For hackathon effort 11 Jul 2025 ICCS summer school, relevant to UKESM evaluation next week
The UK Met Office earth system model (UKESM) will be one of many models used to predict future Earth behaviour for the IPCC Assessment Report 7.
There is a hackathon at Reading to evaluate the UKESM 1.3 output to discuss its output quality currently and what variant should be taken forward for this purpose. To decide that, we need to compare the model predictions against past observations of the real atmosphere.
Today we have some model output from UKESM 1.3 (model run
u-dp226) in NetCDF format, gridded as a function of
(time, latitude, longitude, model level). This data is supplied
separately as it is too large for GitHub.
We also have some observational datasets which some more detailed notes about will be added below.
Final note: we only got as far as converting ozone mass mixing ratio to Dobson Units for comparison with two of the satellite datasets. See also 2025_07_11_ICCS_UKESM_eval_HackathonPresentation.odp for final group presentation.
From ncdump -h this dataset has:
dimensions:
time = 456 ;
latitude = 36 ;
level = 70 ;
...
o3_mean:units = "mole mole^-1" ;
The UKESM output is in mass mixing ratio, so a conversion is neeeded to molar mixing ratio to compare to this. Also the levels will not be the same -- we could average or sum over levels initially to get something comparable.
From ncdump -h this has:
dimensions:
latitude = 180 ;
longitude = 288 ;
time = UNLIMITED ; // (12 currently)
...
tco:units = "DU" ;
tco:long_name = "Total column ozone" ;
DU means Dobson Units, an absolute measure of the total quantity in the column per unit area (from ground level to space).
This has:
dimensions:
longitude = 288 ;
latitude = 120 ;
t = 183 ;
...
ozone_column:units = "DU" ;
So again we will have to convert from mass mixing ratio in the model output to compare with this dataset.