- This topic has 2 replies, 2 voices, and was last updated 5 years, 3 months ago by
.
-
Posts
-
I have two questions about the quality flag provided by RTTOV (radiance%quality(:))
The dimension of radiance%quality is nchanprof = nchannels * nprof. So, for one atmospheric profile, we have one quality value for each radiance simulated by RTTOV.
1- For example, if for one level and one profile, the ozone mixing ratio exceeded the maximum value of the RTTOV validity range for ozone (Regression limits), the whole quality array will be filled by a value other than 0 or only the channels affected by the ozone mixing ratio ?
2- Why the quality array is given per channel ? not per profile ?
Thank you for your answer.
Nicolas.
Dear Nicolas,
To answer your questions:
1. If the ozone value exceeds the regression limit for a given profile, then the regression limit flag will be set for all channels simulated for that profile. This is true for any case where one or more of the regression limits are exceeded. Currently RTTOV does not make a distinction when setting these flags for channels which are affected, for example, by ozone absorption vs channels which are unaffected.
2. The quality flag array is used to indicate various potential issues with simulated radiances. Currently, in addition to informing you that regression limits were exceeded, it is used by the MFASIS fast visible cloud parameterisation: one of those MFASIS-related flags indicates that the optical depths have exceeded the range of values used in training the MFASIS look up tables for a given channel. In this case, this is a per-channel flag, not a per-profile flag, since optical depths vary between channels.
More generally, all RTTOV outputs are per-channel so it is helpful to maintain this for consistency. It also gives us scope to introduce other per-channel flags in the future if the need arises.
Best wishes,
James Hocking
Thank you for your answers.
Nicolas.
- You must be logged in to reply to this topic.