Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.apo.nmsu.edu/35m_operations/35m_manual/Instruments/Grim/OLD/grim/manual.ps Äàòà èçìåíåíèÿ: Fri Aug 22 02:46:00 2003 Äàòà èíäåêñèðîâàíèÿ: Sun Apr 10 10:26:43 2016 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: m 8

GRIM II User's Manual
Alan Watson, Mark Hereld, and Bernie Rauscher
15 March 1997

`Jeeves,' I said, `have you ever pondered on Life?'
`From time to time, sir, in my leisure moments.'
`Grim, isn't it, what?'
`Grim, sir?'
`I mean to say, the difference between things as they
look and things as they are.'
--- P. G. Wodehouse anticipating the
advantages of infrared astronomy
in Very Good, Jeeves!
i

Contents
1 Introduction 1
1.1 GRIM II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 User's Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.3 Mailing List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2 Optics 3
3 Detector 5
3.1 Detector Characteristics . . . . . . . . . . . . . . . . . . . . . . . 5
3.2 Detector Operation . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3.3 Bias and Dark Current . . . . . . . . . . . . . . . . . . . . . . . . 6
3.4 Response Uniformity . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.5 Non­Linearity and Saturation . . . . . . . . . . . . . . . . . . . . 6
3.6 Bad Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.7 Residual Image . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.8 Peculiarities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4 Operation 15
4.1 REMARK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.2 MC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.2.1 Commands . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.2.2 Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2.3 Example Session . . . . . . . . . . . . . . . . . . . . . . . 23
4.3 Hangs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5 Imaging 25
5.1 Cameras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.2 Neutral Density Filters . . . . . . . . . . . . . . . . . . . . . . . . 25
5.3 Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.3.1 Transmittances . . . . . . . . . . . . . . . . . . . . . . . . 25
5.3.2 Zero Points and Backgrounds . . . . . . . . . . . . . . . . 28
5.3.3 Broad Band Filters . . . . . . . . . . . . . . . . . . . . . . 28
5.3.4 Narrow Band Filters . . . . . . . . . . . . . . . . . . . . . 35
5.4 Bright Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
ii

5.5 Flat Field Images . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.6 Basic Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
6 Spectroscopy 39
6.1 Capabilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
6.2 Flat Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
6.3 Wavelength Calibration . . . . . . . . . . . . . . . . . . . . . . . 39
6.4 Absorption Standards . . . . . . . . . . . . . . . . . . . . . . . . 41
7 Extinction 42
8 Standards 44
8.1 Photometric Standards . . . . . . . . . . . . . . . . . . . . . . . . 44
8.1.1 Elias et al. JHK, CO, and H 2 O Standards . . . . . . . . 44
8.1.2 Carter & Meadows JHK Standards . . . . . . . . . . . . 45
8.1.3 Casali & Hawarden JHK Standards . . . . . . . . . . . . 45
8.1.4 Wainscoat & Cowie K 0 Standards . . . . . . . . . . . . . 45
8.1.5 Manufactured K 0 Standards . . . . . . . . . . . . . . . . . 45
8.1.6 Absolute Calibration . . . . . . . . . . . . . . . . . . . . . 45
8.2 Spectrophotometric Standards . . . . . . . . . . . . . . . . . . . 46
8.2.1 Bohlin Spectrophotometric Standards . . . . . . . . . . . 46
8.2.2 Manufactured Spectrophotometric Standards . . . . . . . 46
8.3 Stellar Colours . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8.4 Stellar Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
9 Headers 48
iii

List of Figures
2.1 Optical Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.1 Bias Image (IRAF format; 1 second) . . . . . . . . . . . . . . . . 7
3.2 Bias Image (FITS format; 1 second) . . . . . . . . . . . . . . . . 8
3.3 Flat Field Image in K . . . . . . . . . . . . . . . . . . . . . . . . 9
3.4 Ratio of Flat Fields Images in J and K . . . . . . . . . . . . . . 10
3.5 Non­linearity and Saturation . . . . . . . . . . . . . . . . . . . . 12
3.6 Bad Pixels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.1 The obj4 procedure . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.2 The fexp procedure (for n = 5) . . . . . . . . . . . . . . . . . . . 23
5.1 Camera Orientations for Imaging . . . . . . . . . . . . . . . . . . 26
5.2 Neutral Density Filter Transmittances . . . . . . . . . . . . . . . 27
5.3 Filter Transmittances . . . . . . . . . . . . . . . . . . . . . . . . 30
5.4 Filter Transmittances (continued) . . . . . . . . . . . . . . . . . . 31
5.5 Filter Transmittances (continued) . . . . . . . . . . . . . . . . . . 32
5.6 Filter Transmittances (continued) . . . . . . . . . . . . . . . . . . 33
5.7 Broad Band Filter and Atmospheric Transmittances . . . . . . . 34
7.1 Model Atmospheric Transmitances . . . . . . . . . . . . . . . . . 43
iv

List of Tables
4.1 Modes Values and Names . . . . . . . . . . . . . . . . . . . . . . 17
4.2 Scale Values and Names . . . . . . . . . . . . . . . . . . . . . . . 17
4.3 Filter Values and Names . . . . . . . . . . . . . . . . . . . . . . . 18
5.1 Camera Configurations for Imaging . . . . . . . . . . . . . . . . . 26
5.2 Filter Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . 29
5.3 Filter Zeropoints and Backgrounds . . . . . . . . . . . . . . . . . 29
5.4 Imaging Dome Flat Characteristics at f=5 . . . . . . . . . . . . . 36
6.1 Wavelength Coverage in ¯m . . . . . . . . . . . . . . . . . . . . . 40
9.1 Header Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
v

Chapter 1
Introduction
1.1 GRIM II
GRIM II is the near infrared camera and low resolution spectrograph in service How should
GRIM II and APO
be acknowledged
in publications?
on the Apache Point Observatory 3.5 meter telescope. GRIM II has a 256 \Theta 256
NICMOS­3 detector and works between 1.0¯m and 2.5¯m. In imaging mode
it has three pixel scales of about 0.48, 0.24, and 0.11 arcsec/pixel and a large
number of broad and narrow band filters. In spectroscopic mode it has three
resolutions of about 200, 400, and 800. GRIM II was designed and built by Mark
Hereld, Bernie Rauscher, Scott Severson, and Bob Loewenstein of the Univeristy
of Chicago with engineering support from Dale Sandford, Fred Mrozek, Dave
Fischer, and Jeff Sundwall.
1.2 User's Manual
This manual has two purposes: to provide a basic introduction to near infrared
imaging and low­resolution spectroscopy for astronomers familiar with optical
CCD imaging and spectroscopy and to describe the specifics of performing such
observations with GRIM II. The authors of this manual are
Alan Watson alan@oldp.nmsu.edu
Mark Hereld hereld@bucephalus.uchicago.edu
Bernie Rauscher B.J.Rauscher@durham.ac.uk
Direct and indirect contributions have also been made by
Eddie Bergeron
Jon Brinkmann
Nancy Chanover
Karen Gloria
Jon Holtzman
Mike Ledlow
1

Bob Loewenstein
Dan Long
James Rhoads
Scott Severson
This manual is an evolving document; the lastest version is available as
ftp://oldp.nmsu.edu/pub/alan/grim/man.ps.Z
Versions are identified by the date on the cover. If you find errors or have
suggestions for improvements, please send them to the authors.
1.3 Mailing List
Another source of information on GRIM II is the GRIM II mailing list main­
tained by Michael Strauss. An archive of mail sent to the list and instructions
for subscribing to and sending mail to the list are available from
http://www.astro.princeton.edu/APO/apo35­grim/INDEX.html
2

Chapter 2
Optics
The optical layout of GRIM II is shown in Figure 2.1. All of the optical compo­
nents are contained in a cryogenic dewar (which is purple and very pretty). The
f=10 beam from the telescope enters the dewar and comes to a focus at the slit
wheel. After that it passes through the field lens, collimator lens, grism wheel,
two filter wheels, and then the lenses and fold mirrors of one of the three cameras
mounted on the camera carousel, before coming to a focus on the detector.
The grism wheel contains an aperture stop, the grism, and 3%, 13%, and
25% transmission neutral density filters. The aperture stop is a circular aperture
without a central obscuration to block the high­emissivity central hole in the
primary or rotating vanes to block the spiders. The filter wheels contain a large
number of broad and narrow band filters along with a solid plate known as the
`blank­off' or `dark' filter.
The f=5 camera has only one fold mirror but the f=10 and f=20 cameras
have two. Thus, the image formed by the f=5 camera is flipped about one axis
compared to the images formed by the other two (see x5.1). The throughputs
of the cameras is slightly different as they each have different optics.
The oblique reflections within GRIM II and from the tertiary mirror induce
polarization. Since GRIM II and the tertiary rotate with repect to each other
under normal circumstances, there is a photometric modulation of even unpolar­
ized sources. The amplitude of this modulation is though to be in the vicinity of
1% RMS. Polarized sources will also suffer from modulations and offsets because
of instrumental polarization.
3

Slit Wheel Field Lens
Accessory Ring
Collimator Lens
Grism and Filter Wheels
Camera Carousel
Detector
Figure 2.1: Optical Layout
4

Chapter 3
Detector
3.1 Detector Characteristics
GRIM II has a 256 \Theta 256 NICMOS­3 detector. The detector is sensitive from
0.8¯m to 2.5¯m. The long wavelength cutoff lies at the red end of the atmo­
spheric K window and is sufficiently short that low backgrounds can be obtained
by cooling the detector and optics to 77 K with liquid nitrogen.
The device is split into quadrants each of which has its own amplifier. The
quadrants are read simultaneously. The detector has a gain of about 4.7 elec­
trons/DN and an effective read noise of about 110 electrons. Thus, the read
noise is larger than the Poisson noise for exposures of less than about 2500 DN.
3.2 Detector Operation
GRIM II does not have a shutter. Instead, exposures are controlled electroni­
cally. GRIM II operates in a `reset, read, read' or `double correlated sampling'
mode; this mode has lower noise than the `reset, read' mode, which is analogous
to the mode in which CCDs are operated.
An exposure begins with a reset, which sets the bias in each pixel. A short
time after being reset, the chip is read. The read is non­destructive and merely
samples the voltage in each pixel. After a further time the chip is read once
more. The GRIM II controller does not allow multiple first and second reads.
The signal \Deltan is the difference between the first and second reads. The time
between the reset and the first read is about 0.95 seconds. The time between
the two reads is the exposure time \Deltat and is
\Deltat = 0:901 + OPENTIME (3.1)
in seconds where OPENTIME is a header value. The exposure time is about 0.2
seconds less than the time requested using REMARK or MC. The shortest
exposure time possible is 1.0 seconds (corresponding a requested time of 1.2
seconds).
5

3.3 Bias and Dark Current
Between the first and the second reads, the signal chain bias level changes. Thus,
even a short dark exposure has values which are far from zero. To obtain the
true signal, one must subtract a bias image. The amount by which the bias level
changes depends on the exposure time in the sense that longer exposures have
more negative bias levels. For this reason, one must construct a separate bias
image for each exposure time used. Since bias images can only be constructed
from dark exposures, subtracting a bias image also removes the dark current. As
the dark current is small, this has a negligible effect on subsequent corrections
for non­linearity. Dark images can be taken using the `dark' or `blank off' filter.
IRAF images obtained with GRIM II have a constant of 10000 DN added
to them by the MC. This allows the values to be comfortably represented as
16­bit unsigned integer. FITS images obtained with GRIM II do not have this
constant added because BZERO header value can be used to achieve the same
ends. The values in short dark exposures are about \Gamma2000 DN for FITS images
and 8000 DN for IRAF images. 1 second bias images are shown in Figure 3.1
for IRAF images and Figure 3.2 for FITS images.
The bias fluctuates, giving rise to the bands seen in the lower rows of each
quadrant. The amplitude of the fluctuation seems to depend primarily on the
exposure, so that the bands largely disappear when two similar exposures are
subtracted but remain when two different exposures are subtracted. In conse­
quence, the bands largely disappear in the course of the normal processing of
object frames (as sky frames are subtracted). Unfortunately they remain in flats
(as a low exposures are subtracted from a high exposures) at the few percent
level. One approach to this problem which seems to work well is normalizing
each row individually so that the central columns of the flat have the same
mean.
Low­level diagonal banding in the lower right quadrant has also been re­
ported.
3.4 Response Uniformity
The response of the detector in GRIM II is fairly uniform compared to many
NICMOS­3 detectors. Figure 3.3 shows a flat field image in K. However, the
response changes as a function of wavelength; Figure 3.4 shows the ratio of flat
field images in J and K.
3.5 Non­Linearity and Saturation
Since the influence of the ARC board does not extend to waiving the laws
of physics, the GRIM II detector is by necessity slightly non­linear. The non­
linearity is about 6% over the working range of about 28000 DN and the cor­
rections for non­linearity can be as large as 10%. Furthermore, those unaware
6

Figure 3.1: Bias Image (IRAF format; 1 second)
7

Figure 3.2: Bias Image (FITS format; 1 second)
8

Figure 3.3: Flat Field Image in K
9

Figure 3.4: Ratio of Flat Fields Images in J and K
10

of the mode in which GRIM II operates can inadvertently expose the chip be­
yond its working range; saturation can occur for signals as low as 13000 DN.
Fortunately, the non­linearity can be characterized and corrected and with fore­
warning it is not difficult to stay within the working range. The origin of the
non­linearity suggests that its characteristics should be stable with time.
The non­linearity in a near infrared array occurs because each pixel in the
array is effectively a capacitor. The pixel is read by sensing the voltage across
the capacitor. At the start of the exposure, the capacitor is biased. Thereafter,
it accumulates charge and the bias decreases. If the capacitance of the pixel
were constant, the voltage would be linearly related to the accumulated charge.
Unfortunately, the capacitance increases with decreasing bias and the relation
between accumulated charge and voltage is sub­linear.
Alan Watson and Nancy Chanover investigated and characterized the non­
linearity of GRIM II in June 1996. The full text of their report is available
as
ftp://oldp.nmsu.edu/pub/alan/grim/lin.ps.Z
Their main conclusions are sumarized here.
The non­linearity can be adequately modelled as a straight line, that is the
signal N that would be accumulated by a truly linear detector can be related
to the actual signal n by
n = (1 \Gamma – 1 N)N (3.2)
where – 1 is a constant.
Inverting this expression to correct for the non­linearity is hampered by the
fact that the detector accumulates charge for a time t 1 between the reset and the
first read and then a further time \Deltat between the first read and the second read.
As noted in x3.2, the actual exposure time is different both from the requested
exposure time and the OPENTIME header value. The signal \Deltan is the difference
between the first and second reads. (A bias image needs to be subtracted to
obtain \Deltan.) The value of the signal \DeltaN that would be given by a perfectly
linear detector is
\DeltaN = 1 \Gamma (1 \Gamma 4– 1 \Deltan(\Deltat + 2t 1 )=\Deltat) 1=2
2– 1 (\Deltat + 2t 1 ) \Deltat: (3.3)
From eight sequences of increasing exposures, they found
– 1 = 2:18 \Theta 10 \Gamma6 (3.4)
and
t 1 = 0:95: (3.5)
Contours of the fractional correction are shown in Figure 3.5; the corrections
can be as large as 10%.
11

0 1 2 3 4 5 6 7 8 9 10
Exposure Time Dt
0
5000
10000
15000
20000
25000
30000
Signal
Dn
0 1 2 3 4 5 6 7 8 9 10
0
5000
10000
15000
20000
25000
30000
0 1 2 3 4 5 6 7 8 9 10
0
5000
10000
15000
20000
25000
30000
1.01
1.01
1.02
1.02
1.03
1.03
1.04
1.04
1.05
1.05
1.06
1.06
1.07
1.07
1.08
1.08
1.09
1.09
1.10
1.10
0 1 2 3 4 5 6 7 8 9 10
0
5000
10000
15000
20000
25000
30000
Figure 3.5: Non­linearity and Saturation
The full well of the detector is only about 28000 DN. Since charge accumu­
lates between the reset and the first read, saturation can occur well before the
signal approaches this value. To avoid saturation, the signal must be kept below
the values indicated by the thick line in Figure 3.5. Thus, in a 1 second exposure
the signal must be kept below about 13000 DN. (The `signal' in question here
is the value in the image after subtracting the bias, which is about \Gamma2000 for
FITS images and about 8000 for IRAF images.)
3.6 Bad Pixels
The GRIM II detector is cosmetically excellent compared to many other similar
detectors. Figure 3.6 shows an image of the bad pixels constructed by dividing
a 15000 DN flat field from a 1500 DN flat field and flagging pixels that deviated
from the mean by more than 10%. Most of the bad pixels are isolated, although
there are three clumps of bad pixels, one of which is fairly close to the center
of the detector. The number of bad pixels is expected to increase slowly with
time.
3.7 Residual Image
Like all NICMOS­3 detectors, the detector in GRIM II suffers from residual It might be useful
to characterize
this.
image. This has not been well characterized. The conventional wisdom is that
taking a series of bias exposures helps to eliminate a residual image.
12

Figure 3.6: Bad Pixels
13

3.8 Peculiarities
Columns 128 and 256 are offset down by one row. The values in pixels (128; 128),
(128; 256), (256; 128), and (256; 256) in the image are garbage. The values in
pixels (128; i) and (256; i) in the image actually correspond to the values in
pixels (128; i + 1) and (256; i + 1) on the detector.
14

Chapter 4
Operation
GRIM II can be controlled by normal users with either the REMARK interface
or the MC interface. Both allow the telescope and instrument to be controlled
over the Internet and both can automatically copy images to a remote host
using FTP.
4.1 REMARK
REMARK runs on a networked Macintosh and provides a remote, graphical
interface for controlling both the telescope and instruments. REMARK was
written by Bob Loewenstein. The documentation for REMARK can be found
on the APO home page (http://www.apo.nmsu.edu). At the time of writ­
ing, the documentation describing the operation of GRIM II using REMARK is
incomplete.
4.2 MC
MC runs on tycho.apo.nmsu.edu and provides a command line interface for
controlling both the telescope and instruments. MC was written by Brian
Yanny. The documentation for MC can be found from the APO home page
(http://www.apo.nmsu.edu). The advantage of the MC over REMARK is
that it can be programmed; procedures can be written to perform sequences of
tasks such as exposures and offsets. An MC session can be started on tycho
with
mcnode
An MC status display can be started on tycho with
mcnode ­s
in a 80 \Theta 26 VT100­compatible window (e.g., xterm ­geom 80x26). It is also
useful to monitor the hub log on tycho with
15

tail ­f /home/apotop/syslog/hub.log
In particular, FTP error messages appear in the hub log.
4.2.1 Commands
The most common MC commands are listed here. The descriptions are some­
what abbreviated and often do not show all of the options; consult the MC
documentation for more details on these commands and the less common com­
mands.
priority priority
The priority command sets the priority of this MC session. A priority
value of 0 allows only harmless commands; a priority value of 1 allows all
commands. A new MC session initially has a priority of 0.
grimmove mode scale f ilter
The grimmove command moves the slit wheel, grism wheel, filter wheels,
and camera carousel in GRIM II. The numerical values of mode, scale,
and f ilter are listed in Tables 4.1, 4.2, and 4.3. These values appear as
the MODE, SCALE, and GFILTER header values. Occasionally, a move will
fail and need to be repeated.
grimstatus
The grimstatus command requests a message describing the status of the
grim optical components. The numerical values returned correspond to
the mode, scale, and filter values listed in Tables 4.1, 4.2, and 4.3.
inst instrument
The inst command sets the current instrument for the nexpose com­
mands. An instrument value of grim specifies GRIM II.
nexpose itime=itime n=n [reduce='send?ftp']
The nexpose command takes n exposures each of itime seconds. As noted
in x3.2, the actual exposure time is about 0.2 seconds shorter than the
value of itime. If reduce='send?ftp' is specified and an FTP connection
has been established with the loginftp command, the image is FTP­ed
to the remote host.
grimabort
The grimabort command aborts the current exposure.
slew hh:mm:ss [+j­]dd:mm:ss epoch=epoch
The slew command executes a slew to the specified position. A slew can
be aborted with the stop tcc command.
16

Mode Value Name
image 0 image
grism with slit 1 grism+slit
image with slit 2 image+slit
grism without slit 3 grism
image with 3% ND filter 4 image+nd3
image with 13% ND filter 5 image+nd13
image with 25% ND filter 6 image+nd25
Table 4.1: Modes Values and Names
Scale Value Name
f=5 1 f5
f=10 3 f10
f=20 5 f20
f=20 short 13 f20short
f=20 long 21 f20long
Table 4.2: Scale Values and Names
17

Filter Value Name
open 13 open
dark 0 dark
J 1 j
H 2 h
K 3 k
K 0 4 kprime
K s 5 ks
K dark 7 kdark
1.08¯m 15 1.08
1.09¯m 16 1.09
1.24¯m 17 1.24
1.28¯m 18 1.28
1.58¯m 8 1.58
1.64¯m 19 1.64
1.70¯m 9 1.70
1.99¯m 20 1.99
2.12¯m 21 2.12
2.17¯m 22 2.17
2.21¯m 23 2.21
2.25¯m 24 2.25
2.26¯m 6 2.26
2.30¯m 25 2.30
2.34¯m 26 2.34
Table 4.3: Filter Values and Names
18

stop tcc
The stop tcc command aborts a slew.
offset x y type [abs]
The offset command executes an offset of x arcseconds in the x direction
and y arcseconds in the y direction.
If abs is specified, the offset is relative to the last slew, otherwise the offset
is relative to the last offset of the given type. The type value can be inst
or obj to specify `instrument' or `object' offsets. Instrument and object
offsets are independent.
Object offsets have a coordinate system that always has x west and y
north. Object offsets are reflected in the RA, DEC, RAOFF and DECOFF
header values.
Instrument offsets have a coordinate system which rotates with the in­
strument. At 0 degrees rotation, x is east and y is south. With the f=5
camera, the x and y axes of the coordinate system match the axes of
the detector, but with the other cameras the coordinate system is flipped
about the x axis. Instrument offsets are reflected in the the X and Y header
values.
rotate angle type
The rotate command sets the rotator position angle to angle. The
type values can be object or horizon. The rotation is recorded in the
ROTATION header values. An object rotation of 0 degrees gives the orien­
tations shown in Table 5.1 and Figure 5.1. A horizon rotation of 0 degrees
places the slit parallel with the horizon.
focus focus
The focus command sets the telescope focus to focus. When an in­
strument is mounted, the operator normally sets the focus to something
reasonable.
imdir dir
The imdir command specifies the directory in which images are to be
created. The directory must have write permission for all users. Images
are created in /export/images by default.
diskname prefix
places places
filetype type
seq number
The diskname, places, filetype, and seq commands specify the name
and type of image created.
19

The file name is prefix, followed by the sequence number padded with
zeros to a width of places, followed by either .hhh or —
hhd if IRAF images
are being written or .fit if FITS images are being written.
The filetype command specifies that IRAF or FITS images are created
depending on whether type is iraf or fits.
The seq command sets the sequence number for the next exposure.
subscribe type level
The subscribe command determines which messages are printed in this
MC session. Values for type are message, monitor, and status. Values
of level are 0, 1, and 2, with 0 switching messages off.
The MC is quite verbose. It is useful to have two MC session running
simultaneously, with one being used only for commands and having all
messages turned off and the other being used only for messages.
loginftp hostname username senddir=dir
The loginftp command establishes an FTP connection to the directory
dir on host host. You will be prompted for a password.
4.2.2 Procedures
MC procedures can be written in the TCL language. Information on TCL is
available from http://www.NeoSoft.com/tcl/. Writing MC procedures is not
trivial. The basic model is described by Brian Yanny in the MC documentation.
A somewhat higher­level model has been suggested by Alan Watson in a message
to the GRIM II mailing list on 8 July 1996.
A number of useful procedures, which can also serve as examples from which
to create your own, are available from
ftp://oldp.nmsu.edu/pub/alan/grim/mc.tcl
This file also exists on tycho.apo.nmsu.edu as
~visitor1/alan/grim/mc.tcl
To use these procedures, start an MC session on tycho.apo.nmsu.edu with the
mcnode command and then type:
source ~visitor1/alan/grim/mc.tcl
start
config
loginftp hostname username senddir=dir
Because of the way MC works, many of these procedures return before they
have completed; do not attempt to execute another command until the message
stating that the procedure has completed. These procedures will attempt to
FTP any exposures they take. The available procedures are:
20

start
This command must be issued before any others.
end
This command should be issued to relinquish control of GRIM II.
config
Query grim for the current mode, camera configuration, and filter and
print them in recognizable forms. This command must be issued after the
initial start command but before any others.
mode mode
Set the mode according to mode, which can be one of one of names listed
in Table 4.1. Occasionally, a move will fail and need to be repeated.
scale scale
Set the scale according to scale, which can be one of names listed in
Table 4.2. Occasionally, a move will fail and need to be repeated.
filter f ilter
Set the filter according to f ilter, which can be one of names listed in
Table 4.3. Occasionally, a move will fail and need to be repeated.
north arcsec
south arcsec
east arcsec
west arcsec
Move the telescope arcsec north, south, east, and west.
center x y
Move an object that is at pixel location (x; y) to the center of the detector.
exp exptime n
Take n exposures each of duration exptime. (The procedure actually
requests an exposure of 0.2 seconds longer than exptime so that the actual
exposure corresponds to the exptime. See x3.2.)
obj1 exptime n skyx skyy
Take 2 sequences of n exposures each of duration exptime. The sequences
are object then sky. The object exposures are taken at the current po­
sition. The sky exposures are taken at an instrument offset of skyx and
skyy arcsec. The telescope is left at the initial position.
21

d
d
d
3
d
2
4
5
6 7
8
1
0 skyx
skyy
Figure 4.1: The obj4 procedure
obj4 exptime n skyx skyy d
Take 8 sequences of n exposures each of duration exptime. The sequences
are sky, object, object, sky, sky, object, object, sky. The object exposures
are each centered at the corner of a square of side d arcsec centered on
the current position. The sky exposures are each centered at the corner
of a square of side d arcsec centered on an instrument offset of skyx and
skyy. The telescope is left at the initial position. This is illustrated in
Figure 4.1. The telescope starts at position 0, and then takes a sequence
of exposures at positions 1, 2, 3, 4, 5, 6, 7, and 8, before finally returning
to position 0 again.
home
Return the telescope to its position after the last slew command or the
last center, north, south, east, or west procedure. This is useful if an
obj1 or obj4 procedure has to be aborted.
fexp exptime fstart fend n
Take a focus run of n exposures each of duration exptime with the focus
set to values equally spaced between fstart and fend inclusive. The
focus star should be roughly centered before this procedure is run. The
22

5
4
focus = fend
0
3
2
1 focus = fstart
Figure 4.2: The fexp procedure (for n = 5)
exposures are offset in the north--south direction with a larger gap before
the final image. This is illustrated in Figure 4.2 for n = 5. The telescope
starts at position 0, and then takes a sequence of exposures at positions 1,
2, 3, 4, and 5, before finally returning to position 0 again. The exposures
can be combined to form a single focus image by subtracting the median
from the mean. The operator can advise you of a reasonable value for the
focus.
4.2.3 Example Session
In this example session, command are set in this type and comments in normal
type.
Load procedures and initialize:
source ~visitor1/alan/grim/mc.tcl
start
config
Tell the MC to create FITS files like 970120.0001.fit:
filetype fits
diskname 970120.
places 4
seq 1
Set f=5 imaging mode with the H filter:
mode image
scale f5
filter h
23

Slew to an object:
slew 00:01:02 ­03:04:05 epoch=2000
Take a finder exposure:
exp 1 1
Move the telescope 10 arcsec north and 5 arcsec east:
north 10
east 5
Take 3 exposures each of 5 seconds, offset 600 arcsec to sky, take 3 more,
and return to the object:
obj1 5 3 600 0
4.3 Hangs
Sometimes GRIM II hangs just after starting an exposure. If you are using RE­
MARK, you can recover by closing and reopening the exposure control window.
If you are using MC, you can recover by issuing a grimabort command.
24

Chapter 5
Imaging
5.1 Cameras
GRIM II has three cameras: f=5, f=10, and f=20. Table 5.1 lists the pixel
scales, fields of view, and orientations at 0 degrees rotation of the cameras.
The orientations assume that the origin is at the lower left. The plate scales
were measured by Alan Watson and reported to the GRIM II mailing list on
29 January 1997. The uncertainties in the pixel scales are estimated to be
about 0.001 arcsec. Figure 5.1 shows the orientation of the cameras. If you are
working in f=10 or f=20 and switch to f=5 to acquire an object, remember that
the orientation will change.
5.2 Neutral Density Filters
There are three neutral density filters known as ND03, ND13, and ND25. Un­
fortunately, these filters are visible light neutral density filters; in the infrared
they are not especially neutral and have similar transmissions. Their transmis­
sions are shown in Figure 5.2. The ND13 filter supplies the most blocking in
the infrared and has transmissions of about 10% in J , 25% in H, and 35% in
K.
5.3 Filters
5.3.1 Transmittances
Table 5.2 lists for each filter the central wavelength – 0 j
R
T– d–=
R
T d–, width
\Delta– j
R
T d–= max(T ), resolution R j \Delta–=– 0 , and integrated transmittance
R
T d–. Transmittance curves are not available for a number of filters; the
values for these filters are estimates. Figures 5.3 to 5.6 show the transmittances
of the filters. The transmittances are available in electronic form as
25

Camera Pixel Scale Field of View N E
arcsec arcsec
f=5 0.473 120 down right
f=10 0.236 60 up right
f=20 0.113 30 up right
Table 5.1: Camera Configurations for Imaging
(b) f/10 and f/20
(a) f/5
N
N
E
E
Figure 5.1: Camera Orientations for Imaging
26

Figure 5.2: Neutral Density Filter Transmittances
27

ftp://oldp.nmsu.edu/pub/alan/grim/filters.tar.Z
No significant shifts in focus between the filters are expected and none have
been reported.
5.3.2 Zero Points and Backgrounds
Typical zero points and backgrounds at one airmass for the broadband filters
are listed in Table 5.3. These zero­points can be used to estimated efficiencies,
but should not be used for calibration. The zero points are defined by
m j \Gamma2:5 log —
n + ZP (5.1)
where m is the standard magnitude and —
n is the signal in DN s \Gamma1 .
The backgrounds will vary with the airmass, season, atmospheric conditions,
and mirror reflectivity. These measurments were made on a photometric night in
January 1997, just after the mirrors were realuminized, with an air temperature
of about 275 K.
The throughputs of the narrow band filters can be estimated from the Measurements
would be useful.
throughputs of the broad band filters by scaling by the ratios of the integrated
transmissions from Table 5.2. The background in the near infrared is domi­
nated by many OH lines in the J , H , and the blue end of the K window and
thermal emission at the long end of the K window. This complex structure
makes it more difficult to estimate narrowband backgrounds from the broad
band measurements.
5.3.3 Broad Band Filters
The transmittances of the GRIM II broad band filters are shown in Figure 5.7.
Also shown is a representative MODTRAN­3 model atmospheric transmittance
curve. The JHK filters are shown with solid lines, the K 0 filter with a dotted
line, the K s filter with a dashed line, and the K dark filter with a dashed­dotted
line.
The K 0 filter was designed for deep imaging from Mauna Kea and has been
described by Wainscoat & Cowie (1992). It has a shorter effective wavelength
than the K filter and so has a lower thermal background. However, its blue cut
off encroaches on the atmospheric H 2 O band between the H and K windows; on
Mauna Kea this is not so much of a problem, but at lower sites the absorption
reduces the throughput of the filter, makes the extinction correction more non­
linear, and introduces significant variablitity into the band pass of the filter.
For these reasons, the K s filter is probably a better choice for observations from
APO.
The K s filter was designed for deep imaging from moderate altitude obser­ Does anyone have
a reference for
Ks?
vatories. It will be used in the 2 Micron All Sky Survey. It has a longer blue
cut off than the K 0 filter but a shorter red cut off than the K filter.
The K dark filter is designed for low­background imaging from the South Pole
(Nguyen et al. 1996; Ashley et al. 1996).
28

Filter – 0 \Delta– \Delta–=– 0
R
Td–
¯m ¯m ¯m
J 1.265 0.267 0.211 0.221
H 1.646 0.339 0.206 0.276
K 2.187 0.399 0.182 0.305
K 0 2.114 0.343 0.162 0.310
K s 2.155 0.324 0.150 0.300
K dark 2.359 0.147 0.062 0.113
1.08¯m 1.084 0.013 0.012 0.007 He I
1.09¯m 1 1.094 0.010 0.010 Pfl
1.24¯m 1.235 0.011 0.009 0.006 O II
1.28¯m 1.282 0.015 0.012 0.011 Pfi
1.58¯m 1 1.580 0.010 0.006 continuum
1.65¯m 1.647 0.019 0.012 0.016 [Fe II]
1.70¯m 1 1.700 0.050 0.030 CH 4
1.99¯m 1 1.990 0.020 0.010 H 2 O
2.12¯m 2.120 0.030 0.014 0.018 H 2 v = 1--0 S(1)
2.17¯m 2 2.161 0.032 0.015 0.022 Brfl
2.21¯m 2.210 0.100 0.045 0.062 CO off
2.25¯m 2.248 0.025 0.011 0.020 continuum
2.26¯m 2.260 0.055 0.024 0.047 H 2 v = 2--1 S(1)
2.30¯m 2.297 0.028 0.012 0.022 CO 2--O band head
2.34¯m 2.342 0.091 0.039 0.057 CO on
1 No transmittance curve; values are estimates.
2 1987 transmittance curve (see x5.3.4).
Table 5.2: Filter Characteristics
Filter ZP Background
mag DN s \Gamma1 arcsec \Gamma2
J 23.3 1100
H 23.1 5100
K 22.2 6300
K s 22.4 4700
Table 5.3: Filter Zeropoints and Backgrounds
29

Figure 5.3: Filter Transmittances
30

Figure 5.4: Filter Transmittances (continued)
31

Figure 5.5: Filter Transmittances (continued)
32

Figure 5.6: Filter Transmittances (continued)
33

Figure 5.7: Broad Band Filter and Atmospheric Transmittances
34

Whether K 0 , K s , or K dark give better signal­to­noise than K depends on
colour of the object and the background. Wainscoat & Cowie (1992) perform
this calculation for their K 0 filter on Mauna Kea, but their results may not be
valid for APO. Since the K, K 0 , K s , and K dark filters have different effective
wavelengths and band passes, the K, K 0 , K s , and K dark magnitudes of an
object are all different in general.
5.3.4 Narrow Band Filters
GRIM II has a very wide selection of narrow band line filters, but is somewhat
lacking in narrow band continuum filters. For some observations, broad band
filters are adequate for the continuum. However, when imaging a weak line
against a continuum or in the presense of other lines, this approach can fail.
Some users have had success using nearby narrow band line filters (e.g., [OII]
with Pafi and H 2 v = 1 \Gamma 0 S(1) with Brfl), but obviously one needs to be certain
that the emission line in the `continuum' filter is sufficiently weak.
In earlier documentation, two curves for 2.17¯m filters are given. The 1987
curve has a peak transmission of 0.69 and a FWHM of 0.0286¯m; the 1992 curve
has a peak transmission of 0.80 and a FWHM of 0.0224¯m. On the strength of
spectra taken through various filters with GRIM II, it is believed that the filter
in GRIM II corresponds to the 1987 curve.
5.4 Bright Limits
As discussed in x3.5, the detector saturates if it is exposed above a certain level.
In the minimum exposure of 1 second, this occurs at a signal of 13000 DN above
the bias. This sets a bright limit for the instrument.
With the f=5 camera and 1 arcsec seeing, at most about 20% of the light from
a point source lands in a single pixel. Using the zero points and backgrounds
from x5.3.2, the bright limits for the minimum exposure of 1 second are J ú 11:3,
H ú 11:1, K ú 10:4, and K s ú 10:4.
These limits will be 1.5 and 3.0 mag brighter for the f=10 and f=20 cameras
and will change with seeing roughly as 5 log FWHM=arcsec. The ND13 neutral
density filter can be used to allow observations of objects up to about 2.5 mag
brighter at J , 1.5 mag brighter at H , and 1.0 mag brighter at K. (Note the the
nuetral density filters are not neutral; see x5.2.)
5.5 Flat Field Images
Flat field images can be contructed either from sky exposures or lamps on/off
dome exposures. Whichever you elect, it is worth combining a number of expo­
sures to minimize systematic errors and aiming to get a formal signal­to­noise
ratio well in excess of 100.
The advantage of sky flats is that the sky is in the far field of the telescope;
the disadvantages are that the sky is not a continuum source, has a very different
35

Filter Lamp On Off T S=N
DN/s DN/s s
J faint quartz 2750 0 3.6 175
H faint quartz 2850 0 3.5 175
K faint quartz 5300 3700 1.9 45
K 0 faint quartz 3750 1800 2.7 80
K s faint quartz 3750 2100 2.7 65
K dark faint quartz 3250 2900 3.1 15
1.08¯m bright quartz 750 0 13.3 175
1.09¯m bright quartz 800 0 12.5 175
1.24¯m bright quartz 700 0 14.3 175
1.28¯m bright quartz 1300 0 7.7 175
1.58¯m bright quartz 1150 50 8.7 165
1.64¯m bright quartz 1200 0 8.3 175
1.70¯m bright quartz 5650 1200 1.8 130
1.99¯m bright quartz 2250 50 4.4 170
2.12¯m bright quartz 1000 50 10.0 165
2.17¯m bright quartz 1050 100 9.5 155
2.21¯m bright quartz 3050 550 3.3 135
2.25¯m bright quartz 950 250 10.5 120
2.26¯m bright quartz 2350 650 4.3 120
2.30¯m bright quartz 1050 400 9.5 95
2.34¯m bright quartz 2950 1300 3.4 85
Table 5.4: Imaging Dome Flat Characteristics at f=5
36

color from most astronomical objects, and scattered light and thermal emission
from the telescope and instrument are not removed well.
The advantages of lights on/off dome flats are that the source is a continuum,
has a color somewhat appropriate to many astronomical objects (being a roughly
2000K black body), and scattered light and thermal emission from the telescope
and instrument are removed well; the disadvantages are that the source is in
the near field of the telescope and, at APO, dome flats must be taken off the
mirror covers, which may cause additional problems.
At the time of writing, the mirror covers and lamps can only be controlled
by the telescope operator at APO.
Table 5.4 gives some useful information for taking dome flats. The numbers
are for the f=5 camera but can be trivially scaled to the f=10 and f=20 cameras.
Table 5.4 lists for each filter the appropriate lamps (either the bright or faint
quartz lamps), the rates with lamps on and off, the time to get about 10000 DN
with the lamps on, and the formal signal­to­noise ratio S=N in a flat constructed
from a single on/off pair. The backgrounds in the K window will be very
sensitive the the temperature of the telescope; these data were taken with the
telescope at about 4 C.
5.6 Basic Techniques
Infrared imaging is normally performed by imaging the object field and one or
more sky fields at slightly different pointings. The different pointing mitigate
the presense of flat field errors and bad pixels and allow objects in the sky field
to be rejected when the sky exposures are combined. Sky exposures are required
because it is impossible to flat field with sufficient accuracy to detect objects
that are faint compared to the sky.
The actual details of obtaining and combining the images are the subject of
endless debate. Sky exposures need to be taken close in time to the object expo­
sures as the sky background changes rapidly, but exactly how rapidly depends
on the night. It is best to do quick sky subtractions as you observe to get an
idea for this. A typical sequences might be four object exposures interspersed
with four sky exposures in an ABBAABBA sequence.
The sky exposures would have their bias subtracted and then be linearized.
They would be combined to form a single sky image, perhaps with multiplicative
adjustment to the same median and outlier rejection to eliminate stars.
The object exposures would have their bias subtracted and be linearized.
The sky exposure would be subtracted from each. Since the sky is variable
even on short timescales, the object exposures will often have some residual
sky emission which needs to be removed by subtracting a constant. The offsets
between the object exposures need to be determined and the images shifted and
combined.
When combining object images you need to deal with both bad pixels and
spurious values in normally good pixels (from electronic glitches and particle
events). A good method is to use a bad pixel mask to deal with the persistently
37

bad pixels and outlier rejection to deal with spurious values.
38

Chapter 6
Spectroscopy
This chapter is
incomplete.
6.1 Capabilities
GRIM II can be used as a low resolution spectrograph by placing a grism in the
beam. The J , H , and K windows can be selected by using the J , H , and K
filters. The K band is at 3rd order, the H band at 4th order, and the J band at
5th and 6th order, which unfortunately overlap. The resolution can be selected
between –=\Delta– of about 200, 400, and 800 by using the f=5, f=10, and f=20
cameras with 240, 120, and 60¯m slits. The (design) effective focal length of
the 3.5m is 35.24m. Thus, the slits have widths of 1.40, 0.70, and 0.35 arcsec on
the sky. Three 60¯m slits are provided to cover the entire K window. Table 6.1
shows the wavlength coverage of the different configurations.
6.2 Flat Fields
Lights on/off flat fields can be obtained using the bright quartz lamps.
6.3 Wavelength Calibration
There are three common methods for obtaining a wavelength calibration: arc
lamps, nigth sky lines, and emission line nebulae. Helium, neon, and argon
lamps are available. Line lists for these can be found in the CRC. Long ex­
psoures (of order 300s) are required and it is worthwile taking lights on/off
exposures to remove thermal background. A good source of the wavelengths of
airglow lines is Dick Joyce's manual for the KPNO CRSP spectrograph, avail­
able from the KPNO WWW page. For the high resolution work, the airglow
lines are considered unreliable as many are blends, but they should be fine at
the low resolution of GRIM II. Finally, bright emission line nebulae can be used,
especially in the H window where many Br lines are visible. Some observers use
39

Camera & Slit Order 6 Order 5 Order 4 Order 3
f=5 0.844--1.301 1.013--1.561 1.266--1.951 1.688--2.602
f=10 0.958--1.187 1.150--1.424 1.437--1.780 1.916--2.373
f=20 Short 0.970--1.082 1.163--1.298 1.454--1.622 1.939--2.163
f=20 Mid 1.015--1.127 1.217--1.352 1.522--1.690 2.029--2.253
f=20 Long 1.068--1.180 1.282--1.415 1.602--1.769 2.136--2.359
Table 6.1: Wavelength Coverage in ¯m
40

these lines to obtain a good solution in the H window and then use the grating
equation to transfer to other orders.
6.4 Absorption Standards
Good sources of B and A stars for absorption standards are the Bright Star
Catalog, SIMBAD, or the Tycho/Hipparchos Input Catalog. Kurucz stellar
atmospheres can be used to model their intrinsic spectra. Late­type stars are
sometimes used for programs that require good cancelation in the vicinity of
hydrogen lines (see the discussion in the appendix of Hanson, Conti, & Rieke
1997). Alternatively, one can simply interpolate over the Brfl feature.
41

Chapter 7
Extinction
The atmospheric extinction above 1¯m results largely from molecular absorption
and shows significant variations as the water vapour content of the atmosphere
changes from season to season, night to night, and even hour to hour. MOD­
TRAN3 model transmittances at the zenith for APO in the summer and winter
are shown in Figure 7.1. These models are the standard mid­latitude summer
and winter atmospheres with no aerosol contribution. MODTRAN3 is the latest
atmosphere radiation transfer code from the Air Force Phillips Laboratory. It
is freely available from
ftp://146.153.100.3/pub/chet/
Most of the absorption results from molecular bands and many are saturated
or semi­saturated. This means that the familar linear extinction law
m(X) = m(0) + CX (7.1)
is invalid to some degree, as it assumes that the optical depth at all frequencies
scales linearly with airmass. Instead, the true extinction law is non­linear. This
problem is discussed in detail by Manduca & Bell (1979) and Young, Milone,
& Stagg (1994). For normal differential photometry between 1 and 2 airmasses
the departures fron non­linearity are small (less than 1%) but increase at higher
airmasses; for absolute photometry the non­linearity can be much larger (tens of
%). It is probably wise to avoid standard or object exposures above an airmass
of 2.
Because the transmittance has so much structure, one cannot na¨ively apply
extinctions derived by broad band observations to narrow band observations. If
you do not wish to derive extinction coefficients for each narrow band filter you
use, you should observe standards and objects at similar airmasses.
42

Figure 7.1: Model Atmospheric Transmitances
43

Chapter 8
Standards
8.1 Photometric Standards
Near infrared standards are immature compared to optical standards. There
are many different standard systems (Johnson, Glass, CIT, CTIO, AAO, MSO,
ESO, UKIRT, Carter) not all of which are especially well defined. Glass (1985),
Bessell & Brett (1988), and Carter (1993) discuss these systems and give trans­
formations between them. The differences between systems are smallest at
K and largest at J . Most of these systems were constructed with InSb pho­
tometers in mind and are too bright to use with infrared array cameras on
moderate­aperture telescopes (see x5.4).
There are really only three choices of JHK standards for use with GRIM II: Does anyone have
Ks standards?
the Elias et al. standards, the Carter & Meadows standards, and the Casali &
Hawarden standards. Each has advantages and disadvantages and you will have
to select the one that matches your requirements for photometric accuracy and
easy of use. The only K 0 standards known to the author are the Wainscoat &
Cowie standards.
Observing your standards with a different instrumental configuration (e.g.,
a higher f= cameras or the neutral density filters) will probably introduce sys­
tematic errors in your photometry.
8.1.1 Elias et al. JHK, CO, and H 2
O Standards
For many years the most commonly used standards suitable for the northern An accurate
measurement of
the
transformation
would be useful.
hemisphere have been those of Elias et al. (1982). These standards are on
the CIT system, suitable for both hemispheres, and have a reasonable range
of color. Unfortunately, even the `faint' standards have K ú 7 and are too
bright to observe with GRIM II even at f=20 without defocusing. Many of
these standards have high proper motions. Elias et al. also give CO and H 2 O
indices for many of their standards.
44

8.1.2 Carter & Meadows JHK Standards
Carter & Meadows (1995) have defined a slightly fainter set of standards on the
Carter system with 8 ! K ! 10. These standards cover a range of colors. Al­
though most of these standards are southern, some are located near the equator
and are faint enough for use with GRIM II at f=10 or f=20
8.1.3 Casali & Hawarden JHK Standards
Casali & Hawarden (1992) have defined a set of truly faint standards on the
UKIRT system. These standards are also known as the `UKIRT faint standards'.
Most of the standards have K ú 11 and as such can be used with GRIM II
without defocusing. Unfortunately, almost all the standards have colors close to
zero, making the determination of the transformation between the instrumental
system and the standard system impossible using these standards alone. Beware
that FS24 is known to be variable (Landolt 1990). Coordinates, magnitudes,
and finding charts are available as
ftp://oldp.nmsu.edu/pub/alan/ukirt.ps.Z
8.1.4 Wainscoat & Cowie K 0 Standards
K 0 magnitudes are different from K magnitudes, partly because the filter has
a shorter central wavelength and partly because it is less sensitive to the CO
bands and more sensitive to the H 2 O band in late­type stars. Wainscoat &
Cowie (1992) define the K 0 system and give K 0 magnitudes for 18 of the Elias
faint standards. They derive
K 0 \Gamma K = (0:22 \Sigma 0:03)(H \Gamma K) (8.1)
for unreddened dwarf stars. This transformation is not necessarily valid for
giants, reddened stars, or galaxies. Nevertheless, some idea of the possible errors
can be gained by applying this equation to the objects under consideration.
Consider the case of observing the Casali & Hawarden standards (which have
H \Gamma K ú 0) and nearby galaxies (which have H \Gamma K ú 0:3). The galaxies are
expected to be 5--10% fainter in K 0 relative to the standards than they are in
K.
8.1.5 Manufactured K 0 Standards
One possible solution to the lack of K 0 standards is to `manufacture' a set from
a set of JHK standards using the transformation given by Wainscoat & Cowie
(1992).
8.1.6 Absolute Calibration
Bessell & Brett (1988) derive an absolute calibration for JHK from the Vega
optical absolute flux measurement and a model atmosphere. Campins, Rieke,
& Lebofsky (1985) also derive a calibration using the solar analog method.
45

8.2 Spectrophotometric Standards
There are no empirical spectrophotometric standards in the near infrared; all
rely on model atmospheres to a greater or lesser extent.
8.2.1 Bohlin Spectrophotometric Standards
Bohlin (1996) presents four DA white dwarf photometric standards for use in
the near infrared. The stars have K ú 13. Model atmosphere calculations
provide the relative fluxes and optical photometry sets the absolute flux. The
accuracy of these standards in the near infrared is not well known, but some
indication is given by the 3.5% disagreement between the two sets of models
considered by Bohlin. Finding charts are presented by Bohlin, Colina, & Finley
(1995). Note that HZ 43 has a close, red companion at a separation of only 3
arcsec; this probably makes it unsuitable for use as a near infrared standard for
ground­based observations.
8.2.2 Manufactured Spectrophotometric Standards
Again, spectrophotometric standards can be `manufactured' using broad band
photometry, an absolute photometric calibration, and a model atmosphere or
an observation from a spectral atlas. The effective temperature scale of Code
et al. (1976) and Kurucz model atmospheres can be used with early­type stars.
The model solar spectrum presented by Colina, Bohlin, & Castelli (1996) can
be used with solar analogs (e.g., those of Hardorp 1978, 1980, and 1982).
8.3 Stellar Colours
The following studies of stellar colours may be useful:
Koorrneef (1983)
early­ and late­type dwarfs, giants, and supergiants
Bessell & Brett (1988)
late­type dwarfs and giants
Leggett (1992)
late­type dwarfs
8.4 Stellar Spectra
The following studies of stellar spectra may be useful:
Kleinmann & Hall (1996)
R ú 3000; K; late­type stars
46

Arnaud, Gilmore, & Collier Cameron (1989)
R ú 100; K; late­type stars
Terndrup, Frogel, & Whitford (1991)
R ú 1000; K; local and bulge M giants
Lan¸con & Rocca­Volmerange (1992)
R ú 500; HK; early­ and late­type stars
Kirkpatrick et al. (1993)
R ú 300; J ; M dwarfs
Origlia, Moorwood, & Oliva (1993)
R ú 1500; H ; late­type stars
Ali et al. (1995)
R ú 1350; K; late­type dwarfs
Wallace et al. (1996)
R ú 45000; JHK; Sun
Wallace & Hinkle (1996)
R ú 300000; K; late­type stars
Hanson, Conti, & Rieke (1996)
R ú 1000; K; early­type stars
Colina, Bohlin, & Castelli (1996)
R ú 1250; JHK; Sun; largely based on a Kurucz model
Leggett et al, (1996)
R ú 250; JHK; late­type dwarfs
47

Chapter 9
Headers
Certain information is placed into GRIM II images headers by the MC. The
most useful header values are summarized in Table 9.1.
48

Name Meaning
IMTITLE Image name (without the .fits, .hhh, or .hhd extension)
OPENTIME Exposure time \Gamma 0:901 seconds (see x3.2)
STARTIME Time exposure started (local time)
STOPTIME Time exposure stopped (local time)
UT Time exposure was commanded (UTC)
LT Time exposure was commanded (local time)
UTDATE Date of exposure (UT)
RA Right Ascension
DEC Declination
EPOCH Epoch of RA and DEC
RAOFF Object offset in Right Ascension (arcsec)
DECOFF Instrument offset in Declination (arcsec)
X Instrument offset in y (arcsec)
Y Instrument offset in y (arcsec)
ROTATION Instrument rotation (degrees)
ZD Zenith distance
AIRMASS Airmass
MODE Mode value from Table 4.1
SCALE Scale value from Table 4.2
GFILTER Filter value from Table 4.3
Table 9.1: Header Values
49

References
Ali, B., Carr, J. S., DePoy, D. L., Frogel, J. A., & Sellgren, K. 1995, AJ, 110,
2415
Arnaud, K. A., Gilmore, G., & Collier Camerson, A. 1989, MNRAS, 237, 495
Ashley, M. C. B., Burton, M. G., Storey, J. W. V., Lloyd, J. P., Bally, J.,
Briggs, J. W., & Harper, D. A. 1996, PASP, 108, 721
Bessell, M. S., & Brett, J. M. 1988, PASP, 100, 1134
Campins, H., Rieke, G. H., Lebofsky, M. J. 1985, AJ, 90, 896
Colina, L., Bohlin, R. C., & Castelli, F. 1996, AJ, 112, 307
Bohlin, R. C., Colina, L., & Finley, D. S. 1995, AJ, 101, 1316
Bohlin, R. C. 1996, AJ, 111, 1743
Carter, B. S. in Precision Photometry, eds. D. Kilkenny, E. Lastovica, & J. W.
Menzies, (SAAO: Cape Town), p. 100
Carter, B. S., & Meadows, V. S. 1995, MNRAS, 276, 734
Casali, M., & Hawarden, T. 1992, JCMT­UKIRT Newsletter, 3, 33
Code, A. D., Javis, J., Bless, R. C., Hanbury Brown, R. 1976, ApJ, 203, 417
Elias, J. H., Frogel, J., Matthews, K., & Neugebauer, G. 1982, AJ, 87, 1029
(with erratum in AJ, 87, 1893)
Glass, I. S. 1985, Irish AJ, 17, 1
Hanson, M. M., Conti, P. S., & Rieke, M. J. 1986, ApJS, 107, 281
Hardorp, J. 1978, A&A, 63, 383
Hardorp, J. 1980, A&A, 88, 334
Hardorp, J. 1982, A&A, 105, 120
50

Kirkpatrick, J. D., Kelly, D. M., Rieke, G. H., Liebert, J., & Allard, F., &
Wehrse, R. 1993, ApJ, 402, 643
Kleinmann, S. G, & Hall, D. N. B. 1986, ApJS, 62, 501
Koornneef, J. 1983, A&A, 128, 84
Lan¸con, A., & Rocca­Volmerange, B. 1992, A&AS, 96, 593
Landolt, A. U. 1990, PASP, 102, 1382
Leggett, S. K. 1992, ApJS, 82, 351
Leggett, S. K., Allard, F., Berrimen, G., Dahn, C., C., & Hauschildt, P. H.
1996, ApJS, 104, 117
Manduca, A., & Bell, R. 1979, PASP, 91, 848
Nguyen, H. T., Rauscher, B. J., Severson, S. A., Hereld, M., Harper, D. A.,
Loewenstein, R. F., Mrozek, F., & Pernic, R. J. 1996, PASP, 108, 718
Origlia, L., Moorwood, A. F. M., & Oliva, E. 1993, A&A, 280, 536
Terndrup, D. M., Frogel, J. A., & Whitford, A. E. 1991, ApJ, 378, 742
Wainscoat, R. J., & Cowie, L. L. 1992, AJ, 103, 332
Wallace, L., & Hinkle, K. 1996, ApJS, 107, 281
Wallace, L., Livingston, W. Hinkle, K., & Bernath, P. 1996, ApJS, 106, 165
Young, A. T., Milone, E. F., & Stagg, C. R. 1994, A&AS, 105, 259
51

Index
absorption standards, 41
aperture stop, 3
bad pixels, 12
bias
exposure, 6
IRAF bias, 6
variability, 6
camera, 3
camera carousel, 3
collimator lens, 3
dark
current, 6
exposure, 6
detector, 3, 5
bad pixels, 12
bias, 6
columns 128 and 256, 14
full well, 12
gain, 5
linearity, 6
non­linearity, 6
orientation, 3
peculiarities, 14
read noise, 5
residual image, 12
exposure
time, 5
extinction, 42
absorption standards, 41
law, 42
non­linearity, 42
f ratio, 16, 25, 49
field lens, 3
field of view, 25
filter wheel, 3
filters, 3, 16, 21, 49
2.17¯m, 35
characteristics, 25
focus shift, 28
K 0 , 28
K dark , 28
K s , 28
neutral density, 3, 16, 21, 25,
49
transmittances, 25
flat fields, 6, 35, 39
dome flats, 35
sky flats, 35
wavelength dependence, 6
focus, 19, 22, 28
FTP, 16, 20
grism, 3, 16, 21, 39, 49
grism wheel, 3
hangs, 24
header value, 48
DEC, 19, 49
DECOFF, 19, 49
EPOCH, 49
GFILTER, 16, 49
IMTITLE, 49
LT, 49
MODE, 16, 49
OPENTIME, 5, 49
RA, 19, 49
RAOFF, 19, 49
ROTATION, 19
ROTATTION, 49
SCALE, 16, 49
52

STARTIME, 49
STOPTIME, 49
UT, 49
X, 19, 49
Y, 19, 49
ZD, 49
imaging, 25
instrumental polarization, 3
lamps
quartz, 37
linearity, 6
mailing list, 2
MC, 5, 6, 15, 24
center procedure, 21
commands, 16
config procedure, 21
diskname command, 19
documentation, 15
east procedure, 21
end procedure, 21
example session, 23
exp procedure, 21
fexp procedure, 22
filetype command, 19
filter procedure, 21
focus command, 19
grimabort command, 16, 24
grimmove command, 16
grimstatus command, 16
home procedure, 22
imdir command, 19
inst command, 16
loginftp command, 20
mode procedure, 21
nexpose command, 16
north procedure, 21
obj1 procedure, 21
obj4 procedure, 22
offset command, 19
places command, 19
priority command, 16
procedures, 15, 20
rotate command, 19
scale procedure, 21
seq command, 19
session, 15
slew command, 16
south procedure, 21
start procedure, 21
status display, 15
stop command, 19
subscribe command, 20
west procedure, 21
mcnode UNIX command, 15
mode, 16, 21, 49
MODTRAN3, 42
non­linearity, 6
optical layout, 3
orientation, 25
pixel scale, 25
REMARK, 5, 15, 24
documentation, 15
residual image, 12
response uniformity, 6
saturation, 12
scale, 16, 21, 49
shutter, 5
slit wheel, 3
spectroscopy, 39
standards, 44
absolute calibration, 45
Bohlin, 46
Carter & Meadows, 45
Casali & Hawarden, 45
CO, 44
Elias, 44
H 2 O, 44
JHK, 44, 45
K 0 , 45
manufactured, 45
photometric, 44
spectrophotometric, 46
Wainscoat & Cowie, 45
stellar colours, 46
stellar spectra, 46
53

TCL, 20
user's manual, 1
wavelength calibration, 39
54