Wow, it’s quite a long way up to the upper troposphere there - are there any previous papers on MLS data related to smoke from South America? Could be a good case to put to the pyrocb mailing list…

or could it be part of the CO tape recorder?

The points on the map are points that stick out from the usual statistical spread. They are right at the lower end of the MLS sensitivity; they are the 315 hPa level which is really not a lot of use. The excess CO is probably somewhere between 10 and 13km.