Hi Patrick, Thanks for your explanation, I am more confident now.
Best regards, Pauline ----- Météo-France ----- Dr. Pauline Martinet Chercheur CNRM/GMEI/LISA [email protected] Fixe : +33 561079031 Site web: www.sites.google.com/site/martinetpauline31 ----- Mail original ----- De: "Patrick Eriksson" <[email protected]> À: "Pauline Martinet" <[email protected]>, [email protected] Envoyé: Mercredi 14 Janvier 2015 19:06:59 Objet: Re: [Qpack] humidity jacobians Hi Pauline, At 22.24 GHz you are very close to the centre frequency of the transition and the absorption at this frequency is significant also in the stratosphere. So don't think anything is wrong here. The end point includes the contribution all the way to the end of the model atmosphere, which explains why get a "peak" when moving down the end point of the retrieval grid. Note also that the shape of the Jacobians depends strongly on the unit you have selected. It seems that you are using VMR. IF you switch to 'rel', the stratospheric values will be much smaller compared to the ones in the troposphere. Bye, Patrick On 2015-01-14 17:27, Pauline Martinet wrote: > Hi everyone, > > I am using qpack to retrieve humidity profile from microwave > observations (from 22.24 GHz to 31.40 GHz). > > My vertical grid goes from 1009 hPa to 0.6 hPa in 60 levels. > > I am surprised about the shape of the humidity jacobians for the first > channel at 22.24 GHz that gives very large values at the top of the > atmosphere > where I would expect the jacobians to be almost zero. (see attached > figure humidity_jacobians_60levels.png) > > I tried to used two different absorption models (PWR98 and MPM93) but I > have the same problem. > > To avoid this large value in the stratosphere, I also tried to run my > retrievals on a smaller grid only between level 1 and level 45 (~280 hPa) > and now I have even larger values at 280hPa instead of 0.6 hPa (see > attached figure humidity_jacobians_60levels.png). > > Does anyone know why I get such large values for the humidity jacobians ? > > Maybe there is something wrong when I define my retrieval grid: > Q.ABS_SPECIES(2).GRIDS > > Thank you very much for your help. > > Best regards, > > Pauline > > ----- Météo-France ----- > Dr. Pauline Martinet > Chercheur CNRM/GMEI/LISA > [email protected] > Fixe : +33 561079031 > Site web: www.sites.google.com/site/martinetpauline31 > <https://www.sites.google.com/site/martinetpauline31/> > > > > _______________________________________________ > qpack mailing list > [email protected] > https://www.sat.ltu.se/mailman/listinfo/qpack > _______________________________________________ qpack mailing list [email protected] https://www.sat.ltu.se/mailman/listinfo/qpack
