Why are the units of fluxes, given by the Grib decoder I use, kg/m, and not kg/m/s?
What wgrib decoder, version v1.8.0.9f says (other decoders?), is that eight various fields that are water vapor and condensate fluxes, have units [kg/m]. This does not look right, because fluxes have units [kg/m/s]. The reason for the discrepancy is that variables in these fields are in fact not fluxes, transports (kg) across unit distance, per second, but are transports in 3 hours. Note the “0-3hr acc” in wgrib printout. Thus, the units kg/m are correct. At the time of this writing, we are in the process of changing wgrib to refer to those variables as transports, as opposed to fluxes.
