Oxford MIPAS meeting#27
18 Mar 03

Present

• Oxford: Alastair Burgess, Anu Dudhia, Don Grainger, Vivienne Payne, Chiara Piccolo
• RAL: Victoria Jay, Martin Juckes

Instrument Status [Prev] [Next]

MIPAS switched between 13:26-18:05 13 Mar due to unspecified problem

Entire Envisat payload switched off 04:21 15 Mar due to unspecified problem

L2 Data (CP) [Prev] [Next]

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

• Data received up until 14th March (although locations not yet plotted)
• March data continues to show missing profiles at the equator (probably related to fringe count error)
• Problem with duplicate data probably explained by different processing versions (file extensions .N1 and .N1.1 on ESA data)
• Problem reading ESRIN L2 data (...PDE... files), at end of February, but successfully read from 5th Mar onwards, but now problems reading Kiruna L2 data (...PDK... files).
• ASCII profiles extracted for 4 days in September and sent to Jo Urban for ODIN intercomparison.

ESA Processing [Prev] [Next]

Earlier request from Rob Koopman to the MIPAS-QWG to specify particular orbits for processing activities

• RAL requested Orbits 4125 to 4141 on 14th and 15th Dec (priority ones for which we already have SCIA data are: 4125, 4126,4127,4139, 4140, 4141)

Data Assimilation (MJ) [Prev] [Next]

Assimilating MIPAS O3 in an isentropic model

• Now included SAGE III data for January
• Results for 500 K and 850 K surfaces
  • The upper plot is a scatter plot of SAGE III against an assimilation of MIPAS, showing ozone ppmv. Red is northern hemisphere, blue southern. Time period is January 1-30, 2003. The size of the crosses represents the estimated observational error in the SAGE III data.
  • The lower plot shows histograms of analysis minus observation normalised by estimated observational error. The dark line is for the MIPAS observations used in the analysis, the blue line is for SAGE III observations which were not used in the analysis.
• Still waiting for GOME O3 profile data to be processed at RAL

SO2 Retrievals (AB) [Prev] [Next]

Retrieving SO2 from MIPAS spectra for different days, around a complete orbit and in an area concentrated near Mt Etna

Trying to eliminate influence of a priori

• Persistent high altitude peak (>30km) - due to a priori?
• `Flat' a priori profile still produces same low altitude features.

Retrieval Non-Linearity (CP) [Prev]

Investigating if Jacobians change significantly between nominal retrieval and retrieval + 1sigma

• Previous results indicated large non-linearity, but now corrected: H2O seems very linear
• Plots for MWs [H2O_0001] [H2O_0002] [H2O_0007], [Error Patterns] and [Tabulated Results]
• A few more tests to try but it looks like retrieval non-linearity will not be an issue for data assimilation

Averaging Kernels (CP)

Starting to look at modifying the ORM to calculate AKs for cross-terms such as dT/dp, dO3/dT etc.

RAL Retrievals (Reported by VJ) [Prev] [Next]

Intending to combine MIPAS and SCIAMACHY measurements, but in the absence of SCIAMACHY data working with GOME measurements.

MIPAS-STR (VP) [Next]

Aircraft version of MIPAS being flown on the Geophysika from Kiruna as part of the Envisat validation campaign.

4 flights, plus 1 test flight performed in previous 2 weeks.

REC Analysis (AD) [Prev] [Next]

Performing REC analysis on nominal OMs - 6 latitude bands x 7 months (Aug-Feb) (For MIPAS Cal/Val report)

• Some results: [NO2] (divided by Day/Night) [NH3] [CCl4] [Non-LTE] (divided by Day/Night)
• Shift between Aug-Oct and Nov-Feb results probably related to Pitch Adjustment on 13 Nov (see email from Herb Nett)
• Before 13 Nov MIPAS had been pointing ~1.5km higher than thought, suggesting that, for example, the contaminant profiles VMR assigned to 12km was actually being applied to an atmosphere at 13.5 km
• Therefore expect profiles of species which decrease with altitude to have been overestimated, implying negative fit and becoming more positive from Nov onwards (and vice versa for profiles which decrease with altitude)
• Seems to be true for NH3(dv/dz -ve) and NO2(dv/dz +ve), but opposite for CCl4 (dv/dz -ve)

Date: Mon, 17 Mar 2003 10:15:43 +0100 (MET)
From: Herbert Nett 
To: dudhia@atm.ox.ac.uk
Subject: Re: Pitch Adjustment

hi Anu,

...
> Hi Herb, I'm trying to make sense of a discontinuity I see in the
> residuals of contaminant profiles up to October and from November
> onwards. I suspect these are related to the ~1.5km Pitch Adjustment on Nov
> 13th.
> 
> Can you tell me the sign of the adjustment. Ie, before the adjustment when
> we thought MIPAS was commanded to look at 12km tangent height, was it
> really looking at 10.5km or at 13.5km?

13.5 km
I realised that my e-mail of 12 Nov. 2002 (below) may not have been totally 
clear. Before 13 Nov we were assigning a too low tangent height in L1B 
products.

Note that at the same time with the LOS correction on ground we changed
also the commanding of elevation angles. This means we command a 1.5 km lower
tangent hight, such that, after correction on ground we arrive at exactly 
12 km.


Note also:

1. there is a harmonic component (amplitude = 7 mdeg ~ 420 m at tangent point) 
   in elev. correction

2. on 1 Dec, 0h UTC we switched climatology database (IG2) file
   MIP_IG2_AXVIEC20021119_141033_20021201_000000_20071201_000000


hope this helps
cheers

Herbert

...
> 1. IPF/Level 1B:
> ----------------
>    aux file:          MIP_CL1_AX (line-of-sight calibration data)
>    start of validity: 13-NOV-2002 12:00 UTC
>    Correction:        LOS pointing angles (elevation) 
>                       bias (pitch)           = +25 mdeg
>                       1. harmonic: amplitude = 7 mdeg    
>                                    phase     = +99 deg  (wrt ANX)
>    previous settings: bias & 1. harm. amplitude set to zero
>    
>    Remarks:
>    1. correction based on in-flight mispointing characterisation (IR
>       star acquisition)
>    2. activation of change in IPF done simultanously with correction 
>       in MIPAS commanded elevation pointing angles ('elevation start 
>       angle correction table')
>       Effect: no change to reported tangent heights in L1B products;
>               acquired tangent points will be corrected towards higher
>               altitudes by ~ +1.1 km ... +1.9 km, depending on orbit 
>               position
> 
>    (for details see also L1B ATBD, po-tn-bom-gs-0012, section 4.7.4)
>