Hello Heiko, CF
Many thanks for your feedback, that is really useful. I think I appreciate the utility of these standard names, the feedback I have had is that diagnostics like this are widely used by forecasters and model analysts.
The concerns raised with me relate to the descriptions of these diagnostics, and the focus on 'cloud type identification' for classification.
There are similarly named diagnostics in our models, which explicitly do not categorise clouds; they take defined vertical ranges and return data based on the amount of any cloud in a grid box for that range of model levels.
The concern is that whilst the standard name is likely to be recognised by model developers and forecasters alike, the description which is bound to this standard name does not describe how these diagnostics are calculated within our model.
This has lead to the concern that whilst these seem very useful standard names, they should not be used for our models, as the textual description of the standard names do not reflect the modelled quantity in our case. The |low/medium/high| cloud amount is always determined on the basis of the vertical location of the cloud, not on the cloud type, which is not modelled.
Do you think that the description text statement:
''X_type_cloud_area_fraction is determined on the basis of cloud type and not on the vertical location of the cloud.''
properly represents the standard name and its usage across the community?
Might we be better served by using a different name for these model outputs, even if this hides the generally accepted interpretation of high/medium/low cloud? (if so, what?)
many thanks
mark
________________________________________
From: Heiko Klein [Heiko.Klein at met.no]
Sent: 11 June 2014 18:53
To: Hedley, Mark; cf-metadata at cgd.ucar.edu
Subject: Re: [CF-metadata] cloud amounts
Hi Mark,
I've been the original requestor for the
low/middle/high_type_cloud_area_fraction.
WMO has introduced the low/middle/high clouds already before models
became as famous as they are today, and they still seem to be a
excellent simplification and forecasters still use these 3 types (+fog).
Model output can come in many levels, currently in Europe between 60 and
200. This is definitely too much to transfer and to handle for a
forecaster and a reduction to 3 (4) cloud-types is done at all known
centres (Norway/ECMWF/SMHI). You are right that this reduction is often
just a summation of certain levels, but it doesn't need to - I've seen
at least one model which calculates fog different than just taking
clouds in ground-level.
Examples:
# using correct standard_name
http://thredds.met.no/thredds/dodsC/arome25/arome_norway_default2_5km_latest.nc.html
# layer named e.g. High cloud cover
http://wrep.ecmwf.int/wms/?token=MetOceanIE&request=GetCapabilities&version=1.1.1&service=WMS
Concerning your questions:
1. Forecasts are never accurate, but the definitions are at least well
established by WMO, and it is the models task to translate them as best
as possible
2. No, the cloud types predate models and model levels and eventually
accurate height measurements. If a model-level - cloud type assumption
is used, this is just a guess.
3. Really good question, I'll ask our modellers how they calculate the
different types.
Heiko
On 2014-06-11 15:41, Hedley, Mark wrote:
> Hello CF
>
> I have been having an interesting conversation with some modelling
> colleagues regarding the standard names and descriptions for cloud amount:
>
> cloud_area_fraction:
> 'X_area_fraction' means the fraction of horizontal area occupied by X.
> 'X_area' means the horizontal area occupied by X within the grid cell.
> Cloud area fraction is also called 'cloud amount' and 'cloud cover'. The
> cloud area fraction is for the whole atmosphere column, as seen from the
> surface or the top of the atmosphere. The cloud area fraction in a layer
> of the atmosphere has the standard name
> cloud_area_fraction_in_atmosphere_layer.
>
> low_type_cloud_area_fraction:
> Low type clouds are: Stratus, Stratocumulus, Cumulus, Cumulonimbus.
> "X_area_fraction" means the fraction of horizontal area occupied by X.
> Cloud area fraction is also called "cloud amount" and "cloud cover".
> X_type_cloud_area_fraction is determined on the basis of cloud type and
> not on the vertical location of the cloud.
>
> middle_type_cloud_area_fraction:
> Middle type clouds are: Altostratus, Altocumulus, Nimbostratus.
> "X_area_fraction" means the fraction of horizontal area occupied by X.
> Cloud area fraction is also called "cloud amount" and "cloud cover".
> X_type_cloud_area_fraction is determined on the basis of cloud type and
> not on the vertical location of the cloud.
>
> high_type_cloud_area_fraction:
> High type clouds are: Cirrus, Cirrostratus, Cirrocumulus.
> "X_area_fraction" means the fraction of horizontal area occupied by X.
> Cloud area fraction is also called "cloud amount" and "cloud cover".
> X_type_cloud_area_fraction is determined on the basis of cloud type and
> not on the vertical location of the cloud.
>
> In our local models, we have diagnostics labelled:
> TOTAL CLOUD AMOUNT - RANDOM OVERLAP
> LOW CLOUD AMOUNT
> MEDIUM CLOUD AMOUNT
> HIGH CLOUD AMOUNT
>
> The model calculates LOW CLOUD AMOUNT by finding the maximum amount of
> cloud cover in a model level which exists within a range defined as
> low. The model has no inclination about cloud types and makes no
> evaluation of overlap for these diagnostics.
>
> The model calculates TOTAL CLOUD AMOUNT - RANDOM OVERLAP by evaluating
> the cloud over all vertical levels and its spatial displacement.
>
> Base on this, it seems reasonable to use the CF standard name
> cloud_area_fraction
> for output fields but based on the descriptive text then I do not see
> how we could ever be able to use
> low_type_cloud_area_fraction
> for data output from our model.
>
> I would be really interested to hear from people who use these standard
> names regularly on the applicability to our case; specifically:
>
> 1. Are these definitions explicit, complete and accurate?
>
> 2. Should all uses of these standard names be sure that they are based
> on typing the cloud correctly, not on the level of the cloud within the
> model?
>
> 3. Do these high/medium/low cloud type standard names also assume that
> spatially displaced levels are summed, or are they aiming to report the
> maximum in any one level within the range?
>
> many thanks
> mark
>
>
> _______________________________________________
> CF-metadata mailing list
> CF-metadata at cgd.ucar.edu
> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
>
--
Dr. Heiko Klein Tel. + 47 22 96 32 58
Development Section / IT Department Fax. + 47 22 69 63 55
Norwegian Meteorological Institute http://www.met.no
P.O. Box 43 Blindern 0313 Oslo NORWAY
Received on Tue Jun 17 2014 - 06:57:15 BST
