README -- Describes contents of HEMCO/SOILNOX/v2014-07 directory 28 Mar 2017 GEOS-Chem Support Team geos-chem-support@g.harvard.edu Overview: =============================================================================== This directory contains input files for the HEMCO soil NOx extension. The data was prepared by Bram Maasakkers and converted to netCDF file format by the GEOS-Chem Support Team. Files: =============================================================================== (1) soilNOx.landtype.generic.025x025.nc Contains array LANDFRAC(1440,720,24), which is placed on the GENERIC GRID at 0.25 x 0.25 degree resolution. The array contains the land fractions for the 24 different MODIS/KOPPEN biome types. Because the original data is on 0.25 x 0.25 as well, all the fractions in this file will be either 0 or 1. When this file is regridded, the values should simply be averaged, giving the fraction of a certain biome type within the (for example 2x2.5) gridbox. The data is not pegged to any specific year, so we have timestamped it with a "generic" date of 2000/01/01 @ 0h GMT. -- Bram Maasakkers (maasakkers@seas.harvard.edu) ADDENDUM: Level 1 compression (e.g. nccopy -d1) has been applied to this file in order to shrink the file size. -- Bob Yantosca, 11 Feb 2015 (yantosca@seas.harvard.edu) ####################################################################### ##### NOTE: The netCDF file attribute 'lev:positive = "up"' was ##### ##### added to this file to allow GCHP to eventually read it. ##### ##### -- Bob Yantosca (27 Feb 2017) ##### ####################################################################### (2) soilNOx.landtype.generic.025x025.1L.nc Contains the same data as soilNOx.landtype.generic.025x025.nc, but lacks the netCDF "lev" dimension. This is just as well, since the data is only defined at the surface. This alleviates some problems reading in the data. ADDENDUM: Level 1 compression (e.g. nccopy -d1) has been applied to this file in order to shrink the file size. -- Bob Yantosca, 11 Feb 2015 (yantosca@seas.harvard.edu) (3) soilNOx.climate.generic.05x05.nc Contains the following arrays which are placed on the GENERIC GRID at 0.5 x 0.5 degree resolution: NON_ARID(720,360): Fraction of non-arid soil (excluding water). (The data consists of 1's & 0's at the native resolution of 0.5 x 0.5 degrees.) ARID(720,360): : Fraction of arid soil The soil at grid box (I,J) can be considered ARID if: ARID(I,J) > NON_ARID(I,J) The data is not pegged to any specific year, so we have timestamped it with a "generic" date of 2000/01/01 @ 0h GMT. -- Bram Maasakkers (maasakkers@seas.harvard.edu) ADDENDUM: Level 1 compression (e.g. nccopy -d1) has been applied to this file in order to shrink the file size. -- Bob Yantosca, 11 Feb 2015 (yantosca@seas.harvard.edu) (4) soilNOx.fert_res.generic.05x05.nc Contains the array fert_res(720,360,366), which is placed on the GENERIC GRID at 0.5 x 0.5 degree resolution. The array contains the reservoir of N associated with manure & chemical fertilizer. The data is available per day (1-366). The data is not pegged to any specific year, so we have timestamped as days from 2000/01/01 at 0 GMT (2000 being a leap year). -- Bram Maasakkers (maasakkers@seas.harvard.edu) ADDENDUM: Level 1 compression (e.g. nccopy -d1) has been applied to this file in order to shrink the file size. -- Bob Yantosca, 11 Feb 2015 (yantosca@seas.harvard.edu) ADDENDUM 2: The original time variable vector in the file, 1-366, was not consistent with the units "days since 2000-01-01 00:00:00". The time values are now 0-365 for consistency. -- Lizzie Lundgren, 24 Jul 2017 (elundgren@seas.harvard.edu) ###################################################################### ##### IMPORTANT NOTE ABOUT UNITS! ##### ###################################################################### The actual units of the data are kg/m2 of NO coming from fertilizer. The units listed in the netCDF file are kg/m3. The reason we changed the units from kg/m2 to kg/m3 was because was because of the automatic unit conversion of HEMCO. Earlier versions of HEMCO would automatically convert fluxes to kg/m2/s and concentrations to kg/m3. A file with input units of kg/m2 would be interpreted as concentration, hence the labeling as kg/m3. You can look at it as concentration within the top 1m of the soil layer! So the correct unit is kg NO/m2 soil if looking at it in a 2D space. But the concentrations make more sense in 3D, we hence prefer kg NO/m3. -- Christoph Keller (Christoph.A.Keller@nasa.gov), 28 Mar 2017 (5) soilNOx.Inputs_MODIS_Biomes.nc Contains the various resistances for each of the MODIS/Koppen biome types. NOTE: This data is now inlined into the HEMCO soil NOx extension. (6) soilNOx.restart.geos.2x25.nc (7) soilNOx.restart.geos.4x5.nc Sample restart files for the HEMCO soil NOx extension on the GEOS-Chem 2 x 2.5 and 4 x 5 horizontal grids. These are provided as a convenience to users. These files contain the following data: float DRYPERIOD(time, lat, lon) ; DRYPERIOD:long_name = "Dry period" ; DRYPERIOD:units = "hours" ; DRYPERIOD:_FillValue = 1.e+15f ; // "" DRYPERIOD:missing_value = 1.e+15f ; // "" DRYPERIOD:gamap_category = "RST-SOIL" ; float PFACTOR(time, lat, lon) ; PFACTOR:long_name = "Pulsing factor" ; PFACTOR:units = "unitless" ; PFACTOR:_FillValue = 1.e+15f ; PFACTOR:missing_value = 1.e+15f ; PFACTOR:gamap_category = "RST-SOIL" ; float GWET_PREV(time, lat, lon) ; GWET_PREV:long_name = "Soil moisture" ; GWET_PREV:units = "unitless" ; GWET_PREV:_FillValue = 1.e+15f ; GWET_PREV:missing_value = 1.e+15f ; GWET_PREV:gamap_category = "RST-SOIL" ; float DEP_RESERVOIR(time, lat, lon) ; DEP_RESERVOIR:long_name = "N reservoir" ; DEP_RESERVOIR:units = "kg/m3" ; DEP_RESERVOIR:_FillValue = 1.e+15f ; DEP_RESERVOIR:missing_value = 1.e+15f ; DEP_RESERVOIR:gamap_category = "RST-SOIL" ; (8) DepReservoirDefault.nc Contains the default values to use for the DEP_RESERVOIR field (kg/m3) if a HEMCO restart file cannot be found. Prior to v11-02, the default DEP_RESERVOIR was set to 1e-4, but this was too low. The DEP_RESERVOIR fields in this file were obtained from the HEMCO_restart.201307010000.nc file found in the 4x5_standard directory of the GEOS-Chem Unit Tester (This is the restart file used in the 1-month benchmark simulations). (9) soilNOx.fert_res.generic.05x05.time_corrected_for_GCHP.nc Same as soilNOx.fert_res.generic.05x05.nc except with time variable vector changed from 1-366 to 0-365. The time unit is days since Jan 1 00:00:00 and therefore the first entry should have time=0 for correct interpretation by GCHP ExtData. Time=1 is interpretted as Jan 2 00:00:00 which is the incorrect date for that data slice. This causes GCHP to exit with an error if starting a simulation on Jan 1. This file is being used for the v11-02b GCHP 1-year benchmark. It will replace soilNOx.fert_res.generic.05x05.nc in v11-02c. -- Lizzie Lundgren (elundgren@seas.harvard.edu), 28 Jun 2017