Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://xmm.vilspa.esa.es/docs/documents/CAL-SRN-0274-1-1.ps.gz Äàòà èçìåíåíèÿ: Thu May 12 22:35:03 2011 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 23:00:22 2012 Êîäèðîâêà:

XMM­Newton CCF Release Note
XMM­CCF­REL­274
OM Photometry. Time dependent sensitivity degradation
correction: updating the coe#cients
A. Talavera
May 12, 2011
1 CCF components
Name of CCF VALDATE List of Blocks
changed
CAL VERSION XSCS flag
OM PHOTTONAT 0005 2000­01­01T00:00:00 DEGRADATION No
2 Changes
The table extension ``DEGRADATION'' was introduced in 2006 to contain the coe#cients of the
time dependent sensitivity degradation correction. This correction is defined as
Correction factor = A +B âMJD (1)
Corrected rate = Measured rate â Correction factor (2)
Where MJD is the Modified Julian Date of observation and A and B depend of the filter. Their
current values are given in Table 1.
Since time dependent sensitivity variation is due in part to sensitivity degradation of the pho­
tocathode, it is wavelength dependent and therefore it is di#erent in each of the OM lenticular
filters.
The correction is based in measurements of the count rates of three spectrophotometric standard
stars, BPM 16274, HZ 2 and GD 153, which are observed regularly with OM in all filters. As more
1

XMM­Newton CCF Release XMM­CCF­REL­274 Page: 2
Table 1: OM Time sensitivity degradation correction
filter A B new A new B
UVW2 ­ 2.72723 7.14543e­05 ­1.1448062 4.2029730e­05
UVM2 ­ 2.91093 7.49769e­05 ­1.1701736 4.2607553e­05
UVW1 ­ 1.16448 4.14899e­05 ­0.30229645 2.5461415e­05
U ­ 0.313603 2.51781e­05 0.19793896 1.5670360e­05
B ­ 0.484320 2.84507e­05 0.14699201 1.6715890e­05
V ­ 1.25429 4.32118e­05 ­0.64141692 3.1820188e­05
data became available, we checked that the degradation had not deviated from its original values
more than 1­2 %.
In the last evaluation, including data obtained in 2011, we see deviations from the original trend
reaching as much as 5 % from the current values in the CCF for UVM2 filter, 4 % for UVW2, 3 %
for UVW1 and 2 % for U, B and V.
Therefore we have updated the correction coe#cients in this new version of the CCF. Table 1
gives both old and new coe#cients.
3 Scientific Impact of this Update
The time dependent sensitivity degradation trend has changed with time. Therefore we need to
update the coe#cients to be able to obtain a proper correction.
In Figure 1 we can see the evolution of the count rates with time for the three observed stars.
The current correction was based in data obtained before 2005, i.e. using less than half of the data
points available now. The new fit (red dotted line) is therefore more precise.
4 Estimated Scientific Quality
The correction coe#cients were thoroughly tested before releasing the SAS code that performs
the correction. The time dependent sensitivity degradation is monitored regularly to ensure the
repeatability and stability of all corrections applied by SAS when new observations and new versions
of SAS become available (see test procedures and their results in the corresponding sections below).

XMM­Newton CCF Release XMM­CCF­REL­274 Page: 3
Figure 1: The normalized count rates of OM as a function of time for three white dwarfs, BPM16274
(black), Hz 2 (red), GD 153 (green)

XMM­Newton CCF Release XMM­CCF­REL­274 Page: 4
5 Expected Updates
As the degradation trend changes in the future, then a new version of the correction coe#cients will
be implemented.
We use currently a linear approximation su#ciently accurate, but the structure of the CCF
allows us to use a quadratic term in the correction if necessary.
Other solutions, e.g. an exponential one, would require a change of the structure of the CCF
and therefore a corresponding change in the CAL
6 Test procedures
The testing of the new correction has two parts. First we compute the corrected counts of the
observed stars using the new coe#cients. In Table 2 we show the mean of the measured counts in
all filters and the corresponding error (the standard deviations given as percentage). We can see
the current correction and the proposed new one, as well as a quadratic approach.
In the second part of the test, we use the new CCF to process with SAS all observations of BPM
16274 to confirm the correctness of the CCF. The results of running SAS are presented in Table 3.
7 Summary of the test results
We can see in Table 2 that the errors in the application of the new correction are smaller than the
ones we would obtain with the current SAS correction. We have tried also a quadratic approximation
that gives errors similar to the new linear one. However, the quadratic solution predicts a change
of slope in the degradation trend in the near future, which has no sense. This is the reason for
selecting the linear one.
Table 3 shows that the new CCF works perfectly in SAS 11.
References

XMM­Newton CCF Release XMM­CCF­REL­274 Page: 5
Table 2: Comparison of time sensitivity degradation correction coe#cients
new correction quadratic approach. current SAS correction
filter rate(c/s) error(%) rate(c/s) error(%) rate(c/s) error(%)
BPM 16274
UVW2 14.63 1.07 14.62 0.61 14.65 3.12
UVM2 30.31 1.14 30.29 1.00 30.35 3.58
UVW1 72.83 0.84 72.81 0.85 72.89 1.82
U 112.87 0.67 112.84 0.62 112.92 1.15
B 107.77 0.85 107.75 0.82 107.83 1.47
V 32.57 1.33 32.55 1.21 32.58 1.92
HZ 2
UVW2 23.64 1.73 23.63 0.96 23.77 3.69
UVM2 48.21 1.40 48.20 0.96 48.50 3.51
UVW1 111.63 1.23 111.62 0.99 111.99 2.28
U 169.00 0.84 168.98 0.73 169.32 1.31
B 149.03 0.90 149.02 0.80 149.38 1.50
V 43.25 1.98 43.24 1.63 43.34 2.18
GD 153
UVW2 82.92 1.33 83.08 0.86 82.01 1.09
UVM2 162.69 1.42 162.92 1.02 160.74 1.14
UVW1 329.58 0.70 329.84 0.52 327.53 0.79
U 420.47 1.64 420.75 1.57 418.88 1.86
B 284.22 0.77 284.37 0.74 282.91 1.25
V 70.64 1.64 70.74 1.51 70.31 1.52
Table 3: Processing all observations of BPM 16274 with SAS 11 and the new
OM PHOTTONAT 0005
filter rate(c/s) error(%)
UVW2 14.64 1.1
UVM2 30.34 1.2
UVW1 72.88 0.9
U 112.98 0.7
B 107.94 0.9
V 32.66 1.4