Implementation of ozone ancil

Unified Model
1

Hi,

Sorry to bother you, but I have a couple of (hopefully rather simple) technical questions concerning the implementation of ancillaries, and in particular ozone, in the model. I am fairly certain I know the answer to these, but wanted to doublecheck with you before I do something silly (I have little confidence these days in my own abilities).

Firstly, when looking at my atmosphere ancillary versions file (e.g. at /work/n02/n02/cjrw09/gc31/pliod/ancil_versions/GC31_UM107_atmos_mp_modant_eant_ncs_rs_buff), I can see that the ozone is linked to a particular file ($UM_ANCIL_N96EDIR/ozone/sparc/1994-2005/v3). This is a zonal mean. However, looking at my suite (u-ec126, at um > namelist > Reconfiguration and ancillary control > Configure ancils…), there is another ozone file hardwired into it, rather than using the variable names like elsewhere. This file is 3D.

So my first question: am I right in thinking if something is hardwired into the suite at the above list, this overrides whatever is in the ancillary versions file? So we are not actually using the zonal mean pointed to by the ancillary versions file, but instead the 3D file hardwired in? I fairly sure the answer is yes, but want to doublecheck.

Then my second question: in the suite there is an extra switch for ozone (i_ozone_int, at um > namelist > UM Science Settings > Section 01… > Ozone) in order to specify that the field is either 3D or 2D. Am I right in thinking that the time dimension is ignored here, so a field that is lon,lat,level,time (which is obviously 4D) is actually counted as 3D here? Because there is no 4D option. In contrast, a zonal mean (i.e. lat,level,time, which I would say is 3D) is actually counted as 2D here? Certainly this would reinforce the answer to my first question, because this switch is currently set to 3D and indeed the hardwired version it is currently pointing to (i.e. not in the ancillary versions file) is indeed 3D (or rather 4D, but I don’t think the time dimension is included).

Then my final question, assuming my answers to the other 2 are correct: presumably therefore, if I want to use a zonal mean for my ozone (which I do, because this is more appropriate for palaeo runs), I would need to change this switch to 2D? I have a definite memory that this is what I did before (many years ago), and indeed it failed because I did not change this switch, but just would like to doublecheck.

Thank you very much,

Charlie

2

Further to this…

I went ahead and tried running with my new ozone anyway, but clearly something is wrong because the reconfiguration failed with the following error:

? Error code: 52
? Error from routine: inancila_rcf_inancila
? Error message: Wrong Ozone data provided.
? Error from processor: 115
? Error number: 1

I thought at first this was because I set the option to 2D, but I have tried it as 3D as well and same error.

Are you able to help? My suite is ec186. I don’t think there is anything wrong with my actual ozone file (at /work/n02/n02/cjrw09/gc31/pliod/ancils_mp_modant_eant_ncs_rs_buff/atmos/ozone/sparc/1994-2005/qrclim.ozone_L85_O85_new) but maybe there is? Or maybe the way I have put it into the reconfiguration panel?

Thank you,

Charlie