Pressure Retrieval (AD/IY)
[Prev]
[Next]
Undergraduate vacation project
to investigate whether it is possible to
directly retrieve tangent pressure from L1B spectra by simply taking the ratio
of the average radiances in two specifically chosen microwindows
(analagous to the cloud-detection).
- Microwindow pairs now selected for each nominal
tangent altitude and quadratic
expressions derived for pressure as a function of ratio for each pair
- Algorithm now incorporated into Lv2PP (Oxford Lv2 Preprocessor)
ready for comparison with retrieved tangent pressures.
Quality Working Group Meeting#4 (Paris, 19/20 Jul 04)
[Prev]
[Next]
Some notes
- Introduction (Rob Koopman)
- MIPAS operations may start as soon as the beginning of August - it is
desirable to have MIPAS operating in time to observe the Antarctic Ozone
Hole.
- An ad-hoc MIPAS "Science Team" (replaces Science Advisory Group)
will be convened in ESRIN, possibly October 6th (following QWG#5)
- Mission Planning (Marta de Laurentis)
- Operations will initially resume with 146 scans per orbit at 0.0625cm-1
resolution (ie approx twice as many profiles as operations at
0.025cm-1), but with no orbital correction to align scans with latitude
(although approximate repeat cycle is 3 orbits)
- Future nominal and special modes to be defined by the Science Team.
- Instrument Performance (Roland Gessner)
[Prev]
- Main cause for concern is the chance that mirror#2 may get stuck in
"Launch Lock" position after restart whenever the operation is
automatically stopped due to slide speed anomalies
- Operating just mirror#1 (`Backup Mode') is not currently allowed due
to possible vibrations introduced to satellite
- No estimate of lifetime is possible
- L1B Spectra (Gaetan Perron)
[Prev]
- NESR will continue to be represented every 400 sample intervals,
which now means one value every 25cm-1 instead of every 10cm-1.
- Fwd/Rev Oscillations (Anne Kleinert)
[Prev]
- Difference of radiance from each scan with average of adjacent scans.
No obvious improvement with v4.61 compared to v4.55 (orbit 2975)
- Non-linearity correction (Manfred Birk)
- Non-linearity correction in A and B detectors gives a radiance
uncertainty of 3%
- Pointing Covariance (Marco Ridolfi)
- Using Line-of-sight calibration viewing stars to verify assumptions
on absolute and relative intersweep pointing stability - seems to be
better than specified/assumed in pT a priori covariance
- Jitter Ghosts (Frank Hase)
[Prev]
- Analysis of empty regions of spectrum where ghosts might be expected
(+/- 13.5cm-1 from spectral lines) - no evidence found
- Spectral Shift v Tangent Height (Frank Hase)
- Examine possible cause due to asymmetry of ILS, but this would result
in a variation in shift proportional to wavenumber which is not evident.
- Wind Retrievals (Claude Camy-Peyret)
- Started investigation of whether line-of-sight wind speed can be
detected in MIPAS
spectra - signature would be 3 orders of magnitude smaller than the
0.025cm-1 sampling
- Approach is to use discrete Fourier Transforms to average signals from
regions of uniformly spaced spectral lines, and look for signature
in Fourier space rather than spectral space
- Current Anomalies (Roberta Mantovani)
- Level 1
- Level 2
- Exactly 1 sec discrepancy between NRT and OFL profiles
(possibly due to leap seconds?)
[Prev]
[Next]
- Anomalous values in PCD (Product Confidence Data)
- O3 pT error propagation covariance matrix
- Occurrence of one L2 file with 22 (rather than 17) sweeps per scan
- connected with transition to IF9 measurements (which have 22).
- ORM Tests (Simone Ceccherini)
- Conducting tests of preliminary delivery of pT and H2O microwindows
for reduced resolution operation using original L1B spectrum with
degraded resolution [Prev]
- Chi-square values for pT comparable to high resolution but H2O
about a factor 3 larger, and both differ by more than the random
errors compared to original retrievals
- H2O microwindows 201 and 204 cause particular problems
- Processing is approx 2x slower per scan, and with twice as many
scans per orbit, NRT processing options are to reduce number of
species or to process one orbit in two (however, CPU is less of a
consideration for off-line processing).
- University of Leicester Work (John Remedios and Alison Waterfall)
- IG2 (Initial Guess climatology) CO2 will be updated for for current
levels (368ppmv instead of 363ppmv - a 1.4% increase)
- IG2 Temperature and pressure will be updated to use CIRA (which
should remove the anomalously large polar spring and autumn
temperature differences compared to actual retrieved temperature)
- Still unexplained discrepancies between clouds as detected in
v4.61 processing and by their own analyses of L1B spectra.
- Non-LTE Effects (Manuel Lopez-Puertas & Mariliza Koukouli)
- Retrieving H2O at high altitude from UA modes results in problems
with Jacobians in mesosphere - possibly related to the
H2O problem in nominal mode data thought to arise from the
combination of Doppler-broadened lines and increasing temperature
(therefore broader lines) with altitude.
- (Corrections from MLP to my original notes) NLTE enhancement in the NO2(v3) hot bands is about a factor 50
smaller. Hence NLTE in v3 HOT bands and potential contamination to H2O
lines is much less important.
- However, NLTE in the v3 fundamental band, which is used in the NO2
vmr retrieval, is affected by NLTE in a SIMILAR magnitude than before
but with opposite sign. The figure I presented shows an underestimation
of NO2, if assumed in LTE, ranging from 10 to 25% at 45-60 km. The
reason is that chemical excitation does NOT compensate now for the
spontaneous emission and hence we get populations for 001 smaller than
in LTE.
- Publications
- Send all publications using MIPAS data (post launch) to IFAC for
their website
