from Equator to Pole.
(repeated 96 times in each hemisphere)
Ooops cell height wrong.. redoing table.
Image not complete yet, but this plot is hopefully useful in the meantime.



Before I go further I will have to check that what I have done so far is correct, and
that the Poles in particular are handled by the models as the CPDN definition seems to imply (to me).
However, Figure 1 attributed thusly " [Figure courtesy of the Hadley Centre] " on the CPDN explanation page
appears to show one of the cell lines at what I interpret as 51.25 degrees North? Rather than 50 or 52.5 degrees North.
If so, then presumably there is one cell straddling the equator, and
what happens with the resulting 1.25 degrees at the poles is not explained.
It could be the regional models use a different base line to start from, and
this may account for the apparent difference in the illustration to the definition given by CPDN of at what latitude the cell line should / could be at.
Literally the illustration depicts the cells as about half a cell out (approx 1/2 a cell hieght) from,
their actual modelled positions according to CPDN's definition as given, vertically speaking.