Ray:

I think I know what happened. I found the following in the code that
wrote the data volumes (the part that gets the acd species):

c
c Some species on acd histories are in mass mix ratios (ispecpsi(ip)=1),
c   and some are in number density (ispecpsi(ip)=0) -- 
c   We need to convert the number density species to mass mixing ratios 
c   for the data volume (use rho from zatmos (rholb(jmx) is in no.dens)):
c
      do ip=1,nspec
        if (ispecpsi(ip).le.0) speclb(:,ip) = speclb(:,ip) / rholb(:) 
      enddo

where:

c                   h2o, h2, co, co2,  h, oh,  o, o3, ho2, no2, no ch4
      data ispecpsi/  1,  1,  1,   1,  1,  0,  0,  0,   0,   0,  0,  1/

that is, h2o through h, and ch4 are in mass mixing ratios in the acd
model, and the others (oh through no) are in number density. The
former ones I did not convert, and the latter ones were converted
to volume mixing ratios using rho from zatmos (Forbes). The comment
above is wrong (number densities were converted to volume ratios,
not mass ratios).

So species oh through no should be correct (ispecpsi(i)=0), and the
others are wrong (mmr) (ispecpsi(i)=1)

--Ben