PACS Spectrometer Spatial Calibration plan in PV phase

1
PACS SpectrometerSpatial Calibration plan in PV
phase
A.Contursi D. Lutz and U. Klaas
2
GOALS
Spec. central pointing (Req. 4.1.1.PCD)?
FOV distortion (Req. 4.1.2 PCD)?
At module level
As function of chopper position
As function of wavelength
PSF determination (Req. 4.1.3 PCD)?
Ghosts (Req. 4.1.4 PCD)?
Straylight (Req. 4.1.5 PCD)?
3
Spectrometer central pointing

This is only a verification of the pointing for
the spectrometer. The Herschel pointing is
calibrated with PACS bolometer.
9
9
Req. 4.1.1 PCD
4
Spectrometer central pointing

MEASUREMENTS - Chopped 99 raster, - Step size
1/4 of spectrometer pixel step size (2.35
assuming a spectral - pixel size 9.4), -
Chopper throw to 1.5 , - Dwell time on each
raster position 12 s - Fixed grating position
(77 and 154 µm), - Point source on Spec virtual
aperture (module 12 BLUE), - Duration (XHSPOT)
0.61 h. ANALYSIS and SIAM UPDATE - 2d Gaussian
fit to get peak coordinates - Fed back to SIAM -
New measurement to check the update ERRORS and
DEPENDNCIES FROM SOLAR ASPECT ANGLES - Repetition
on 7 other sources at different solar aspect
angles Is this necessary or already done during
CP? TOTAL DURATION - 9 measurements 5.5 h
5
Spectrometer central pointing
SOURCES - No need to be perfectly point source
but peak should be clear - Several hundreds Jys

6
FOV DISTORTION
AT Module level

Req. 4.1.2 PCD
7
FOV DISTORTION
As function of chopper

8
FOV DISTORTION
As function of wavelength

9
FOV DISTORTION
MEASUREMENTS I - 27x27 chopped raster (as before
but larger)? - Repeat minimum 5 chopper position,
better all (Opt0,/- Large, /- Medium, /-
Small)? - Fixed grating (77 and 154 µm)? - Dwell
time 12 sec per raster - Total duration 18.5 h
(26 h)? MEASUREMENT II - Chopped scan maps at
fixed grating position. - Chop frequency 1 Hz -
Chop throw medium ( larger than the
spectrometer FOV). - Scan speed 3/sec -
Off-position while chopping before and after scan
map. - Leg length 1.5', - legs 31 legs -
Leg distance 3 - Source in the positive beam -
Final map size 1.51.5 (bit larger than
raster)? - Repeat minimum 5 chopper position,
better all - Fixed grating (77 and 154 µm)? -
Total duration17.5 h (24.1 h)?

10
FOV DISTORTION
SOURCES - Several hundreds Jys - If no point
sources available, small extended OK but clear
peak

11
PSF CHARACTERIZATION
Characterize the PSF profiles Check whether
they are nominal at given wavelength

Req. 4.1.3 PCD
12
PSF CHARACTERIZATION
MEASUREMENTS I (PSF in each spatial pixel at a
given ?)? - 27x27 chopped raster (same as FOV
distortion)? - 1 chopper position - Fixed grating
position (77 and 154 µm)? - PSF in all pixels -
Total duration 3.7 h MEASUREMENTS II (PSF in
central module at other ?)? - 99 chopped raster
(same as for pointing)? - Dwell time per raster
position 12 s - Raster step size in S/C x and
y-direction 2.5 - Source centered on the
raster for the positive beam, - Chop throw
1.5' - Off-positions while chopping before and
after the raster. - Source centered on central
module - Mapped area 22 22 arcsec2 (to get at
least the first Airy ring in the - diffraction
limited case.)? - For 5 chopper deflections at 0,
0.5, 1.5. - Fixed grating positions at 55 /
110 µm and 90 / 180 µm. - Total duration 1.5
h

13
PSF CHARACTERIZATION

SOURCES - Strictly point like source - Several
hundred Jys - Best (and almost unique) candidate
Neptune !!!!! Constraints on observation
period !!!! IMPORTANT NOTE If Neptune available
in PV (unlikely) FOV distortion
measurements give also PSFs measurements. No
need to repeat big raster. Only 9x9 rasters
needed
14
SPECTROMETER GHOSTS

Large map(s) to check the existence of
ghosts. If found, characterize them
100
Req. 4.1.4 PCD
15
SPECTROMETER GHOSTS

MEASUREMENT I - 50x50 chopped raster (same as
FOV distortion)? - 1 chopper position - Fixed
grating position (77 and 154 µm)? - total size
188 - Large chopper throw - Total duration 14
h Further combinations of chopper deflections
and wavelengths will be only implemented, if
analysis of the available data sets suggests any
dependence on chopper position or wavelength.
16
SPECTROMETER GHOSTS

MEASUREMENT II - Very large scan map - Chopper
frequency 1 Hz - Chopper throw 3' - Scan
speed 3/sec - Off-position while chopping
before and after the scan map - Leg length 3' -
legs 45 - Leg distance 4 - Source in the
positive beam, would give a - Final map size
3'x3' (3 times the 27x27 raster size)? - Scan
map repetition 8 - Fixed grating position at
77 and 154 µm - Total duration 8 h The use of
the scan was first thought to be more efficient
than rasters to observe large areas. This turned
out not to be true if measurements are not too
long.
17
SPECTROMETER GHOSTS

UNIFYING PSF AND GHOSTS MEASUREMENTS - The scan
map is 5 h shorter than the raster. - But since
ghosts have to be done on point sources too, we
can combine the PSF measurements at chopper
0 with the ghosts measurements. Instead of
having 83.7 12h we would have 14h
SOURCES - Strictly Point sources, as isolated as
possible and with clean background. - Very
bright (1000 Jy) (Neptune still the best
candidate!)?
18
STRAYLIGHT

5 strips (scan map) around very very bright
source (i.e. Jupiter)? repeat the same
measurement w/o source
X Jupiter movements _at_ 30 Oct 2009
Req. 4.1.5 PCD
19
STRAYLIGHT
MEASUREMENTS - 5 strips around Jupiter (Jupiter
moves 3' per day)? - Each strip unchopped
scan maps - Leg length 15' - legs 4 - Leg
separation 30 - Scan speed 3/sec - Repeat
maps twice to disentangle glitch effects and
increase S/N. - Point to a clean off-position
before after scan. - Strips orientation in TRUE
sky coordinates to avoid Jupiter - If e.g..
Jupiter is selected fix grating at prominent line
so that instantaneous spectrum of any intensity
feature could support origin by Jupiter. -
Repeat same types of measurements with Jupiter
far away. - Total duration 19 h Rasters would
be too long due to overheads!

20
CANDIDATES SOURCES

PLANETS Best candidates small and bright!
_at_ 154 µm
250 Jy
600 Jy
1.5e4 Jy
2.5e4 Jy
PV
21
CANDIDATES SOURCES

Neptune (best candidate for brightness and size)
BUT basically not visible during PV Uranus
(good for brightness not for PSF
measurements)? is visible most of PV. We could
use Uranus for all Spectrometer Spatial
Calibration measurements but PSFs. We still need
other than Uranus sources for the Central
Pointing dependency from the solar aspect
angle.
22
CANDIDATES SOURCES

Neptune
27x27 raster lasts 3.7 h If we observe where
vlt0.5/h Neptune moves 2.3 in the all raster.
Start to become significant for PSF.
Velocity (arcsec/hour)?
0.5/h
Uranus
Velocity (arcsec/hour)?
0.5/h
Days since 1 May 2009
23
CANDIDATES SOURCES

Planets can be observed as Fixed objects
not advisable for PSF Moving object
tracking must be good reduction
could become an issue Point like
candidates Not moving objects reduction much
more straightforward They might be not point
sources after all. They will be checked with
photometer
24
Proposed strategy
SOURCES Use point like candidates for all but
PSF and ghosts measurements. (The important
requirement is to get the intensity peak)? Use
Neptune ONLY for PSF and ghosts, as moving
target. Since this will be observed after PV at
that point in time we might have
1) gained experience for reduction strategy
2) have s/w available for reduction
Type of measurements Since scanning does not
seem to be significantly more efficient than
rastering, do scanning only when rastering is
prohibitive (i.e. Straylight). Try to double some
observations with both techniques for exploration
purposes (FOV Distortion).

25
Summary measurements
Spec Central Pointing 9 9x9 rasters , total
time 5.5 h
source with good peak FOV distortion 7 (each
chop position) 27x27 rasters , total time 26
h 5 scan maps, total
time 17.5 h source
with good peak PSF and Ghosts 6 27x27
rasters (all chop but center)
total time 16.2h
One 50x50 raster at chop0, total time
13.2 h 2 9x9 rasters
at chop 0 at other ?, total time 1.5 h
Source Neptune (POST PV !!!)
Straylight 5 scan maps, total time 19h
TOTAL SPEC SPATIAL CALIBRATION CAMPAIGN
DURATION 99 h ( 68 h in PV, 31 h after PV)?

26
CONCLUSIONS

Time expensive campaign. Raster type data
format already known (analysis should be
straightforward)? Scan map never tried.
Moreover no such pointing mode is available for
spectrometer science data. This means no piece of
s/w is available neither. If scan map can be
reduced at a desired precision level, we will
have the following advantages 1) have redundant
data for the spatial calibration 2) the
possibility to introduce this mode for the
spectrometer later in the mission.....