Oxford MIPAS meeting#90
07 Feb 06

Present

• Oxford: Anu Dudhia, Alastair Burgess, Jane Hurley, Joanne Walker, Claire Waymark, Chiara Piccolo.
• RAL: Alison Waterfall.

Instrument Status [Prev] [Next]

The QWG reported that the number of drive errors has significantly reduced from 1 per 4 measurement orbits to 1 per 70 orbits. This is thought to be due to the decreased duty cycle (3-4/16 per day) as well as the switch-on of the interferometer heaters.

Functioning normally as far as we know, currently switched on for 4-5 orbits per day.

• Some details on the Envisat Web-Site including Instrument Availability web-site
• Latest version of Mission Planning Document covering 1 Jan 06 to 05 Mar 06: [.xls] [.html].

QWG #9 [Prev] [Next]

The main results of the meeting are covered in the [L1] and [L2] sections, below. Overall, the instrument status is improved. The L1 processor development is progressing, although no date has been given for the availability of automatically produces L1 files.

There were significant systematic NO2 differences observed between ESA and IMK. Also, the high altitude polar winter (enhanced NO2) has indications that the MW are affected by line saturation.

The presentation by Anu is here.

The presentation by Alastair is here.

The presentation by Jane is here.

The minutes are here for L2 and here for the complete set.

L1 Data [Prev] [Next]

BOMEM manually greated two versions a a L1 file. One was resampled to the reduced resolution; the other truncated. The differences were pronounced in the raw spectra, with the resampled spectra showing a strong oscillatory component.

HiRDLS Comparisons (CW) [Prev] [Next]

(Continuing to attempt to use MIPAS retrievals as input to RFM to simulate radiances in the 11/21 HIRDLS channels which are not fully overlapped by MIPAS bands.) - The simulations look OK. They are similar to measured radiances, especially T. Problem exists with simulating the long wavelength channel (possibly due to using the wrong spectral database).

TES/MIPAS/ODIN Comparisons (CP) [Prev] [Next]

MIPAS/ACE within 5K for temperature, generally better. Water withing 20-30%. Ozone variable, but good in some regions.

MIPAS/ODIN provisionally shows ozone to be systematically lower in mid profile and higher at the top and bottom of the profile, possibly reflecting lower ODIN sensitivity to ozone in these regions.

Cloud Parameters (JH) [Prev] [Next]

Looking at cloud top heights from September 2003. A Tech report is here.

Report on D-band cloud flag issues from QWG is here.

Summary of cloud band 'D' discoveries. Overall these result in a limited range of altitudes and latitudes where we can trust the D band cloud flag.

• Cases where A shows cloud and D does not
• D shows more cloud in the day than at night
• D shows cloud free when looking at scenes that are totally cloudy due to flatness of Planck function in D band
• D band more noisy

2RR Microwindows (ie 0.0625cm^-1 resln) (AD) [Prev] [Next]

pT, H2O, CH4 and N2O completed so far. Remainder in progress...

NOx (JW) [Prev] [Next]

Some N2O5 profiles are unstable, but correlated with high aerosol. Consider changing the aerosol uncertainty.

Found a [routine] to calculate sunrise / sunset times from a geolocation and date.

T validation with ECMWF (AW) [Prev] [Next]

Comparisons show good agreement at most latitudes. This is in line with results shown by M. Ridolfi at the QWG.

Also looked at O3 comparisons vs GOME data. MIPAS results degraded to the lower GOME resolution.

L2 validation (AD) [Prev] [Next]

New routine (IDL) to generate a set of MORSE-compatible .atm atmosphere profile files for the operation species from an ESA L2 file. This saves us having to repeat the retrieval.

[Presented at QWG] Investigated differences between differing versions of orbit 17540 and 14404 (see here).

• 17540 - Truncated vs resampled interferogram at the L1B level does not seem to have a huge impact on the L2 retrievals using MORSE. This is unexpected, as the L1B was shown to contain a significant oscillatory component in the resampled version. It is suggested that the effect of the apodisation before retrieval significantly reduces the effect, hence the small differences in the retrievals.
• 14404 - Manually processed 1 year apart, the only difference is supposed to be the altitudes assigned to each spectrum within the L1B file. However, there are (small) numerical differences in the spectra. The retrievals showed no difference in VMR between the two versions, when compared by sweep number. The retrieved altitudes were slightly different.

ABB Feb 06