Good evening,
I am attempting to analyze budgets of a number of trace gases with stratospheric sinks (N2O, CH4, CFC-12) in the CESM2(WACCM6) runs archived at NSF NCAR GDEX Dataset d651045. Monthly mean amounts of these species are unsurprisingly specified as mixing ratios (e.g. mol N2O/mol dry air). However the associated monthly mean loss rates (e.g. N2O_CHML) are specified in units of /cm^3/s. I would like monthly mean loss rates with dimensions of mixing ratio per unit time, and it is tempting to estimate these by rescaling the /cm^3/s loss rates using a monthly mean dry air density computed from the pressure and monthly mean temperature.
But this isn't quite rigorous, because the time averaging and units conversion operators don't actually commute--unless, of course, the monthly mean /cm^3/s loss rates are just a relabeling of a monthly mean mixing ratio loss rate. I am trying to understand how, exactly, monthly mean _CHML with units of /cm^3/s are computed--is it
Thanks,
Todd Mooring
Associate
School of Engineering and Applied Sciences
Harvard University
I am attempting to analyze budgets of a number of trace gases with stratospheric sinks (N2O, CH4, CFC-12) in the CESM2(WACCM6) runs archived at NSF NCAR GDEX Dataset d651045. Monthly mean amounts of these species are unsurprisingly specified as mixing ratios (e.g. mol N2O/mol dry air). However the associated monthly mean loss rates (e.g. N2O_CHML) are specified in units of /cm^3/s. I would like monthly mean loss rates with dimensions of mixing ratio per unit time, and it is tempting to estimate these by rescaling the /cm^3/s loss rates using a monthly mean dry air density computed from the pressure and monthly mean temperature.
But this isn't quite rigorous, because the time averaging and units conversion operators don't actually commute--unless, of course, the monthly mean /cm^3/s loss rates are just a relabeling of a monthly mean mixing ratio loss rate. I am trying to understand how, exactly, monthly mean _CHML with units of /cm^3/s are computed--is it
- An unweighted average over all relevant instantaneous /cm^3/s loss rates
- An unweighted average over all relevant instantaneous mixing ratio loss rates, rescaled to /cm^3/s to write to files
- An inverse-density weighted average over all relevant instantaneous /cm^3/s loss rates, to facilitate correct calculation of monthly mean mixing ratio loss rates or
- Something else I haven't thought of?
Thanks,
Todd Mooring
Associate
School of Engineering and Applied Sciences
Harvard University