[CF-metadata] New area_types for LUMIP subgrid land-use tile reporting

From: alison.pamment at stfc.ac.uk <alison.pamment>
Date: Tue, 22 Nov 2016 14:34:12 +0000

Dear Lena,

Thank you for the clarification. I think I understand better now that primary and secondary land has a related but different meaning to primary and secondary vegetation. I still think it is OK use the name primary_and_secondary_land for the new area type, but obviously it is important to have the correct definition and we should also add some words to make a clear distinction from the primary/secondary tree names.

Here is my second attempt at the definition:
primary_and_secondary_land
"Primary and secondary land" is all land that is not in active use as cropland or pastureland, including forests, grasslands, bare ground and vegetated wetlands. "Primary land" is land that has never been subjected to human disturbance. "Secondary land" is land that has undergone human disturbance and has subsequently been abandoned. "Primary and secondary land" refers to land use, rather than to specific categories of vegetation, and is distinct from the area types for primary and secondary trees.

Is that better? I have removed the references to primary and secondary vegetation. Are there any published references for primary and secondary land use that could be mentioned in the definition? (I am thinking here of a general reference, rather than something that is intended specifically for CMIP6).

Best wishes,
Alison

------
Alison Pamment Tel: +44 1235 778065
Centre for Environmental Data Analysis Email: alison.pamment at stfc.ac.uk
STFC Rutherford Appleton Laboratory
R25, 2.22
Harwell Campus, Didcot, OX11 0QX, U.K.



> -----Original Message-----
> From: Elena Shevliakova [mailto:elena at Princeton.EDU]
> Sent: 21 November 2016 15:20
> To: Pamment, Alison (STFC,RAL,RALSP); dlawren at ucar.edu; cf-
> metadata at cgd.ucar.edu
> Subject: RE: [CF-metadata] New area_types for LUMIP subgrid land-use tile
> reporting
>
> Hi Alison,
> thank you for working on the variable list refinement. I have a few comments. I
> think there is a misalignment between existing definitions of secondary which
> refer to land cover as opposed to land use. For example, land use term
> "secondary" refers to lands which may or may not have trees, grasslands which
> were used as pasture and then abandoned would be considered secondary
> lands under CMIP6 land use scenarios. The same issues applies to primary - in
> CMIP6 LU scenario primary would refer to forested, non-forested and even bare
> lands
>
> Lena
>
>
> ________________________________________
> From: alison.pamment at stfc.ac.uk [alison.pamment at stfc.ac.uk]
> Sent: Monday, November 21, 2016 9:41 AM
> To: dlawren at ucar.edu; cf-metadata at cgd.ucar.edu; Elena Shevliakova
> Subject: RE: [CF-metadata] New area_types for LUMIP subgrid land-use tile
> reporting
>
> Dear Dave,
>
> Thank you for proposing these new area_types for LUMIP.
>
> > primary_and_secondary_land - all primary and secondary vegetation that is
> not used as cropland or pastureland, should include forests, grasslands, bare
> ground and
> > vegetated wetlands.
>
> I think the name itself is fine and we do already have some entries that refer to
> primary and secondary deciduous/evergreen trees. I think we should add
> something into the definition to explain 'primary' and 'secondary' as for the
> existing terms. Based on their definitions, I'd suggest something like:
> 'Primary land contains only naturally regenerated species where there are no
> clearly visible indications of human activities and the ecological processes are
> not significantly disturbed. Reference: 'Global Forest Resources Assessment:
> Terms and Definitions', Forestry Department of Food and Agriculture
> Organization (FAO) of the United Nations, Rome 2010,
> www.fao.org/forestry/14241-0d7b74f45b0d2cfef31599cc17e4c28cd.pdf.
> Secondary land has undergone human disturbance and has then recovered
> naturally. Reference: 'Report of the ad hoc technical expert group on forest
> biological diversity', United Nations Convention on Biological Diversity,
> www.cbd.int/forest/definitions.shtml. Primary_and_secondary_land includes all
> primary and secondary vegetation that is not used as cropland or pastureland,
> including forests, grasslands, bare ground and vegetated wetlands.
>
> Is this OK? If so, I think the name can be accepted for inclusion in the area
> types table.
>
> > urban - urban settlements
>
> Is there a reference that we could point to that contains a formal definition of
> 'urban'? If so, it would be good to include it in the area type definition. I think
> 'urban' is fine as a name.
>
> You also suggested that we should expand the definition of the existing
> area_type 'pastures':
> > pastures - Includes anthropogenically managed pastureland and rangeland
>
> Currently the definition reads 'Pastures are assumed to be anthropogenic in
> origin.' Certainly I think we could expand this to 'Pastures are assumed to be
> anthropogenic in origin. They include anthropogenically managed pastureland
> and rangeland.'
>
> OK? If so, I think this change can be accepted for inclusion in the area types
> table.
>
> > Note that in our GMD LUMIP paper, we suggest that subgrid tile variables
> should be submitted according to the following structure, using Leaf Area Index
> (LAI) as an
> > example: laiLut (lon, lat, time, landusetype4) - where the landusetype4
> dimension has an explicit order of primary_and_secondary_land, crops,
> pastures, urban. Please
> > advise if this is not correct.
>
> This looks perfectly good to me! CF doesn't impose any order on the list of land
> use types, but it is fine to adopt an additional convention for LUMIP.
>
> > Sum of fractional area for psl+crp+pst+urb may not add up to 1 for grid cells
> with lakes or glaciers. If model does not represent one of the requested tiles,
> then should be
> > reported as missing value. In cases where more than one land use tile shares
> information (e.g., if pastureland and cropland share same soil column), then
> duplicate
> > information should be provided for both tiles. Further details on tile reporting
> available in the LUMIP GMD paper.
>
> Again, I think it is fine to adopt these conventions for LUMIP to supplement the
> CF conventions.
>
> Best wishes,
> Alison
>
> ------
> Alison Pamment Tel: +44 1235 778065
> Centre for Environmental Data Analysis Email: alison.pamment at stfc.ac.uk
> STFC Rutherford Appleton Laboratory
> R25, 2.22
> Harwell Campus, Didcot, OX11 0QX, U.K.
>
>
> From: CF-metadata [mailto:cf-metadata-bounces at cgd.ucar.edu] On Behalf Of
> David Lawrence
> Sent: 21 October 2016 20:27
> To: cf-metadata at cgd.ucar.edu; Elena Shevliakova
> Subject: [CF-metadata] New area_types for LUMIP subgrid land-use tile
> reporting
>
> LUMIP is requesting subgrid output on four land-use tiles
> (primary_and_secondary_land, crops, pastures, and urban).
>
> area_types for crops and pasture are already defined in the existing cf-
> conventions.
>
> New area_types required:
>
> primary_and_secondary_land - all primary and secondary vegetation that is not
> used as cropland or pastureland, should include forests, grasslands, bare
> ground and vegetated wetlands
>
> urban - urban settlements
>
> Additionally, the existing area_type pastures could include expanded definition
>
> pastures - Includes anthropogenically managed pastureland and rangeland
>
> Note that in our GMD LUMIP paper, we suggest that subgrid tile variables
> should be submitted according to the following structure, using Leaf Area Index
> (LAI) as an example: laiLut (lon, lat, time, landusetype4) - where the
> landusetype4 dimension has an explicit order of primary_and_secondary_land,
> crops, pastures, urban. Please advise if this is not correct.
>
> Sum of fractional area for psl+crp+pst+urb may not add up to 1 for grid cells
> with lakes or glaciers. If model does not represent one of the requested tiles,
> then should be reported as missing value. In cases where more than one land
> use tile shares information (e.g., if pastureland and cropland share same soil
> column), then duplicate information should be provided for both tiles. Further
> details on tile reporting available in the LUMIP GMD paper.
> Regards,
> Dave
Received on Tue Nov 22 2016 - 07:34:12 GMT