slabocean.brian

Here are the initial conditions: [png][pdf] Note that there is an Antarctic ghost because I didn't make these hemispherically symmetric. I'm hoping that the model won't mind too much.

EXPERIMENT 1: NO Q-FLUX

The following plots show results from the first 20 years of a control and a double CO2 integration. The points in timeseries are monthly mean values, and the gray line is the control (355ppmv) while the red line is the doubled CO2 (710ppmv).

First some time series (global mean values) showing the evolution of a few variables. The idea was to wait until the global means come into some kind of equilibrium before doing any substantial analysis. Unfortunately, as demonstrated by the time series, no equilibrium seems to be reached in 20 years. The warmer world shows some signs of stabilizing, but the control world continues to cool at quite a clip. The problem is that there is no ocean heat transport. The lower boundary is simply an isolated column (50 m) of water.

• FIGURES:
  • Temperature at 900 hPa [pdf][png]
  • Total precipitable water [pdf][png]
  • Short wave cloud forcing [pdf][png]
  • Sea surface temperature [pdf][png]
  • Specific humidity at 900 hPa [pdf][png]
  • Long wave cloud forcing [pdf][png]
  • Downwelling solar flux at the surface [pdf][png]
  • Upwelling long wave flux at the top of the model [pdf][png]
  • Net long wave flux at the surface [pdf][png]

Also, just to show how different these two simulations have become, I made cute little plots showing sea-ice fraction as a colored contour and SST as a simple line contour overlain. I've put the pngs to the right, with the control version above the warmer climate. Notice the significant difference in sea-ice extent. These are the each from 243rd month of simulation, near the end of the timeseries above.

• Control
  • [png][pdf]
• CO2 doubled
  • [png][pdf]

EXPERIMENT 2: FIXED, SYMMETRIC Q-FLUX

The following plots show results from the first years of a new control integration. The only difference between this run and EXPERIMENT 1 (above) is the q-flux, which is a prescribed oceanic heat flux.

The q-flux used for this run was based on a previous aqua planet experiment ("apeozone"), which used the observed zonal mean, annual average SST and a symmetric ozone data set. I used the packages included with the CAM to generate the mixed layer depth and then the q-flux. However between those steps I made the mixed layer depth equal to 50 m everywhere, which might bias the fluxes. After the q-flux was calculated, I edited the variable called QFLUX to be equal to the annual averaged zonal mean flux, and then averaged the hemispheres to make it symmetric. The result is then uniformly adjusted to have a vanishing global mean. The result is shown here: [png][eps]. It should be noted that I did not edit the other variables generated by 'defineqflux,' which I hope will not matter.

The following figures show timeseries of globally averaged quantities, with points at each "month." This run is still going (as of 19 September 2005), and the plots will be updated as more data is generated.

• FIGURES:
  • Temperature at 900 hPa [pdf][png]
  • Total precipitable water [pdf][png]
  • Short wave cloud forcing [pdf][png]
  • Sea surface temperature [pdf][png]
  • Specific humidity at 900 hPa [pdf][png]
  • Long wave cloud forcing [pdf][png]
  • Downwelling solar flux at the surface [pdf][png]
  • Upwelling long wave flux at the top of the model [pdf][png]
  • Net long wave flux at the surface [pdf][png]

Note the very warm climate. Whether this is due to the q-fluxes being too large is not yet known. There is no sea-ice after about 9 months, although polar SST doesn't become very warm in the first few years. To show this, I made an animation of the first ~10 years: [gif]. Note that the annotation is incorrect, the colors are only the SST, since there is no sea-ice after the first months. If you are really interested, here is the animation with the sea-ice colored and the SST contoured: [gif].

Here's a quick glimpse at the changing surface energy budget [png][eps]. The figure shows the 12-month mean surface fluxes from the 5th and 15th years, along with the net surface flux. Here's the global mean values for each term:

A more comprehensive look at the evolving surface flux can also be seen. I have plotted the components of the surface energy budget (excluding q-flux) for each year up to year 18 [png][eps]. There is also the zonal mean SST for each year [png][eps], and the zonal-mean net surface imbalance, calcuated by Fnet - Qflux for each year [png][eps]. Finally, but probably most useful is the global mean net surface flux timseries (from annual averages) [png][eps].