we require help with generating the nitrogen deposition ancillary file for our future Scenario SSP3-7.0 with ACCESS-ESM1.5. I turns out we used the SSP2-4.5 nitrogen deposition for those runs in the past and we don’t actually have the correct SSP3-7.0 data set.
The nitrogen deposition fields are available from input4MIPS and the UKMO has a script to get this in the right format for the UM. This has also been briefly discussed with @MartinDix and Martin already provided some suggestions on how to get this done.
Thank you.
Cheers,
Tilo
MartinDix
(Martin Dix ACCESS-NRI Associate Director for Model Development)
2
Sorry I have starting looking at it more in details only today.
However, I already have almost all the steps sorted out.
I need to check that the script I wrote is effectively producing what you need, and I will then send you the path to the Nitrogen deposition ancillary file for SSP370 (together with the script used to create it).
I had a look at the file and I think the numbers are too small. From your script it looks like you are only including the “drynhx” component. I think there are four components: dry and wet nitrate (drynoy and wetnoy) and dry and wet ammonium (drynhx and wetnhx) that go into the nitrogen deposition field. This should all be all available from input4mips.
Also, the model requires an additional year of data at either end (for interpolation), which means we need the nitrogen deposition data set from 2014 to 2101 to run our simulation from 2015 to 2100. The 2014 data can be obtained from the last year of the historical period and the 2100/2101 I think are just a repeat of the last year available.
In the ancillary file you provided (/scratch/public/dm5220/tilo/ndep_2015_2099_ssp370_esm1.5.anc) there is only one field (STASH 884) for the nitrogen deposition.
Where are the other fields you mention?
For the values being too small, they are sensible with the source file I used.
I calculated the units for the Nitrogen deposition ancillary to be gN m-2 d-1/, while the source file has kgN m-2 s-1.
Do you think that makes sense?
Do you think the source file is the correct one?
I did not know about it.
In that case, where is the historical source file (or SSP370 ancil file) from where I can take year 2014?
Also, which website/search did you do to get the outputs of the image you shared?
We only provide one nitrogen deposition field (STASH 884) in our model which I think is the sum of the four components (drynoy, wetnoy, drynhx, wetnhx). The units of g/m2/d are correct. You can download the individual components here:
I see now.
I will download the source files from the website you pointed out and try producing the field as sum of the four components.
I will also prepend the last year from the historical run and then duplicate the final year so the new ancil file will span from 2014 to 2101 (end of 2100 techically).
I should be able to provide you with the updated files within the end of the week.
Please find the modified version of the file and script in the same directory: /scratch/public/dm5220/tilo
The updated gist is also available at the same link:
In the final ancil data:
Year 2014 has been taken from the historical data.
Historical data’s last year is actually 2015, not 2014.
However, year 2015 is present in the source files.
Therefore, in the final data, year 2014 has been taken from the historical data’s second-last year, whereas year 2015 comes from the source files;
Years from 2015 to 2099 have been computed from the source files as the sum of the 4 fields you indicated: wetnoy, drynoy, wetnhx and drynhx;
Years 2100 and 2101 have been computed by repeating year 2099 twice (since the source data ends at the end of 2099).
Please let me know if the processing done for the different years make sense to you, and if the new ndep ancil file has sensible values.
the data is looking good and in line with the other scenarios. I have just set-up a new run for testing and it seems to run fine with the new ndep ancillary file.
