# changes_PMIP2_tables.txt,v 1.1 2005/03/14 16:51:32 jypeter Exp History of changes made to the PMIP2 CMOR tables ================================================ (most recent first) [Up-to-date information about the PMIP2 CMOR tables can be found on http://www-lsce.cea.fr/pmip2/design/output_format/ipccadapt.shtml#pmip2_cmor_tables ] 14 March 2005 ============= * PMIP_table_A_MO Added 'rlut' and 'rlutcs' * PMIP_table_A_SE Added 'TOA' in front of the long_name attribute for : 'rsutcs', 'rlut', 'rlutcs' Added 'Radiation' in back of the long_name attribute for : 'rsdscs', 'rsuscs', 'rsutcs', 'rlutcs' 10 March 2005 ============= * PMIP2_table_A_MO, PMIP2_table_A_SE The units of intuaw and intvaw have been changed from "1.e6 kg m-1 s-1" to "kg m-1 s-1" * PMIP_table_O_SE, PMIP2_table_O_AN The cell_methods attribute of the zonal mean variables (thetaozm, sozm, stfmmc) has been changed from "time: mean" to "longitude: mean time: mean". * PMIP_table_O_MO, PMIP2_table_O_SE The standard name of hfo has been changed from heat_flux_into_ocean to surface_downward_heat_flux_in_sea_water. The ziso20 and hcont300 have been changed slightly to be more CF compliant (altered standard name, and use of a scalar dimension) A new water flux variable has been added, wfofs, but the exact definition and use of wfo and wfofs are not clear enough yet... * There have been miscellaneous small changes made in the tables updated by Ian Ross. 2 March 2005 ============ Renaming of tables: all tables now have names of the form PMIP2_table__ where is a single character specifying the variable type: A - atmosphere O - ocean S - vegetation and land-surface I - ice and is a two character (more characters later, if the need arises) string specifying the output frequency: * Frequencies requested for the PMIP2 database FX - fixed data The fixed fields were previously stored in the MO (or subtype "1" table) to be consistent with IPCC AN - annual SE - seasonal MO - monthly (IPCC subtype "1") DA - daily (IPCC subtype "2") This new naming scheme for the frequencies is more consistent with the PMIP2 lists of variables and should be easier to remember (even if it is a bit different from the IPCC naming scheme). * Higher frequencies or variables NOT requested for the PMIP2 database (but useful for some sub-projects). Note that these variables will probably NOT be stored in the official PMIP2 database X1 - extra high-frequency data (daily resolution) X2 - extra high-frequency data (12-hourly resolution) X4 - extra high-frequency data (6-hourly resolution) 21 February 2005 ================ * Renaming and reorganisation of tables. The new naming scheme is based on the organisation of variables in the MOTIF variable tables. Each table has a name of the form PMIP2_table_, where is the domain: A - atmosphere O - ocean S - surface I - ice and is the output frequency: D - daily M - monthly S - seasonal A - annual F - fixed, time-invariant data X - extra high-frequency outputs for special applications This makes it easier to find variable definitions and correlate them with the variable lists on the web pages. * The tables have all been tidied up, with all non-MOTIF variables and axes being removed. * MOTIF variable names have been made consistent and filled in where they were missing: the following variables have been renamed for consistency: 'soil_carbon_content' -> 'scarcont'; 'lcarbon' -> 'lcarcont'; 'lcpft' -> 'lcarcontpft'. Variables that did not have MOTIF names before have been assigned them: Ocean: hfo Total ocean surface heat flux ziso20 Depth of 20 deg. C isotherm hcont300 Heat content of upper 300m mpuoeiso Mean prod. east. EIV current and salinity mpuoeito Mean prod. east. EIV current and temperature mpuoso Mean prod. east. current and salinity mpuoto Mean prod. east. current and temperature mpvoeiso Mean prod. north. EIV current and salinity mpvoeito Mean prod. north. EIV current and temperature mpvoso Mean prod. north. current and salinity mpvoto Mean prod. north. current and temperature Ice: siprod Sea ice production siarea Sea ice area siext Sea ice extent sildarea Sea ice lead area sivol Sea ice volume sisnvol Sea ice snow volume sifram Sea ice transport at Fram Strait * CF names - the following changes have been made: - The CF names for the per PFT vegetation variables have been modified to remove the _per_pft component - for instance, 'vcarcont' and 'vcarcontpft' measure the same quantities, one is just the total over all PFTs of the other; they should thus have the same CF name. - The CF names of the atmospheric transport variables have been modified to (as far as possible) follow the conventions described on the CF web-site. (Note that the modifier integral_of_X_wrt_Y is not currently included in the list of acceptable CF modifiers, but it probably should be.) -> intuaw, intvaw, intuadse, intvadse - Where appropriate CF names do not exist for variables, new CF names have been created. It will be necessary at some point to negotiate these with the people who look after the CF name list. The new names assigned are: heat_flux_into_ocean hfo: Total surface heat flux depth_of_20deg_isotherm ziso20: Depth of 20m isotherm heat_content_to_300m hcont300: Heat content of upper 300m / mpuoeiso: E EIV current x salinity | mpuoeito: E EIV current x temperature | mpuoso: E current x salinity / mpuoto: E current x temperature product_of_X_current_and_Y \ mpvoeiso: N EIV current x salinity | mpvoeito: N EIV current x temperature | mpvoso: N current x salinity \ mpvoto: N current x temperature sea_ice_lead_area sildarea: Sea ice lead area sea_ice_snow_volume sisnvol: Sea ice snow volume sea_ice_transport_at_fram_strait sifram: Sea ice Fram Str. transport litter_carbon lcarcont*: litter carbon content * The 'lcarcont' and 'lcarcontpft' variables are not litter carbon /fluxes/, they are the amount of litter carbon existing at a particular time, and thus are measured in kgC/m^2. They should not have the CF names litter_carbon_flux and litter_carbon_flux_pft. They have been changed to indicate this - the new CF name (which is not in the main CF table) is litter_carbon. * The 'ts' variable has been elided from the ocean variable set, since it is identical to 'tos' there. It has been included in the monthly and seasonal ice datasets as 'ts' for "Surface temperature of ice or snow". * The two possible representations (volume for free surface models, or concentration/dilution effect) for the water flux into the ocean ('wfo') have been included as a single 'wfo' variable in the ocean CMOR tables. * The variables "Average total heat flux at the oceanic surface" and "Average total salt flux at the oceanic surface" have been elided from the ice datasets, since they are identical to the 'hfo' and 'wfo' variables from the ocean datasets. * The 'fpar' variable (in the high-frequency atmospheric variables for the dust model) has had its units changed (to a fraction) to reflect that it really is the FRACTION of photosynthetically active radiation absorbed. * The evaporation and sublimation variables in the atmospheric datasets have been modified slightly: the total evaporation plus sublimation (CF name water_evaporation_flux, "Total surface evaporation") has been assigned the PCMDI name 'evspsbl' (EVaporation at Surface Plus SuBLimation); the "Sublimation (per unit area)" variable has been assigned the (non-PCMDI) name 'sbls' (SuBLimation at Surface) and given the CF name water_sublimation_flux (which is not a standard CF name, but is in analogy with water_evaporation flux). * The singleton pressure axes pressure_pm and pressure_pb have been moved to the end of the axis definitions ("dimensions:" keyword) in PMIP2_table_AD for the variables for which they are used. This change was needed because CMOR does not appear to support the interspersal of singleton axes among normal axes - using configuration files set up this way sometimes results in a "CMOR error 104a" message when an attempt is made to write the variable. Putting the singleton axes at the end (as is done for the height axes for surface measurements) circumvents this problem. Ian Ross, University of Bristol * the units of albs were changed from "%" to "1" * some "cell_methods" attributes were updated * typos were corrected (and some others may have been added) 18 February 2005 ================ * PMIP2_table_A1, PMIP2_table_A2, PMIP_table_S1 The tables have been cleaned to remove non PMIP2 variable information, and the variable entries have been reordered to match the web site order http://www-lsce.cea.fr/pmip2/design/variables/atmosphere.shtml http://www-lsce.cea.fr/pmip2/design/variables/vegetation.shtml * PMIP2_table_A1 'ta', 'zg', 'hus', 'ua', 'va', 'wap' defined to be on 'pressure_p3' axis (3 pressure levels) instead of the standard 'pressure' axis 'evspbl', 'intuaw', 'intvaw', 'intuadse', 'intvadse' moved from PMIP2_table_AS to PMIP2_table_A1 'prw', 'cl', 'clivi', 'clwvi', 'uas', 'vas', 'rsds', 'rsdscs', 'rsus', 'rsuscs', 'rsdt', 'rsut', 'rsutcs', 'rlds', 'rldscs', 'rlus', 'rlut', 'rlutcs', 'rtmt' moved from PMIP2_table_A1 to PMIP2_table_Ase 'mrso' moved from PMIP2_table_A1 to PMIP2_table_S1 and PMIP2_table_Sse 'mrros', 'mrro', 'evspsblveg', 'tsoil' moved from PMIP2_table_A1 to PMIP2_table_Sse 'albs', 'rss', 'rsscs', 'rst', 'rstcs', 'rls', 'rlscs' added to PMIP2_table_A1 'mrsos', 'prsn', 'mrfso', 'snw', 'rsntp', 'rsntpcs', 'rlntp', 'rlntpcs', 'mrsofc', 'tro3', 'clisccp', 'trsul', 'trsult', 'prveg', 'zmla', 'rootd' removed from PMIP2_table_A1 'tauu', 'tauv' removed from PMIP2_table_A1 (now in PMIP2_table_O1) * PMIP2_table_A2 'ta' defined to be on 'pressure_pbm' axis (2 pressure levels) instead of the standard 'pressure' axis 'psl', 'hfls', 'hfss' removed from PMIP2_table_A2 (daily 'psl' not requested in PMIP2) 'zg', 'wap' added, on 'pressure_pm' axis (1 pressure level) 'hus' defined to be on 'pressure_pb' axis (1 pressure level) instead of the standard 'pressure' axis 'ua', 'va' defined to be on 'pressure_pbt' axis (2 pressure levels) instead of the standard 'pressure' axis 'rlds', 'rlus', 'rsds', 'rsus', 'uas', 'vas', 'rlut', 'rsdt' moved from PMIP2_table_A2 to PMIP2_table_Ase * PMIP2_table_AS 'evspbl', 'intuaw', 'intvaw', 'intuadse', 'intvadse' moved from PMIP2_table_AS to PMIP2_table_A1 and PMIP2_table_Ase PMIP2_table_AS is now EMPTY and has been removed! * PMIP2_table_A3, PMIP2_table_A4, PMIP2_table_A5 These IPCC-inspired tables made references to variables not requested in PMIP2 and have been removed! * PMIP2_table_Ase New table for the mean seasonal cycle atmosphere data 'hfns', 'albs', 'rss', 'rsscs', 'rst', 'rstcs', 'rls', 'rlscs' added to PMIP2_table_Ase * PMIP2_table_Aan No need for PMIP2_table_Aan, because the only requested Aan fields were the fixed geographical fields ('orog', 'sftlf' and 'sftgif'), and the fixed fields are now associated with the 'MO' (IPCC subtype 1) frequency, to be more consistent with the IPCC tables (i.e 'orog', 'sftlf' and 'sftgif' are now defined in PMIP2_table_A1) * PMIP2_table_S1 Existing 'pft' axis has been removed, and 'pseudo_level' axis has been renamed into 'pft' (and the axis data was corrected) 'vcarcont', 'vcarcont_pft', 'soil_carbon_content', 'pftfrac', 'lcarbon', 'lc_pft' moved from PMIP2_table_S1 to PMIP2_table_Sse and PMIP2_table_San Standard names have been transformed to all lowercases (PFT_fractions, LAI_pft, Biomass_burnt, River_discharge, Potential_evaporation, Potential_evaporation_per_pft) Variable names ending in '_pft' all consistenly end now with 'pft' (no underscore) Standard names of '_pft/per_pft' variables have been consistenly set to '_per_pft' 'mrros', 'mrro', 'dis', 'trans', 'transpft', 'esoil', 'potevap_pft' moved from PMIP2_table_S1 to PMIP2_table_Sse * PMIP2_table_Sse New table for the mean seasonal cycle vegetation and land surface data Added canopy height (by pft): 'cheight' ('cheightpft') * PMIP2_table_San New table for the yearly average vegetation and land surface data 14 February 2005 ================ * PMIP2_table_A5 Added "ignored: cell_bounds" for pressure1 axis to stay consistent with the matching IPCC table (IPCC_table_A5) 02 February 2005 (05_02_02) =========================== Tables cleaned for consistency with IPCC and put on the web site Jean-Yves Peterschmitt 19 January 2005 =============== Tables updated by Michel Crucifix "Update of PMIP2_O1 Added : ------ 1 variable_entry: sos conforms to cf standards but not in original ipcc table 2 variable_entry: uos long_name: Sea Surface Eastward Sea Water Velocity !NOT RECORDED IN STANDARD CF CONVENTION 3 ditto for vos (Northward) 4 variable_entry: tauu long_name: Eastward Momentum Flux not recorded in standard cf convention there is a surface stress in atmosphere, on atmosphere grid is here only provided on ocean grid. the PMIP2 WebPage on ocean variables should be corrected accordingly 5 ditto for tauv 6 variable_entry: wfo added in table _o1 for pmip because is here provided on ocean grid. appears only in table _a1 in ipcc Update of PMIP2_A2 Added : variable_entry: rsdt long_name: TOA Incident Shortwave Radiation (frequency : 1 day, over one year only) Michel Crucifix Wed Jan 19 13:30:54 GMT 2005 Exeter." 14 January 2005 =============== Tables updated by Michel Crucifix "Got feedback from Jonathan Gregory. PMIP_table_S1 was updated to include naming conventions already established for net primary productivity, gross productivity, etc. Previously existing conventions were taken from the following address: http://www.cgd.ucar.edu/cms/eaton/cf-metadata/standard_name.html Other modifications: - suppression of canopy evaporation in PMIP2_table_S1 (already present as "water_evaporation_ flux_from_canopy in PMIP2_table_A1) - correction of units in Biomass burnt - litter carbon becomes "litter_carbon_flux" - Total Living biomass becomes "vegetation_carbon" and "vegetation_carbon_ pft") - soil_respiration becomes "heterotrophic_respiration_carbon_flux" - soil evaporation becomes "water_Evaporation_flux_from_soil" A solution might still need to be found for "River discharge" Some reflexion is required about the use of "per PFT" variables PMIP_table_AS was updated to include the standard name for total evaporation. Variables previously named "transport" referring to vertical integrals were renamed, with new names inspired from the pre-existing conventions for wind_temperature and wind_specific_humidity products. "Surface_relative_humidity" was changed in relative humidity with the appropriate z coordinate in order to refer to the screen level (2 metres) Michel Crucifix Fri Jan 14 13:36:49 GMT 2005 Exeter." 9 December 2004 (04_12_09) =============== Initial version of the tables on the web site, provided by Michel Crucifix "Most of atmospheric variables are already referred to in the IPCC tables that will be used for the CMIP2 project. Therefore, these IPCC tables were just renamed into PMIP2_tables, with the same numbering. Modifications just lie in the title of the project and allowed experiment names. For those variables which are *not* in the IPCC project, a new table was built, called PMIP2_table_AS ("S" = 'S'upplemental). There are a few variables that do not appear in PMIP2_table_AS: the net clearsky shortwave and longwave fluxes, as well as the net surface heating. Indeed, the downward and upward fluxes are already available, which allow to easily reconstruct the net fluxes. This is the option retained for CMIP2. A brand new table was built for vegetation variables : PMIP2_table_V1 which should cover all our needs. At several places, I had to choose myself some conventions. Feedback from CF experts (Jonathan Gregory, Karl Taylor) might impose us some further modification. Michel Crucifix Thu Nov 25 17:13:03 GMT 2004 Exeter."