Oxford MIPAS meeting#25
18 Feb 03

Present

Oxford: Alastair Burgess, Anu Dudhia, Don Grainger, Chiara Piccolo
RAL: Victoria Jay, Martin Juckes, Brian Kerridge, Jolyon Reburn

Instrument Status [Prev] [Next]

Envisat Orbit Control manouevre scheduled for 25 Feb, 0200-1200UTC. MIPAS unavailable during this and until several hours later.

L2 Data (CP) [Prev] [Next]

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

Data received up until 15th February

Now using a script (supplied by Herb Nett) which automates the ftp data transfer.

Software Updates (AD) [Prev] [Next]

Software in /home/crun/eodg/mipas

/Lv2PP/ (extract apodised microwindows from L1B spectra) allow any size microwindows to be defined

Auxiliary Data Updates (AD) [Prev] [Next]

New data in /home/crun/eodg/mipas

/OM/ (occupation matrices)
- Fix bug in new version of OMs to be delivered to ESA (previous version had extra rows of '0's)

Occupation Matrix Statistics (CP) [Prev] [Next]

Some unidentified OMs due to AD replacing all the current operational OMs in the /OM directory with new versions and not telling CP!
However, other unidentified OMs still detected in ESA L2 data - no explanation yet
Problem with 24 and 27 km sweeps appearing to be corrupt more often than other altitudes connected with missing Antarctic data

NO2 > pT profiles problem (CP) [Prev]

At a previous meeting, it was noted that when counting the number of profiles averaged in the ESA residuals files there sometimes appear to be more NO2 profiles than pT profiles.

Identified a specific file where this is true: MIP_NL__2PNPDK20030202_173516_000005672013_00284_04847_4329.N1
Using Enviview (or ULe/RAL L2 routines) file appears to contain a total of 7 profiles for each species
File also contains three sets of residuals, containing 6 + 1 + 6 profiles (total 13 profiles) for pT and other species, and 7 + 7 + 7 profiles for NO2
Conclusion: seems to be a more general problem counting profiles averaged in residuals
Reported to Guy Brooker (Enviview)

Missing Antarctic Data [Prev] [Next]

Very few L2 profiles of pT generated from early December until late-January, although now starting to reappear.

Now raised by ESA as an anomaly report traced to L1 processing.

OPTIMO (AB) [Prev] [Next]

Now satisfied that OPTIMO runs on Linux and Alpha machines are compatible within the numerical precision of the machines

Data Assimilation (MJ) [Prev] [Next]

Assimilating MIPAS O3 in an isentropic model and comparing with SAGE III observations Sep 21-28 2002.

SAGE III O3 agrees well with MIPAS O3 at 850K (~10mb)data, but not so well at 500K (~50mb)
At 500K SAGE III often indicates low O3 where MIPAS values are more normal (2-4ppmv)
Check whether these instances are associated with polar vortex and PSCs.

REC Retrievals (AB) [Prev] [Next]

Retrieving N2O5 from ESA L2 residual spectra using REC analysis. Aim is to try and identify clear diurnal signature in non-retrieved species as validation of REC approach

No clear signature observed between day and night residuals
Complicated by the fact that the N2O5 initial guess has latitudinal variation (difference in residual signatures at band boundaries is detected
N2O5 diurnal variability is low at low altitudes and also slowly varying throughout diurnal cycle
Suggested first step is to simply find a species for which only a fixed IG profile is used and look for latitude variability.

Retrieval Non-Linearity (CP) [Prev] [Next]

Examining the effect of non-linearity on retrieval convergence and use of singular vectors.

Plotting F(xp) - F(x0) - K(xp - x0 ), where x0 is a mid-latitude day-time profile and xp is the same profile perturbed by +1 sigma (retrieval random error SD) at all altitudes
Surprisingly large differences for pT, H2O, O3 and HNO3, even in expected linear regions (low emission/absorption)
Suggest trying perturbations at single altitude as check.

Data Access (VJ) [Prev]

RAL Attempting to obtain MIPAS L1/L2 data for 14-15 Dec for use in conjunction with SCIAMACHY nadir-only special viewing mode

No data appears to have been distributed
Unable to determine from ESA whether data has actually been processed (no response from EO Help Desk)
Note the absence of any central archive describing MIPAS operating status/data availability on a day-to-day basis

NB: BIRA (Belgium) also stores distributed MIPAS L2 data

Gain Oscillations (VP) [Prev] [Next]

Previously Roberta Mantovani had delivered a reprocessed version of L1B data for orbit 2081 where they had attempted to remove the gain oscillation thought to arise from the sweep-direction dependence of the non-linearity correction

Plot of mean A-band radiance (sweep v scan) before and after suggests that the problem (ie the diamond pattern) has largely been eliminated
Comparison of OPTIMO retrievals also show smaller, and more random oscillations with the newer data before and after
Note that the ORM retrievals were relatively unaffected by the gain oscillation, possibly due to smoothing towards the previous profile caused by the Levenberg-Marquardt parameter - this would significantly damp any component which oscillates from profile to profile.

Methyl Cyanide (CH3CN) (DG)

Methyl Cyanide is a useful indicator of biomass burning (95% source)

Plot of absorption cross-section indicates numerous spectral features covered by MIPAS bands, notably prominent feature around 1500cm^-1
Will need tabulated cross-section data to investigate retrievability

CO Retrieval (AD) [Prev] [Next]

No recent developments, just a summary of status

Previous student project (Ben Fletcher) had investigated joint retrieval of CO and Vibrational Temperature and obtained reasonable-looking retrievals for day/ngt mid-latitude from Orbit 504 data Plot
Values at 6, 9km unrealistically large, but similar problem also obtained with simulated data so thought to be a shortcoming in the (very basic) retrieval scheme rather than anything fundamental
Plot of day/night spectra (upper) used in that retrieval shows significant day-time non-LTE emission enhancement which appears to be successfully removed by the retrieval
Lower plot shows position of isotopic lines, isotopic ratios may also be retrievable
Note that limitation to 3cm^-1 microwindows usually limits microwindows to single CO (main isotope) lines - better retrievals might be obtainable if wider microwindows are allowed

Oxford MIPAS meeting#25 18 Feb 03

Oxford MIPAS meeting#25
18 Feb 03