ACCESS-CM2/AM2 DMS Ancillary file error

Earth System Coupled Model
, , , ,
1

Hello,

I have recently been looking at the aerosol fields in ACCESS-AM2. Following some interesting results I was digging in to the DMS ancillary files and have found that the ancillary file that the model is pointing to is incorrect. It should be the Lana et al. (2011) DMS surface water climatology, but it appears to be the Kettle et al. (1999) surface water climatology. This should have been updated with GA7.1. I have attached a slide which is showing what I have found… DMS_ancils.pdf (369.4 KB)

So I guess my questions are:

  • can we confirm that this is the case (I’m pretty sure so!)?
  • are we sure that there are no other ancillary files that are outdated?

I assume that this will not now be corrected for ACCESS-AM2 or CM2, but is good to be aware of. I will do another run now to see what the impact is for my work at least. I have the Lana DMS climatology on an N96 grid already (from my PhD), I can put it in p66 if needed?

As an aside, there is a new climatology available (hence the reason I was looking) and it might be good to consider for future releases - I’m working on looking at its performance now.

Let me know if you want me to go into more detail/provide the updated ancillary file!
Cheers,
Sonya

4 Likes
2

The HadGEM ancillary is documented at CMIP6/ForcingData/ModelChloroDms – ANCIL (metoffice.gov.uk) if that is helpful

3

Thanks Scott but I think the HadGEM model uses a parameterised DMS from MEDUSA, so not a climatology which is standard when no BGC model (with sulfur cycle) available :slight_smile: The Walters et al. 2019 paper clearly states it should be the Lana et al. (2011) climatology.

1 Like
4

Thanks Sonya,

Looks like we missed this. I’ll go back and check what else changed with 7.1.

Have you done an AMIP run with this fixed?

5

Hi Martin, thanks for checking! I have just completed a short (6 years) AMIP control run (though nudged) with this fixed, plus a few other experimental runs I was working on. Details here: Experiment Proposal: Fixing DMS Ancillary files in ACCESS-AM2 runs