This is a brief description of the format of the adjustments file produced
during the updates. It is updated periodically when the dataset is updated.
It provides the results of our adjustments procedure from all updates (for
further details of this procedure see the report on the updating of HadAT in
near real-time). Each line has one of two ASCII code formats:
An adjustment - '(I8,I7,I5,I8,I8,I8,X,13(F8.3),I6)'
OR
A deletion - '(I8,I7,I5,I8,A8,A8,I4,A8)'
In column order these variables are:
For both cases:
1. The last time in the series at the time the adjustment/deletion was made (YYYYMM)
2. WMO station identifier
3. Pressure level (hPa)
4. Time assigned to the break point (YYYYMM)
In the case of an adjustment:
5. Last point in time series used to create the adjustment factor estimate (YYYYMM).
6. First point in time series used to create the adjustment factor estimate (YYYYMM).
7. Median estimate of all adjustment factors calculated - this is the adjustment applied.
8. Mean estimate of all adjustment factors calculated.
9-19. percentiles of the adjustment factors calculated under our approach:
(1,2.5,5,10,25,50,75,90,95,97.5, and 99)
20. The number of months either side of the breakpoint used to calculate the
adjustment (Note: a shorter period may have been used on one or both sides if
another breakpoint was found within this period)
In the case of a deletion:
5. Empty
6. Empty
7. Missing data identifier of -999.
8. Deletion type. This is one of three types:
manual - manual deletion at the time specified.
auto - automatic deletion at the time specified. These are made at points
that exceed a previously determined critcal value (see the report
on the updating of HadAT in near real-time)
onward - manual deletion of all points after and including the deletion
time specified.
Holly Coleman 02/08/05