Added 2 new unit converstions

[siligam]

Auto-unit converstions for recom variables CO2f and pCO2s are added.
The negative sign used in "mmolC/m2/d" -> "kg m-2 s-1" is to correct the sign change for into ocean or into atm.

[chrisdane]

Before merging can you please add comments how these factors are obtained? Maybe similar like I did here: https://notes.desy.de/F8hTyk3PRielZrVdZRtyaQ?both#Seamore-missing-features-FESOM1-and-2

[pgierz]

Better yet, define and use a documented variable. I personally hate magic numbers.

[siligam]

@chrisdane Thank you for asking the question. You can find the detail explanation on how factor were derived at this link https://notes.desy.de/F8hTyk3PRielZrVdZRtyaQ?view#Seamore-missing-features-FESOM1-and-2 (please see section 1.1 and 1.2)
Additionally, there are notes on the negative sign used in this factor. Instead of combining the factor and the negative sign adjustment in auto-unit conversion section, deferring the negative sign adjustment as a separate action in on cmorization steps might be a better approach. Please let me know your thoughts on aspect: if this extra step is required or is it good enough leave things as is.

[siligam]

Oh yeah.. now I get it, to change the commit message to include you comments from the hedgedoc. will do it rightway.

[siligam]

Here is the link to updated commit which includes notes on how the factors were obtained.

seamore/lib/step.rb

Lines 218 to 239 in edde2d7

	when ["mmolC/m2/d", "kg m-2 s-1"]
	# notes on how this factor was derived (by chrisdane "Christopher Danek")
	# e.g. recom variable: CO2f => CMIP6_Omon.json: fgco2
	# 1) mmolC -> molC: /1e3
	# 2) molC -> gC: *12.0107
	# 3) gC -> kgC: /1e3
	# 4) d-1 -> s-1: /86400
	# -> mult_factor = 1/1e3*12.0107/1e3/86400 # = 1.390127e-10
	# In the CMIP6_Omon.json:fgco2 case, an additional sign change is necessary:
	# >0: into ocean -> >0: into atm: *-1
	# -> mult_factor = mult_factor*-1
	cmds << CDO_MULC_cmd.new(-1.390127e-10) # negative sign to correct into ocean or into atm
	when ["uatm", "Pa"]
	# notes on how this factor was derived (by chrisdane "Christopher Danek")
	# e.g. recom variable: pCO2s => CMIP6_Omon.json: spco2
	# 1) µatm -> atm: /1e6
	# 2) atm -> Pa: *1.01325e5 (1 atm = 1.01325e5 Pa)
	# -> mult_factor = 1/1e6*1.01325e5 # = 0.101325
	#
	# these notes are available on the following link
	# https://notes.desy.de/F8hTyk3PRielZrVdZRtyaQ?view#Seamore-missing-features-FESOM1-and-2
	cmds << CDO_MULC_cmd.new(0.101325)