Model Output Overview  

• Introduction
• File Format: NetCDF
• Directory Structure & Data Organization
• Calendar Used In Experiments
• Using ts_catalog Files To Find Variables

  Model Components  

• Atmospheric GCM
• Land Surface
• Ocean GCM
• Sea Ice
• Flux Exchange Among Model Components

  Control & Transient Forcing Experiments

• ctrl_06
• IPCC_A2_01 & IPCC_B2_01
• IS92a Experiments: gps_01, gps_02 & gps_03

References & Citations

• How To Cite The GFDL R30 Coupled Model Results
• List Of References Cited On These Web Pages

If you have comments or questions regarding these GFDL R30 Coupled Climate Model webpages please e-mail Keith.Dixon @ noaa.gov

The coupled model output stored on GFDL's NOMAD server is taken from six experiments conducted at GFDL using the GFDL_R30_c model. Data files produced by one long-running control integration (one with no changes in external forcings, e.g., constant CO2) and five climate change scenario experiments have been made accessible to interested researchers.

> The Control Experiment: ctrl_06

The ctrl_06 model run was integrated for 900 years, and produced relatively stable results in terms of climate drift ("climate drift" refers to unforced trends that can arise when the various components of the coupled model are joined together). Achieving minimal drift in the GFDL coupled model solutions has been an important consideration in the model development process. The existence of a long control integration (greater than several hundred years) with minimal drift is an important pre-condition for the successful design and interpretation of climate variability and change experiments. If substantial drift occurs over the course of an experiment, the base state of the model can be altered, potentially changing the characteristics of the model's response to external forcing and its internal variability.

The output from a long, stable control experiment such as ctrl_06 provides a database for assessing the possible range of internal variability in a model, and hence the significance of the model's response to external forcing. Also, in the absence of sufficient observational data, such long control integrations are often used as surrogates for internal variability in the real climate system on decadal and longer time scales.

Delworth et al. (2001) report that the drifts seen in the ocean component of ctrl_06 correspond to a long-term, global mean net fresh water addition to the global ocean of approximately 0.24 cm per year and a global mean net heat flux out of the ocean of approximately 0.2 W m-2. The North Atlantic meridional overturning circulation exhibits relatively little drift, but is characterized by substantial multidecadal variability. The drift in ctrl_06 sea ice extent is also small (much less than 5% per 1000 years).

On the GFDL/NOMADS server we have made available time series of monthly mean quantities for the 900 years of the ctrl_06 experiment.
For the atmosphere - land surface component go to... http://nomads.gfdl.noaa.gov/dods-data/DecCen/r30/coupled/ctrl_06/atm/ts/mon/raw/900yr/ and for the ocean - sea ice component go to... http://nomads.gfdl.noaa.gov/dods-data/DecCen/r30/coupled/ctrl_06/ocn/ts/mon/raw/900yr/.

The individual atmospheric netCDF files (10800 time levels of 96 by 80 2-D fields) are about 317 MB in size, while the ocean model's 900-year long monthly time files are about 633 MB in size, due to the fact that the ocean model has twice as many grid points in the east-west direction than does the atmospheric model's transform grid.

A subset of the atmospheric model variables have been stored out on a daily basis for a 50 year period (model years 131 to 181). These data files are about 535 MB in size, contain 18250 time levels, and are located at http://nomads.gfdl.noaa.gov/dods-data/DecCen/r30/coupled/ctrl_06/atm/ts/dly/raw/50yr/.

> Climate Change Scenario Experiments IPCC_A2_01 & IPCC_B2_01

The IPCC_A2_01 model experiment experiences transient greenhouse gas levels based upon the IPCC's SRES A2 scenario. Similarly, the transient forcings derived from the IPCC's SRES B2 scenario were used in the IPCC_B2_01 experiment. Because each of the various greenhouse gases (CO2, CH4, N20, halocarbons, etc.) are not carried separately in this version of the model, the greenhouse gas changes are applied as equivalent CO2 increases. Aerosols are not prognostic variables in the model, so the direct effects of tropospheric sufate aerosols are included as temporally and spatially varying changes in surface albedos consistent with the SRES A2 or B2 forcing scenarios.

More information on the IPCC's SRES forcing scenarios may be found at the SRES Web Page. Additional information on the A2 scenario storyline can be found Chapter 4.3.2 of the SRES (SRES is the acronym for Special Report on Emissions Scenarios). Additional information on the B2 scenario storyline can be found Chapter 4.3.4 of the SRES.

The monthly mean time series files for the IPCC_A2_01 and IPCC_B2_01 experiments are each 140 years in length. Model years 111 to 250 in the A2 and B2 experiments corresponds to scenario years 1965 to 2104. Daily model output was archived for a subset of variables during three separate 20-year periods; namely model years 156-175 (scenario years 2010-2029), model years 186-205 (scenario years 2040-2059), and model years 216-235 (scenario years 2070-2089).

The IPCC_A2_01 and IPCC_B2_01 experiments were each initialized from model year 111 of the gps_01 experiment described below. Because the the SRES scenarios forcing estimates for the period 1865 to 1964 differed somewhat from the IS92a scenario used to integrate the gps_01 coupled model experiment, some blending of the forcing data was done during the scenario period 1965-1989 for consistency. Therefore, we recommend that analyses of the IPCC_A2_01 and IPCC_B2_01 experiments be started at scenario year 1990 (model year 136).

The Climate Change Scenario Experiments: gps_01, gps_02 & gps_03

The three GFDL R30 coupled model experiments named gps_01, gps_02 and gps_03 were all forced by the same IS92a-like forcing scenario, but differ in that they were initialized from different points in the long-running ctrl_06 experiment.

The model's equivalent CO2 level rises 6% at the very start (scenario year 1866) of these three experiments, matching the difference between pre-industrial CO2 levels and that estimated to had been present in 1865. The specified equivalent CO2 levels continue rising at rates following the IS92a scenario until year 1990, and at a rate of 1% per year thereafter. The direct effect of tropospheric sulfate aerosols was included in the model as modified surface albedos that varied in time and space (after Haywood et al., 1997).

Each of the three "gps" (Greenhouse gases Plus Sulfates) experiments starts at scenario year 1865 and ends at 2089 (225 years). Experiment gps_01 was initialized from year 11 of ctrl_06, experiment gps_02 was initialized from year 51 of ctrl_06, and experiment gps_03 was initialized from year 101 of ctrl_06.

On the NOMADS server we have made available time series of monthly mean quantities for the 225 years of the three "gps" experiments. Daily model output from these experiments is not available on the server.