Dear Jonathon,
On 22/11/2006 7:16 a.m., Jonathan Gregory wrote:
> Dear Richard
>
> Thank you for your helpful explanation of what these quantities mean as stats
> of the spectrum. I agree with you that in view of their diversity they are
> better regarded as independent quantities and given standard names as such,
> rather than being described as statistics. Your definitions in terms of moments
> would be useful to include in the description of these standard names.
>
> However, I still wonder whether there are any more self-explanatory names we
> can give to them than tm-1, tm1 and tm2. Is this what they are invariably
> called, or do there exist any longer, more obvious, designations?
At best, people would mostly say "second moment mean period", etc. At
worst, just "mean period" and leave you guessing which one. Also, there
is a "zero upcrossing period" Tz which can be calculated from a time
series record, which under certain circumstances matches Tm2. This is
already in the standard name table as
"sea_surface_wave_zero_upcrossing_period"
So maybe a solution would be a formulation like
"sea_surface_wave_mean_period_from_X", where X gives the method, eg.
Tm2="sea_surface_wave_mean_period_from_variance_spectral_density_second
frequency_moment"
Tm1="sea_surface_wave_mean_period_from_variance_spectral_density_first
frequency_moment"
Tm-1="sea_surface_wave_mean_period_from_variance_spectral_density_inverse_frequency_moment"
We already have "sea_surface_wave_significant_height". Like Tm2/Tz,
this can be calculated from the spectrum or from zerocrossing of a time
series, with not necessarily identical results, so maybe this could be
further qualified, e.g.
Hm0="sea_surface_wave_significant_height_from_variance_spectral_density_zeroth_moment"
>
> The frequency range could certainly be indicated for any such quantity by
> including a size-one or scalar coordinate variable of e.g.
> sea_surface_swell_wave_period, whose bounds would supply the range.
That sounds OK for quantities derived by integrating over a fixed
frequency range. The "ideal" default for the above quantities would be 0
to infinity, but the actual range used could be specified in this way.
Unfortunately this is not the only method used to distinguishing swell
or wind wave components. Sometimes a formulation is used that depends on
wind speed and direction, and the spectral wave direction component (as
in the WAM model that I expect Heinz and Beate would be using). In that
case, to fully describe what is meant by "swell" and "wind wave"
variables, there really should be some reference to the formulation used
to distinguish them.
We could
> also define names for frequency, rather than period, if that is preferable,
> as it seems it might be given your formulae are in terms of frequency. No
> cell_methods entry would be needed, because these are integral quantities, and
> their default interpretation is to apply to the range indicated by the bounds.
>
Sometimes a mean frequency (= 1/Tm1) is used, but period variables seem
to be a bit more user friendly.
Regards,
Richard
--
=========================================================
Richard Gorman
National Institute of Water and Atmospheric Research
PO Box 11-115, Hamilton, 3251, New Zealand
Tel: +64 7 856 1736 Mob: 021 074 7490 Fax: +64 7 856 0151
Email: r.gorman at niwa.co.nz Web: http://www.niwa.co.nz
=========================================================
Received on Wed Nov 22 2006 - 16:11:12 GMT
