|
|
<Prev
TOC
Next >
5. Observation Monitoring
Within each month the characteristics of observations were fixed with few exceptions. The use of
observations, and the exclusion of undesirable data were controlled by means of "blacklists". For each
month of production there exists a corresponding blacklist file. From the observation diagnostics alone
(or combined with analysis monitoring results) undesirable observations were manually blacklisted for
subsequent months. The following is a list of the main observation monitoring diagnostics routinely
produced during ERA:
- Total counts of all observation types for each analysis cycle.
- Observation distribution maps for each analysis cycle.
- Reject observation distribution maps for each analysis cycle.
- Vertical profiles of observation fit to first guess, uninitialised analysis and initialised analysis
for each analysis cycle.
- Monthly time-series graphs of CCR brightness temperatures for all channels.
- Monthly time-series graphs of CCR brightness temperature - first guess for all channels.
- Once monthly maps of geographical distribution of CCR bias correction fields.
- Monthly maps of radiosonde - first guess biases.
- Vertical profiles of monthly mean and standard deviation of radiosonde biases, for bias-
corrected areas.
- Monthly maps of mean & rms errors of drifting buoys
<Prev
TOC
Next >
|
