What initial conditions files should be used for an aqua-planet simulation? Is it possible to start with a set of standard OpenIFS initial conditions? Also how easy would it be to specifiy a highly idealised basic state / initial conditions? The reason I ask is that i am wondering if it would be possible to do a series of aqua-planet simulations changing things like the SST gradient, the position of the jet stream ect. to see how this affects certain aspects of the circulation (e.g. extra-tropical cyclones).

 

Victoria Sinclair

3 Comments

  1. Hi Victoria,

    There is some information about configuring Aqua-planets with OpenIFS. I expect you've seen this.

    The aqua-planet in the OpenIFS code is based on the Neale & Hoskins, 2001, ASL, paper. If you look in ifs/phys_ec/gp_sstaqua.F90, the different SSTs described in the paper are set up there. Other factors in the model, like gravity, are also altered in OpenIFS to match the Neale & Hoskins experiments.  The various idealized starting conditions are shows in ifs/setup/suspecg2.F90 and see also ifs/module/yomdyncore.F90.

    The difficulty is that these options are not always maintained and some work would be required to check. Changing the SST gradient would be possible in a similar way to the code in OpenIFS now but it would require some work - hard to say how much.

    The model can be started from real conditions as long as the land-sea-mask and surface geopotential (orograph) are set to zero (by using grib_set). I am adding more information to the OpenIFS webpage for this setup.

     

    Glenn

     

  2. Hello,

    I have a few questions related to the aqua-planet set-up of OpenIFS.

    1. How are the initial conditions for the aqua-planet experiments which are available on the ftp server made? Previously, I think Gabi told me that these come from the end of a one year spin up experiment but how do you start this one year experiment?
    2. By default, does the aqua-planet set up have a diurnal cycle in the incoming solar radiation?
    3. My last question is about the radiative balance of the model. My 10 year simulations have shown that the model has a huge imbalance in the top of atmosphere radiation balance - it is about 20 Wm-2. Originally I thought that this was not a problem with fixed SSTs as obviously the SSTs can not response and therefore the mean atmospheric temperature does not  change / respond either. However, now I am a bit worried as in some recent experiment where I have changed the local SST gradients (but kept the global mean temperature the same) the TOA radiation balance change a lot from my control run (e.g. from 20 Wm-2 to 13 Wm-2). So the question is has anyone accessed the radiative balance before and is it possible to constrain this / force the model into a more radiative balanced state?

    Thanks in advance,

    Victoria


    1. Hi Victoria,

      Many thanks for your careful inspection!

      1. The 1-year spin-up experiment we used to initialize your aqua-planet experiment has been started from a normal ERA-Interim field and date, there was nothing special. We used these settings: http://www.met.reading.ac.uk/~mike/APE/ (Experimental design), excluding the recommendations in point 1 instead of which we kept the standard IFS values.

      2. There is a diurnal cycle in solar radiation, please see the link above.

      3. We checked the radiation balance at the top of the atmosphere in our spin-up run now, and I have to say, we concluded the same: the difference between the top net thermal radiation and the top net solar radiation is around 15 Wm-2 (-237 and 252). Typically the global values in our normal runs are -238 and 239 (see https://www.ecmwf.int/en/forecasts/charts/physics/physics_clim2000  TOA LW CEREA ebaf, TOA SWnet CERES ebaf).

        Consulting with colleagues from physics about the problem with short wave radiation: the possible reason is that the low cloud cover is much lower than expected, especially there are no subtropical Sc areas, also less clouds in polar regions. They proposed you to decrease either autoconversion in microphysics and/or decrease erosion rate for cloud fraction.

      Gabi