Features of CMIP experimental implementation and output data for participating models. The links in column headings connect to further explanations of the symbols used in that column. Links in table cells connect to further explanation of specific details for that model.
 

CMIP Model	Initialization Method	Flux Adjustment	Run Length (years)	Surface Air Temp	Missing Data
BMRC1	2 A,B,D	none	105	I(1.5 m)	ocean data, standard deviations
CCCma	4A,B,C,D, some 3	heat, water	150	I(2 m)
CCSR	2A,B,C,D , some 3	heat, water	40	L(sigma=0.995)
CERFACS1	1A,B,C	none	40	I(2 m)
COLA 1	1A,B,C	none	50	Veg. canopy T, ocean skin T
COLA 2	1A,B,C	none	190	Veg. canopy T, ocean skin T
CSIRO	4A,B,C,D	heat, water, momentum	100	I(2 m)
ECHAM1+LSG	4A,B,C,D	heat, water, momentum	960	I(2m)	missing all, except sfc temp
ECHAM3+LSG	4A,B,C,D	heat, water, momentum	1000	I(2m)	flux-correction fields
ECHAM4+OPYC3	3A,B,C,D,E	annual mean heat, water	240	I(2m)
ECHAM4+HOPE-G	3A,B,C,D	annual mean heat, water (global mean set  to 0)	100	I(2m)
GFDL_R15_a	4A,B,C,D	heat, water	1000	L(sigma=0.990)
GISS (Miller)	4A,B,D	none	89	I(10 m)
GISS (Russell)	1A,B,C	none	98	I(sigma~0.985)	barotropic strmfcn, decadal standard devns
IAP/LASG1	3A,B,C,D,E,F	heat, water (SSS restored to obs), momentum	50	L(sigma=0.991)
LMD/IPSL1	1A,B,C	none	24	L(sigma=0.979)	decadal stndrd devns
MRI1	4A,B,C,D	heat, water	100	I(0 m)	ocean heat transports
NCAR (CSM)	2 A,B,C,D	none	300	I(2m)
NRL1	2 A,B,C,D	annual mean heat, water	36	L(sigma=0.997)
UKMO (HadCM2)	3 A,B,C,D,E,F	heat, water	1085	I(1.5 m)

 

Initialization Method (Cf. the review by Stouffer and Dixon 1997 concerning the rationale for the following categories.)

• 1.  Observational Method
• A.  Integrate the atmospheric model to equilibrium using prescribed SSTs or by coupling to an oceanic mixed-layer model.
• B.  Initialize the oceanic model, whole volume, to the observations (e.g., after Levitus), then run a "short time".
• C.  Couple the atmospheric and oceanic models without flux adjustments.

• 2.  Reduced Coupling Shock Method
• A.  Integrate the atmospheric model as in 1A above.
• B.  Integrate the oceanic model as in 1B above, or start from a constant temperature/salinity distribution with zero currents, and integrate to equilibrium.
• C.  Take fluxes from 2 A, and integrate the oceanic model to equilibrium by repeating several years of atmospheric fluxes as forcing.
• D. Either couple the atmospheric and oceanic models and integrate the fully coupled model without flux adjustments, or compute flux adjustments from 2 A and 2 C and integrate the coupled model.

 
• 3.  Coupled Restore Method
• A.  Integrate the atmospheric model as in 1A above.
• B.  Integrate the oceanic model as in 1B above, or start from a constant temperature/salinity distribution with zero currents, and integrate to equilibrium.
• C.  Couple the models from 3 A and 3 B, then integrate the coupled system while restoring the SST, surface salinity, and (optionally) sea ice towards observed values.
• D.  At some point, stop and compute the flux adjustments (which are the time average of the restoring terms in 2 C above).
• E.  Continue the coupled integration using the flux adjustments without restoring terms.
• F.  Possibly, also, iterate steps 3 C, 3D, and 3 E.

 
• 4.  Equilibrium Startup Method
• A.  Integrate the atmospheric model to equilibrium using prescribed SSTs.
• B.  Integrate the oceanic model to equilibrium while restoring the SST, surface salinity, and sea ice fields.
• C.  At equilibrium, compute the flux adjustments from 4 A and 4 B.
• D.  Couple the models from 4 A and 4 B.  Using the flux adjustments from 4 C, integrate the coupled model.

Surface Air Temperature: The definition of each model's "surface air temperature" is summarized. L denotes the temperature of the lowest atmospheric level, with the associated sigma value given in parentheses.  I denotes an air temperature that is intermediate between the temperatures at the ground and at the lowest atmospheric level, with the nominal height noted in parentheses (expressed either in meters above the surface or as an intermediate sigma level).

Missing Data: Data not provided in the CMIP1 requested data list.
 
 

Return to CMIP Model Documentation Directory

Return to CMIP Home Page
 

LLNL Disclaimers
UCRL-MI-127713