L2 validation (AW/AD)
[Prev]
[Next]
More general validation issues
- It has been decreed that only MIPAS v4.61 should be used for validation,
although this seems to be based on a single example where v4.61 and v4.62
data disagreed with v4.61 looking more plausible.
- According to Thorsten Fehr (ESA) there should be no difference between
v4.61 and v4.62 processors, although presumably differences may arise from
changes in input data, eg calibration.
- It is now known that L2 profile values are limited to 1E-10: 1E+10
[Prev] when
the retrieval produces results outside this range, but there does not yet
seem to be any clear policy for the validation papers whether these values
should be included/excluded in the profile comparisons.
- Marco Ridolfi (Bologna) has sent the following regarding flagging
data quality
The current status of MIPAS Level 2 products flagging is the following:
- cloudy limb-views are filtered-out before the
Level 2 analysis, therefore in the ESA Level 2
files you can find only cloud-free profile data
points.
- It is in the ESA plans to include also the cloud-filtering
flag in the output files but so far there is no
cloud flag in the MIPAS Level 2 files.
- MIPAS Level 2 products come with two flags:
- quality_flag: set to -1 if the retreival failed
set to 0 otherwise.
- conv_id:
= 0 convergence reached
= 1 too many macro-iterations
= 2 too many micro-iterations
= 3 allowed run-time exceeded
= 4 failed (e.g. division by zero...)
- The validation activity should be limited to the profiles
with quality_flag = 0 and conv_id = 0.
Atmospheric Science Conference 2006
[Prev]
http://earth.esa.int/atmos2006
To be held at ESRIN 18-12 May 2006
Poster abstracts on the following topics have been submitted from
the Oxford group.
- Measurements of CF4 (CP)
- Enhanced HNO3 in the 2005 Arctic Stratosphere (CW)
- Retrieval of Cloud Top Height (JH)
- Trends in CFCs (AB)
- Measurements of NOx (JW)