Oxford MIPAS meeting#48
06 Feb 04

Present

Oxford: Alastair Burgess, Anu Dudhia, Don Grainger, Chiara Piccolo, Analia Quijano
RAL: Victoria Jay

Instrument Status [Prev] [Next] (Some details on the Envisat Web-Site including Instrument Availability web-site)

Off since 2nd Feb, undergoing passive decontamination, and due back in operation on 13th Feb.

L1 Data (VJ) [Prev] [Next]

RAL are now also looking for L1B data from the off-line processor. Contact with IMK suggests that this is a non-trivial task.

L2 NRT Data [Prev] [Next]

In /home/crun/mipas/L2/NL/ [Plots of profile locations]

NRT Data received up until 2nd Feb, plotted up until 29th January. Note that a map is generated for each day where there is a L2 file received, but only locations of valid pT profiles are plotted.
Software version 4.59 in use since 23 Jul 2003.
Although L2 data is being received there have been no valid profiles since 11 Jan. (See 30km daily mean traces).
NRT L2 files just contain message
```
Failed to select an occupation matrix for scan [i] Retrieval [j]
```
Presumably associated with L1 spectra being flagged as unusable
See email from Rob Koopman

L2 OFL Data (AQ) [Prev] [Next]

In /home/crun/mipas/L2/2PODPA/

L2 OFL (off-line) data now received covering 21 Oct - 24 Dec
Profile location plots] now show availability of both NRT and OFL data
No further OFL data received since 20 Jan 04.

NH3 Retrievals (AB) [Prev] [Next]

Using local retrieval code OPTIMO to retrieve NH3 from MIPAS spectra.

Currently investigating influence of a priori on NH3 retrieval, (which appears to be disturbingly large).

O3 Isotopic Retrievals (CP) [Prev] [Next]

Microwindows and error analyses derived for isotopes treated as independent gases

[Plot] of microwindow locations and Jacobian spectra
Error budgets for isotopes [666] [668] [686] [667] [676]
Apart from main isotope (666) error budget is dominated by random errors, as expected, although (surprisingly) accuracies of better than 10% appear to be achievable for all the minor isotopes, at least near the O3 maximum.

RAL Retrievals (VJ) [Prev] [Next]

Attempting to retrieve O3+H2O from S6 mode (ie 0.1cm-1 resolution, high horizontal sampling).

The plan is to use ECMWF T(p) fields,
Currently investigating methods of determining MIPAS tangent point pressures (with normal data these are available from the L2 profiles, but S6 data are not processed to L2).
First, need to try out pointing retrieval on standard data

REC Analyses (AD) [Prev] [Next]

Analyses completed for Jan 2004

The main changes from the previous month are in the non-target microwindow residuals for polar latitudes.
These may simply reflect the discrete change in the seasonal `initial guess' climatology that is introduced in January.
There has also been an improvement (ie reduction) in the derivative signatures associated with the NO2 microwindows.

Occupation Matrix Statistics (AD) [Prev]

Analyses completed for Jan 2004

No valid retrievals were received in January after the 11th
Prior to that the average number of profiles/day (532) was also below normal (~850 for Oct/Nov 2003).
OM statistics for the profiles which were successfully retrieved appear normal and the previous month's problem with corrupt AB and C band spectra appears to be resolved.

Monthly Means (AD) [Prev] [Next]

Analyses completed for Jan 2004

Main changes are in the N.Polar region, generally towards less anomalous behaviour compared to previous month
Note that the Initial Guess data is updated in January, so differences wrt to initial guess compared to the previous month mostly arise from the change in the IG rather than the data
No day-time retrievals were obtained for the north polar region, although since there were some retrievals for both day and night-time in the south polar region, symmetry suggestd there should also be both in the north polar region.

Email from Rob Koopman (ESA) 5 Feb 04

After the non-linearity calibration, MIPAS reprocessing will resume with a slightly modified configuration to adapt to the higher noise level and to avoid an infinite loop in the processor. Unfortunately the frequent level 1 processor hiccups could not be circumvented by the workaround we tested last week, as it turns out that the coding error also corrupts fringe count error values. I will be informed today on the date that the debugged operational processor will become available.

Oxford MIPAS meeting#48 06 Feb 04

Oxford MIPAS meeting#48
06 Feb 04