|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Pre-processing On receipt, IASI observations are pre-processed at the Met Office through the AAPP software. The data are thinned to one pixel in every four for each scan position, with the thinning selecting the observation believed to be the most likely to be cloud-free. Forward modelling The monitoring plots displayed on this site compare IASI observations with simulated brightness temperatures calculated using the appropriate 6hr forecast fields produced by the Met Office's global model and the RTTOV-7 fast radiative transfer model. Observations within 180 minutes either side of the analysis time are processed, and the forecast fields are interpolated to the observation time and location. These calculations are carried out in a processing stage (which includes a 1D-Var step) known as the Observation Processing System (OPS) Channel selection 314 IASI channels, 15 AMSU-A and 5 MHS channels are processed together in OPS. The IASI channels were selected for use in NWP using the method described in Collard and Matricardi 2005 (Proceedings of ITWG14, Beijing). In the 314 channel set, 30 channels were selected for detailed monitoring:
From this 314 channel set, most channels are used in the 1D-Var processing step except for the ozone band, and Band 3 channels for day time observations. Only clear observations over sea are currently processed. Cloud detection Cloud detection is done currently on a per-field-of-view basis rather than a per-channel basis. Five channels are used in the cloud detection scheme which is based on the ATOVS scheme of English et al. 1999 (QJRMS 125 2359—2378.)
Bias correction All observations are bias corrected before processing in 1D-Var. The correction currently consists of a different constant offset for each channel and each scan angle plus a linear function of the 850-300 hPa thickness and 200-50 hPa thickness, with different coefficients for each channel. The coefficients are static and updated periodically as required depending on model or data assimilation changes and any drifts detected with the monitoring plots. Land/sea/ice mask A surface type is assigned to observations during pre-processing in AAPP based on a 1/6° dataset. This is compared to the global model land/sea mask and an estimate of the surface type derived from AMSU data, and where there is a mismatch, the most conservative channel combination is used (in the case of IASI, this means rejection). Sea Ice is identified using the model sea ice field, with the AMSU surface type used to identify likly mismatches as with the land/sea definition.Quality control Not all of the observations included in the monitoring plots are actually processed in 4D-Var. All observations with a valid background calculation are included in the monitoring plots (or all observations with a valid retrieval). For data assimilation purposes, observations are rejected for many reasons, ranging from the profile going out of the bounds of validity of the forward model or failure to converge on a solution, to the retrieved profile not adequately fitting the observation. Thinning Observations are subject to thinning during OPS before processing in 4D-Var, but the plots on this site are of the data before these thinning steps are applied. Aggregation of data for monitoring purposes The data are aggregated into 5°x15° lat-lon boxes over land, sea and sea-ice separately, and for night-clear, night-cloudy, day-clear and day-cloudy situations. The scan position information is aggregated into 5° latitude bands and for each surface type separately, combining all clear observations. No information on individual observations is retained in these monitoring files, and there is no information available on rejection statistics. Initally the stats will only update once per day, i.e., will be four cycles combined. It is possible this will be increased to four times per day in the future. The global map plots are produced once per week, for a week's worth of aggregated data. |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
This section defines some of the parameters which are displayed in the monitoring plots.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The following table gives details of the currently available monitoring plots which are produced on a daily/weekly basis for IASI data quality monitoring. N.B. Your browser must support javascript and javascript must be enabled in order to use the interface. Which statistics are plotted? In most cases the mean or standard deviation plots will display with a number of observations plot below. Number of observations plots are not displayed on their own and are not available from the drop-down menu. In the case of time series and scan position plots, mean and standard deviation are displayed together on the same plot. Scan position plots The four IASI pixels for each of the 30 scan angles are combined together. Corrected - Retrieved statistics are not available for scan position plots.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Please see the list of recent occurences to see known data outages and processing changes. |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||