Dear Susana, Cc to qpack mailing list as this is of general interest, and we prefer if all questions go to any of the arts and qpack lists.
The setting of angles with antenna is confusing. Sorry about that, but could not be avoided with less of accepting calculation overhead for some cases. I am not sure what you want to achieve with your settings. Anyhow, these settings result in multiple spectra explaining the later error for Sx. But I assume that you invert one spectrum at the time. Then you should have something like this Q.SENSOR_RESPONSE.ANTENNA_LOS = 0; Q.MBLOCK_ZA_GRID = linspace(-Apat(1,1),Apat(end,1),10)'; The important part is that MBLOCK_ZA_GRID covers all angles in Apat(:,1). These are "relative angles" exactly as in Apat (that is, the angles in Apat should be centered at zero). ANTENNA_LOS is just of concern if you invert "scans", i.e. multiple spectra from different angles. Otherwise it should be zero. The zenith angle of each spectrum you invert is set by Y.ZA as usual. If you want to invert a "scan", get back to me. Correction, the qpack list! Regards, Patrick On 2014-11-25 18:00, [email protected] wrote: > Dear Patrick, > > I'm working on the retrieval of ozone profiles from the 110 GHz line > measured with our radiometer Gromos-C. So far I had always worked with a > pencil beam but I'd like to implement the antenna pattern. I use a > gaussian beam of 5° (here Apat) and set the following fields: > > > Q.ANTENNA_DIM = 1; > Q.SENSOR_RESPONSE.ANTENNA_RESPONSE.name = 'Antenna response'; > Q.SENSOR_RESPONSE.ANTENNA_RESPONSE.gridnames = > {'Polarisation','Frequency','Zenith angle','Azimut angle'}; > Q.SENSOR_RESPONSE.ANTENNA_RESPONSE.grids = > {{'1'},110.836e9,Apat(:,1),0}; > Q.SENSOR_RESPONSE.ANTENNA_RESPONSE.dataname = 'Response'; > Q.SENSOR_RESPONSE.ANTENNA_RESPONSE.data(1,1,:,1) = Apat(:,2); > Q.SENSOR_RESPONSE.ANTENNA_DO = 1; > Q.SENSOR_RESPONSE.ANTENNA_LOS = [50:5:70]'; > Q.MBLOCK_ZA_GRID = [40:2:80]'; > Q.MBLOCK_AA_GRID = {}; > > Q.SENSOR_LOS = Y.ZA; > Q.SENSOR_POS = Y.Z_PLATFORM; > > > But I have encountered the following error when running the retrievals: > > > Error using arts_sx (line 85) > Wrong size of Q.POLYFIT.SX0. > > Error in qpack2 (line 100) > Sx = arts_sx( Qm, R ); > > > For the polyfit I'm using a polynomial of degree 2, and SX0 Sx1 and SX2 > are just 3 scalars. > Why is there a problem with the size of SX0? I never had this problem > when using a pencil beam. > Any suggestion? > > > Thank you in advance for your help. > > Best regards > > Susana Fernandez > > > > _______________________________________________ qpack mailing list [email protected] https://www.sat.ltu.se/mailman/listinfo/qpack
