[CF-metadata] An area type for non-ocean water surfaces
cote.helene at ouranos.ca
Wed May 10 13:33:32 MDT 2017
I would like to point out that an increasing number of RCMs include a lake
At Ouranos, we are performing our CRCM5 regional simulations at 0,22 deg
and 0,11 deg resolutions; meaning that many lakes are resolved - not only
the Great Lakes but also many others in the Canadian boreal forest.
In fact, we are using the same lake model (FLake for now) for both
resolved and sub-grid lakes.
To do so, we had to define a lake fraction and a new realm (lake) in order
to identify where the lake model was activated. The land surface scheme
was applied on the land fractions.
I am not an hydrologist but I am under the strong impression that river
and lake models differ greatly. This is why I have concerns with the
inland waters definition that include both lake and rivers.
That could become quickly a problem for RCMs simulations, especially in
multiple nest experiments reaching very high resolutions.
If those inland water bodies treatment is part of the land surface scheme,
I guess they could be « part of land ». This was not option for us.
This might not be a crucial issue for ESMs, but it is for RCMs. Beside,
CORDEX datasets try to follow the same conventions as CMIP5 and CMIP6.
cote.helene at ouranos.ca Tel: (514) 282-6464 poste 260
Fax: (514) 282-7131
Hélène Côté, MSc
Groupe Simulations et analyses climatiques
550, Sherbrooke ouest, 19e étage, tour ouest
Montréal, Qc, CANADA H3A 1B9
Consultez le document de synthèse « Vers l¹adaptation »
Le 2017-05-10 14:07, « CF-metadata on behalf of Daniel Neumann »
<cf-metadata-bounces at cgd.ucar.edu on behalf of
daniel.neumann at io-warnemuende.de> a écrit :
>Dear Martin, Dear Jonathan,
>I wonder whether there are applications, for which it might be
>reasonable to distinguish between rivers (or more general "running
>inland waters") and lakes?
>The reasons for distinguishing between ocean and inland water
>- ocean is a large flat area and impacts wind speed differently than
>small inland lakes
>- The ocean has a higher salinity and, hence, hosts different ecosystems
>than inland waters. Therefore, the ocean emits other substances (or more
>of them): sea salt, halogenated organic compounds, DMS, ...
>Rivers and lakes also host different ecosystems. Additionally, the
>air-water exchange might be more pronounced at the surface of running
>waters. Therefore, it might be reasonable to have both individually.
>On 09.05.2017 16:15, Jonathan Gregory wrote:
>> Dear Martin
>> As a suggestion, lakes and rivers together could be called "inland
>> that is a commonly used phrase. I think it would be most naturally
>> the land area too.
>> Best wishes
>> ----- Forwarded message from martin.juckes at stfc.ac.uk -----
>>> Date: Fri, 5 May 2017 14:00:16 +0000
>>> From: martin.juckes at stfc.ac.uk
>>> To: cf-metadata at cgd.ucar.edu
>>> CC: stephane.senesi at METEO.FR
>>> Subject: [CF-metadata] An area type for non-ocean water surfaces
>>> Hello All,
>>> Stephane Senesi has raised the issue that some CMIP6 models may have a
>>>finite area of river water, and this should be recorded as an area type
>>>/85> I suspect that such models may also represet lakes.
>>> We could just add a "river" area type. Ideally, we should make clear
>>>how this relates to other area types. E.g. is "river" part of "land",
>>>or separate from it? The definition of the CMIP5 residualFrac variable
>>>implies that lakes are considered as part of land (though it is not
>>>clear whether this applies to all lakes, or just sub-grid scale lakes
>>>-- resolved by perhaps 1 or 2 CMIP5 models).
>>> CF-metadata mailing list
>>> CF-metadata at cgd.ucar.edu
>> ----- End forwarded message -----
>> CF-metadata mailing list
>> CF-metadata at cgd.ucar.edu
>CF-metadata mailing list
>CF-metadata at cgd.ucar.edu
More information about the CF-metadata