Hi Pete,
Thank you for helping this effort by collecting the feedback from EUMETSAT.
On Fri, Sep 21, 2012 at 4:15 AM, Peter Miu <Peter.Miu@xxxxxxxxxxxx> wrote:
> For the following standard name, should the units be set to 1 since it does
> not have Canonical units?
>
> :standard_name="sensor_band_identifier"
> :units=1
Variables of type char/string don't have units.
> The unit of radiance is a difficult one to defined as the SI version is not
> easy to work with.
> The one defined in the proposal is the most usable for the users even though
> it is not SI.
> Units definition should take usability into consideration.
I interpret this comment as a support for the proposed units, yes?
> platform_zenith_angle & sensor_zenith_angle could be changed to be more
> generic by:
>
> :standard_name="platform_look_angle"
> :standard_name="sensor_look_angle
You are probably referring to "scan" names not "zenith" names. I agree
with the suggestion; "look" is more generic and yet still captures the
meaning.
> Another observation was made for platform_zenith_angle & sensor_zenith_angle.
> The platform_zenith_angle (units=degree) is already in the CF convention
> (this was also observed by Mike Grant; cf-satellite Digest, Vol 25, Issue 4).
I removed platform_zenith_angle from the proposed names.
> However, this is not the best name for the instruments' scan angle but
> the CF convention also includes zenith_angle and angle_of_incidence.
> It is suggested that this sufficient.
>
> Do we need sensor_zenith_angle as this may be described by the
> Angle-of-incidence ?
angle_of_incidence is not applicable to passive remote sensing because
its definition mentions a surface-approaching beam of radiation.
zenith_angle in my opinion is not precise enough for remotely sensed
data.
> We would like to add the following to your proposal to the CF:
>
> :standard_name="spectral_toa_irradiance"
I think a more compatible name would be "toa_spectral_irradiance".
What about direction: outgoing or downward?
> :standard_name="atmosphere_reflectivity_ratio"
> :unit="2*pi / (cos(SZA)) * I/I0"
This is not a valid units string.
> Note, the standard names proposed at the end of the page are GSICS related
> that have been work with & adopted by EUMETSAT and
> the other GSICS partners. This should be made clear that a wider community
> have agreed on these proposed names & units which
> could help with their acceptance.
I agree. Thanks for pointing this out.
-Aleksandar
