bonygrams.brian
Using results from the simulations described elsewhere on this site, I have made some preliminary so-called Bony-grams. These plots show mean values of variables ordered by "dynamical regime." To make each plot, I use the annual climatology (unless otherwise noted) and first construct the probability distribution of vertical pressure velocity at 500 hPa. This is used as a proxy for the dynamical regime, with negative values being convective (ascending) regions and positive values subsiding regions. These plots, so far, are applied only to the tropics (within 30 degrees of equator).

Once the PDF of w500 is made, variables can be binned according to dynamical regime. That means, all values in the geographic region of interest are are assigned to groups according to the value of w500 at that each location. For the plots shown here, the value at each dynamical regime is obtained by taking an area-weighted average of all the locations in that dynamical regime.

The utility of these plots is really in breaking them into terms, to examine changes in experiments and ascribe those changes to dynamical or thermodynamical processes. This analysis will be posted here shortly.
Several sets of plots have been produced so far. Each one retains the PDF of w500 as the upper panel. The lower panel will be some variable; in these preliminary plots the variables used are cloud longwave and shortwave radiative forcing and total precipitable water. The default CAM configuration has a couple extra variables included. Each figure contains two curves per panel. In each case the gray line is considered the "control" simulation, and the red curve is the "perturbed" climate.
  • CAM with slab ocean model. Perturbed climate is doubled CO2.
    • Longwave cloud forcing [png][eps]
    • Shortwave cloud forcing [png][eps]
    • Precipitable water [png][eps]
    • Total grid box cloud liquid water path [png][eps]
    • Total grid box cloud ice water path [png][eps]
    • Incoming shortwave flux at the surface [png][eps]
  • Aqua planet with analytic prescribed SST. Perturbed climate is SST (60S < latitude < 60N) warmed by 2K.
    • Longwave cloud forcing [png][eps]
    • Shortwave cloud forcing [png][eps]
    • Precipitable water [png][eps]
  • Aqua planet with prescribed observed zonal mean SST. Perturbed climate is CO2 concentration at 560 ppmv. [Note there is essentially no climate repsonse in this experiment.]
    • Longwave cloud forcing [png][eps]
    • Shortwave cloud forcing [png][eps]
    • Precipitable water [png][eps]

A table with the different terms in the Bony budget is now available for the default CAM (2xCO2) and aqua planet (+2 K and -2 K) experiments. As has been found before (Refs), the thermodynamic term seems to dominate changes in cloud-related variables. [table, pdf][no browser-friendly version yet]

Differences in this table and the values found for CAM in Wyant et al.'s Table 4 are most likely due to my use of the annual mean instead of using monthly means. For the aqua planet experiments, it is unlikely to make much difference. I will post the alternative method when I have a chance, though I only have climatological monthly means for this CAM run.
REFERENCES:

Bony, S., J.-L. Dufresne, H. Le Treut, J.-J. Morcrette, and C. Senior, 2004. On dynamic and thermodynamic components of cloud changes. Climate Dynamics 22, pp. 71-86. DOI: 10.1007/s00382-003-0369-6.

Wyant, M. E., C. S. Bretherton, J. T. Bacmeister, J. T. Kiehl, I. M. Held, M. Zhao, S. A. Klein, and B. A. Soden, 2005:  A comparison of tropical cloud properties and responses in GCMs using mid-tropospheric vertical velocity.  J. Climate, submitted 1/05, revised 6/05.
BACK TO THE RESEARCH PAGE