Long-Term Model Intercomparison Project


INTRO

For details of the Long-Term (or Long Tail) Model Intercomparison Project aims and methodology, see the LTMIP website.

Results from the GENIE-1 model are provided as tar-balls (*.tar.gz) archives of the entire output results directory structure. The LTMIP relevant sub-directories are:

/biogem
Ocean biogeochemical (and ocean-sediment and ocean-atmosphere interface) properties, of which the most pertinant are:
  1. fields biogem 2d.nc, netCDF format 2-D fields of (mostly) ocean bottom, ocean surface, and sediment surface properties.
  2. fields biogem 3d.nc, netCDF format 3-D fields ocean dissolved and particulate tracer properties.
  3. biogem series *, ASCII time-series files. Time-series are saved for properties for which calculation of a single value global or surface (ocean or sediment) mean is ... mean-ingful *ouch*. e.g., atmospheric CO2 concentration, global total ocean carbon and alkalinity inventories and equivalent concentrations, ocean biological export, surface sediment composition.
  4. biogem year * diag GLOBAL.res, ASCII format file, one for each time-slices, containing miscellaneous global diagnostic information. It is saved at each requested time-slice with the file-name string containing the mid-point of the time-slice (as years).
/sedgem
Sediment (and ocean bottom) properties, of which the most pertinant are:
  1. fields sedgem 2d.nc, a netCDF format file containing a single time-slice taken from the end of the model experiment of 2-D fields of sediment surface and ocean bottom (biogeochemical) properties.
  2. seddiag misc DATA GLOBAL.res, a text file containing a summary of mean global sedimentation, dissolution, and preservation fluxes, and surface sediment composition.

Each experiment is run for 10,001 years, from a 2-part spin-up consisting of 25,001 years of model integration as a 'closed system' with no bioturbational mixing of the sediments, and 50,001 years as an 'open system' (with bioturbational mixing). Refer to the GENIE-1 HOW-TO and Tutorial documents on the mygenie website.

CO2 releases of 100, 1000, and 5000 PgC, in addition to a CO2 doubling (590 PgC release) are considered. CO2 release is to the atmosphere, evenly in time over the first year of model integration. Note that for model development historical reasons, release is distributed in the 2-D atmospheric module only over 'wet' (ocean) grid points (but trace gas composition homogenized at each time-step anyway).

Annual average time-slices are constructed with mid-points at the following times (years): 0.5, 10.5, 20.5, ... 90.5, 100.5, 200.5, ... 900.5, 1000.5., 2000.5, ... 9000.5, 10000.5; defiend in this file.
Time-series data (annual average) is saved at more frequent intervals than time-series, with mid-points defined here.

GENIE-1 results: PHASE-1

The experimental designs are summerized in an Excel sheet.

For the 8-level version of GENIE-1 (GENIE8) tar-balls can be found here.

For the 16-level version of GENIE-1 (GENIE16) tar-balls can be found here.


NOTE: The actual filenames of the tar-balls and experimental runs themselves have a different mnemonic system, which is translated as follows:
  1. genie_xx_yy_zz_ ... ??? == flavor (selected science modules) and resolution, e.g.,
    genie_eb_go_gs_ac_bg_sg_rg == GENIE-1 climate module with 36x36 8-level ocean; (atmosphere+)ocean+sediment+weathering carbon cycling,
    genie_eb_go_gs_ac_bg_sg_rg_itfclsd_16l_JH == as above (ish), 16-level ocean.
  2. yymmdd == creation date.
  3. wor??? == topography. be2 == 8-level ocean, jh2 == 16-level ocean.
  4. ?????PgC_1yr == CO2 release (units of PgC).
    OR CONTROL for control experiment (no CO2 release).
  5. FEEDBACK_?????? == Feedbacks enabled in the model:

NOTE: The time-series files are to be found within the /biogem subdirectory of each unpacked tar-ball. Each file is in simple ASCII format, has a header line with sort description and units where relevant, and has its first column as model experiment time (in years). The most LTMIP-relevant file are:

[home]