Zonal shift in N-deposition 1850 climatology

[thiasB]

Dataset source ID(s)

FZJ-CMIP-nitrogen-1-2

Describe the issue

All four N-deposition variables from the monC climatology show a fairly strong zonal shift compared to the monthly data from 1850 from the historical time series.

Expected data

The pattern should differ slightly, as we are comparing climatology with values from a single year. However, they should generally correspond to the spatial distribution.

Screenshots

This is 1850 mon minus monC. Negative values in blue, positive is red.

Additional context

Using the monC data is recommended for for pre-industrial control runs.

[thiasB]

Tagging @hegglinm here as well

[vnaik60]

ping @hegglinm and @ccmi1-test - your thoughts on this would be helpful.

[hegglinm]

we are looking into the problem but have not found the underlying issue as of yet. we will try to find a solution by the end of this week (hopefully with an updated historical dataset). @vnaik60 need to get this out to the modellers, can they wait to run this simulation? how many modellers are using the N-dep dataset?

[thiasB]

Upon closer examination of the location of the prevailing patterns, e.g., over East Asia and sub-Saharan Africa, it seems to me that the historical(1850) data is accurate, while climatology shows a shift to the north. Could it be that something went not as expected when calculating the temporal mean for the climatology?