ARC 3.5m | GIFS

Table of Contents

1. Instrument Description 1.1. Imaging 1.2. IFS 1.3. Comparison with Other APO Facility Instruments 2. Instrument Operation 2.1. Software Control 2.2. GIFS Configuration 2.3. Changing Mounted Filters 2.4. Using GIFS Presets 2.5. Taking GIFS exposures 2.6. Direct Imaging with GIFS 2.7. IFS Operations 2.8. Focusing the Telescope 2.9. Checking Instrument Focus 2.10. Slewing and Offsetting 3. Instrument Performance 3.1. Detector Summary with Leach (analog) electronics

3.2. Detector Layout/Cosmetics 3.3. Instrument Sensitivity 3.4. Typical Exposure Times 4. Instrument Calibration 4.1. Calibration Lamps and Controls 4.2. Direct Imaging Calibration 4.3. Wavelength Calibration 4.4. Flux Calibration 5. Instrument Issues 6. End of Your Run 7. Data Storage and Retrieval 8. Data Reduction 9. Appendices A. Instrument Design B. Scripting Commands C. Sample Data D. GIFS Instrument History

1. Instrument Description

The Goddard IFS and Imager (GIFS) is a visible-wavelength multi-purpose imager and imaging spectrograph for use in natural seeing at the Apache Point Observatory 3.5m telescope. As of May 2014, the instrument is available in direct imaging, coronagraphic imaging (using a wedge), and integral field spectroscopy modes. The figure shows the instrument viewed from behind together with Bruce Woodgate.

The instrument is used at the Nasmyth 2 f/10.3 focus of the APO 3.5 m telescope. Behind the telescope focus, a field lens and collimator lens collimate the light through a clear optical path, or through a lenslet array for an integral field unit. The instrument focus is wavelength-dependent, with different collimator positions to ensure collimation. A filter or other element restricting the wavelength range imaged should always be in the optical path. Finally, a camera lens refocuses the light onto the CCD, which currently can be used in a conventional analog mode. Future plans include installation of a photon-counting EMCCD camera system for photon-counting and lucky imaging.

1.1. Imaging

Table 1.1: GIFS Imaging Modes

Mode	Spatial Sampling	Wavelength Coverage	Spectral Resolution
Filter Imaging	0.165"/pixel	4,000 - 10,000 Å	Set by filter
Coronagraphic Imaging	0.165"/pixel	4,000 - 10,000 Å	Set by filter

GIFS uses 2” round filters, the details of which are presented in Tables 1.2-1.5. Filters noted as degraded have defects near the edge, which should be monitored for changes with time. Where known the manufacturer is specified to aid in locating filters in the storage boxes. Up to six filters can be mounted in the filter wheel at a time (Section 2.3). Please notify the observatory of your filter requests at least 24 hours in advance of your run (no later than the preceding Friday morning, if observing over a weekend). When using the IFS, the corresponding blocking filters must go in the filter slots, limiting the slots available for direct imaging.

Table 1.2: GIFS Broad Band Filters

Filter	Wavelength (Å)	Width (Å)	Quality
B	4400	1040
V	5300	986
R	6000	1545
I	8000	3136	Degraded 5mm from edge

Table 1.3: GIFS Medium Band Filters

Name	Measured		Manufacturer	Transmission	Quality
	Central Wavelength (Å)	FWHM (Å)
filt4015	4020	50	Andover	48%
filt4050a	4050	160
filt4050b	4050	57	Andover	47%
filt4085	4086	58	Andover	48%
filt4120	4124	52	Andover	49%
filt4155	4156	50	Andover	46%
filt4160	4160	160
filt4200	4206	108	Andover	55%
filt4257	4253	102	Barr	64%
filt4300	4313	116	Andover	60%	Degraded 8mm from edge
filt4400	4411	120	Andover	57%	Degraded 15mm from edge
filt4500	4507	122	Andover	66%	Degraded 15mm from edge
filt4600	4596	129	Andover	69%	Degraded 10mm from edge
filt4700	4701	109	Andover	76%	Degraded 10mm from edge
filt4760	4756	100	Barr
filt4800	4816	107	Andover	77%	Degraded 10mm from edge
filt4865	4864	91	Barr	81%	Degraded 8mm from edge
green_ifs	4893	317		99.06%	Green IFS blocker
filt4900	4909	105	Andover	72%	Degraded 4mm from edge
filt4982	4982	15			Degraded 5mm from edge
filt4994	4994	15			6mm, just one spot
filt5000	5007	106	Andover	74%	Degraded 4mm from edge
filt5007a	5007	20			Degraded 6mm from edge
filt5010	5011	106	Barr	74%	Degraded 3mm from edge
filt5020	5020	15			Degraded 5mm from edge
filt5032	5032	15			Degraded 8mm from edge
filt5100	5111	106	Andover	76%	Degraded 7mm from edge
filt5200	5210	116	Andover	72%	Degraded 10mm from edge
filt5265	5263	115	Barr	75%	Degraded 3mm from edge
filt5300	5304	86	Andover	71%	Degraded 5mm from edge
filt5340	5343	132	Barr	86%	Degraded 10mm from edge
filt5400	5406	124	Andover	74%	Degraded 5mm from edge
filt5500	5501	94	Andover	75%	Degraded 2mm from edge
filt5552	5549	138	Barr	77%	Degraded 8mm from edge
filt5600	5605	139	Andover	74%	Degraded 6mm from edge
filt5700	5703	139	Andover	74%	Degraded 2mm from edge
filt5800	5806	101	Andover	74%	Degraded 1mm from edge
filt5852	5845	154	Barr	88%
filt5900	5902	109	Andover	76%	Degraded 1mm from edge
filt6000	6002		Andover	79%	Degraded 3mm from edge
filt6052	6062	120	Barr	86%
filt6056	6056	100	Andover		Degraded 5mm from edge
filt6100	6103	111	Andover	76%
filt6200	6194	106	Andover	72%	???missing from Dave's list
filt6245	6235	117	Barr	94%
filt6300	6307	126	Andover	69%	Degraded 3mm from edge
filt6400	6403	116	Andover	68%	Degraded 2mm from edge
filt6455	6450	119	Barr	96%
filt6500	6512	122	Andover	74%	Degraded 10mm from edge
halpha_a	6564	10
halpha_b	6563	117	Barr	92%	Halpha-b
red_ifs	6572	265		98 or 96%?	IFS red blocker
filt6600	6602	124	Andover	60%	Degraded 2mm from edge
filt6700	6698	124	Andover	60%
filt6702	6702	124	Andover	63%
filt6724	6724	73	Barr	?
filt6750	6750	200
filt6766	6766	138	Barr	86%
filt6800	6811	99	Andover	71%	Degraded 3mm from edge
filt6900	6891	102	Andover	62%	Degraded 3mm from edge
filt6940	6945	138	Barr	96%
filt7000	7000	103	Andover	68%
filt7100	7106	176	A-wd	95%
filt7230	7230	190	Omega
filt7240	7240	177	??	95%
filt7360	7350	185	Andover	90%
filt7480	7499	185	Andover	89%	Degraded 4mm from edge
filt7650	7650	212	Andover	92%
filt7800	7785	218		98%	One side degraded
filt7950	7964	235	Andover	96%
filt8100	8101	223	Barr	96%	Degraded 5mm from edge
filt8240	8228	206	Barr	96%	Degraded 6mm from edge
filt8380	8368	244	Barr	92%
filt8520	8502	208	Barr	93%
filt8660	8668	218	Barr	94%	Degraded 8mm from edge
filt8840	8829	262	Barr	92%
filt9020	9013	251	Barr	97%	Degraded 7mm from edge
filt9200	9200	316	CS	85%
filt9380	9364	256	Barr	95%
filt9560	9580	266	Barr	95%
filt9770	9775	331	Barr	93%
filt9980	9978	329	Barr	90%
filt10190	10220	348	Barr	94%
filt10500	10500	200	SF
filt10830	10830	200	A-wd

Table 1.4: GIFS Blocking Filters for High Resolution

Name	Central Wavelength (Å)	FWHM (Å)	Manufacturer	Notes
filt4861	4861	38	Barr	Hβ
filt5007b	5007	40		OIII
halpha_c	6563	63	Barr	at GSFC
filt6724	6724	73	Barr

Table 1.5: Other 2" Round Filters Compatible with GIFS

Name	Central Wavelength (Å)	FWHM (Å)
York5059	5060	12
430-610nm/Interference	5200	1800
York5202	5202	13
York5685	5685	14
York5720	5720	14
CU5893	5893	15
York5917	5915	20
York6007	6008	15
York6178	6008	15
York6409	6008	15
York6569	6569	15
CU6731	6731	15
York6903	6903	17
York6931	6929	17
600-940nm/Interference	7700	3400
York7740	7740	19
York8212	8207	20
York8238	8236	20
York8678	8662	21
York9088	9089	22

Observers wishing to use their own filters (2” diameter, round) should contact Bill Ketzeback at APO.

1.2 IFS

GIFS can also be configured as an IFS with choice of 14” x 14” or 7” x 7” fields of view. The IFS operates in the green and red (λ ≤ 7000 Å). A slot partially illuminating the lenslet array can be manually installed in the instrument to enable broader wavelength coverage over a reduced field of view.

Table 1.6: GIFS Integral Field Spectroscopy Capabilities

FOV (arcsec x arcsec)	Spatial Resolution (arcsec/spaxel)	Spectral Bandwidth (Å)	Spectral Sampling (Å)	Central Wavelength (Å)	Spectral Coveragea (Å)	Modeb	Lines in Range
BLUE/GREEN grating R~1475
7" x 7"	0.215	310	1.05	4890	4740-5040	NSG	Hβ, [O III]
14" x 14"	0.43	310	1.05	4890	4780-5040	WSG	Hβ, [O III]
7" x 1.68"c	0.215	107	1.05	4890	4330-5400	NLG	Hβ, [O III], Fe XIV, Ar IV, He I, Cl III, Fe III, He II, Ne IV
14" x 3.36"	0.43	107	1.05	4890	4330-5400	WLG	Hβ, [O III], Fe XIV, Ar IV, He I, Cl III, Fe III, He II, Ne IV
RED grating R~1823
7" x 7"	0.215	420	1.44	6640	6430-6850	NSR	Hα, [N II], [S II]
14" x 14"	0.43	420	1.44	6640	6430-6850	WSR	Hα, [N II], [S II]
7" x 1.68"c	0.215	1250	1.44	6640	5750-7000	NLR	+[O I], [S III], FeX, He I
14" x 3.36"	0.43	1250	1.44	6640	5750-7000	WLR	+[O I], [S III], FeX, He I

^aSpectral Regions adjacent the short spectral ranges may be observed by choice of blocking filters (see Tables 3.2 - 3.5).
^bMode descriptors: N or W for narrow or wide field; S or L for short or long spectrum; R or G for red or green band.
^cRequires manual installation of a field stop for IFS mode, and therefore is only available for on-site observers.

The bare lenslet array has higher throughput than the pinhole array, but has some light scattered and diffracted into the inter-spaxel region. Each spaxel illuminates more than one pixel, but spaxels illuminate fewer pixels with the pinhole array in place than with the bare array in place. The ratio of pinhole/bare throughput is 77% at the peak pixel, but when light from a given spaxel is integrated over all the pixels, the throughput ratio is 72%. This loss in transmission is due to the mask blocking scattered light outside the core. The bare lenslet array also has a smudge in the lower left quadrant, and the team is working to correct it. In the meantime, observers are advised to keep targets out of the affected area as necessary.

1.3 Comparison with Other APO Facility Instruments

As an imager, GIFS has a circular field of view which is only ~60% that of SPICAM. The main strength of GIFS as an imager in analog mode is the large suite of medium-band filters (FWHM=135Å) and the availability of the coronagraphic wedge, or in the future as an imager with photon-counting and lucky imaging modes. As an IFS, GIFS offers contiguous 2-D spatial coverage of a field which is larger than the DIS slit widths, with higher spectral resolution than DIS, but with wavelength coverage limited to the green and red. DIS offers more complete (or user-specified) wavelength coverage, and simpler target acquisition and data analysis, at the expense of less complete spatial sampling near an object of interest.

Table 1.7: Comparison with Other APO Facility Instruments

Instrument	Field of View	Filter Complement	Spectral Resolution
SPICAM	4.7'	SDSS u'g'r'i'z' MSSSO UBVRI	Few; set by bandpass
GIFS Imaging	1.7' x 1.7' with some vignetting	BVRI, medium band	~2-3 for broadband ~20 grating blockers 48 medium band, red 70 Rutgers broad blockers and higher for a few filters
DIS	1.5", 2", or 5" x 6'	Long-slit	300-400, 1200 0.4 - 1.0 microns simultaneously
GIFS as IFS	14" x 14" or 7" x 7"	IFS	R = 1475 (green) R = 1823 (red) R = 5000 (hi-res)

2 Instrumentation Operation

2.1 Software Control

Software Control for GIFS is provided by the Telescope User’s Interface (TUI) package written by Russell Owen (UW). Commanding for GIFS requires version 2.3.1 or later. You can check which version you are running by pulling down TUI>>About TUI on the main menu bar.

At startup, TUI produces a selection of windows to control the telescope, the science instrument, and the calibration lamps, to provide messaging with the observing specialist, and command line commanding of the science instrument.

Figure 2.1.1: TUI Introduction

TUI at startup	Telescope pointing and status information is located on the Status window.

2.2 GIFS Configuration

Using the "Inst" pull down menu on the menu bar, select GIFS (Inst>>GIFS). This brings up the GIFS configuration window. You should see a display similar to Figure 2.2.1a with a list of mechanisms on the left and values on the right.

Figure 2.2.1: GIFS configuration window.

Current instrument status should be shown.	If this is what you see upon launching GIFS, the GIFS actor needs to be restarted; ask the observing specialist

If no data are shown (Figure 2.2.1b), and where they should appear is pink, you will need ask the observing specialist to restart the GIFS actor.

Magnifier, lenslet, and disperser values are normally changed using preset instrument configurations, which can be accessed using the "Presets" pull-down menu, while the filter in the optical path can be changed using the "Filter Wheel" pull-down menu.

Changes to configurations are first selected, at which point they will have a colored background. Actually moving the mechanism is done using the "Apply" button at the bottom of the GIFS window. At this point, you will see count down timers appear and count down as the mechanisms individually move.

The GIFS initial default is that the "Config" box is not checked. To change the instrument configuration, tick the "Config" box.

2.3 Changing Mounted Filters

The six filters you requested to be mounted are listed in the "Filter Wheel" pull-down menu and are named as specified in the confirmation email from the observatory. Click on the filter you want, and then press "Apply." Wait until the countdown timers have counted down and there is no colored background to the filter pull-down menu before changing other configurations. Filter changes normally result in automatic changes to the collimator setting to compensate for the change in instrument focus.

2.4 Using GIFS Presets

Configuring GIFS for either filter imaging or IFS spectral imaging requires coordinated movement of several of the mechanism drives. Present configurations for the most heavily used of these modes have been defined and can be accessed via the "Presets" pull-down menu on the GIFS window. These include options for direct imaging with and without a coronagraphic wedge in the optical path, Fabry-Perot operation (not currently available), and use of the IFS with its choice of 2 magnifications (14/7), 2 lenslet arrays (bare/pinhole), and imaging or spectroscopic modes. The syntax of the presets is generally:
mode,magnification,lenslet,disperser,collimator

Table 2.1: Preset Instrument Modes

Preset Command	Mode	Use
Direct_imaging,clear,collg	Direct imaging	Initial target acquisition in green
Direct_imaging,clear,collr	Direct imaging	Initial target acquisition in red
Direct_imaging,wedge,collg	Direct coronagraphic imaging	Coronagraphic imaging in green
Direct_imaging,wedge,collr	Direct coronapgraphic imaging	Coronagraphic imaging in red
fp,wedge,collg	Fabry-Perot imaging	Not available at this time
fp,wedge,collr	Fabry-Perot imaging	Not available at this time
fp,wedge,collg	Fabry-Perot coronagraphic imaging	Not available at this time
fp,wedge,collr	Fabry-Perot coronagraphic imaging	Not available at this time
ifs,14,bare,clear,collg ifs,7,bare,clear,collg	IFS acquisition (no disperser, no pinhole mask)	Source placement in field of view in green
ifs,14,pinhole,clear,collg ifs,7,pinhole,clear,collg	IFS acquisition (no disperser, with pinhole mask)	Source placement in field of view in green
ifs,14,bare,clear,collr ifs,7,bare,clear,collr	IFS acquisition (no disperser, no pinhole mask)	Source placement in field of view in red
ifs,14,pinhole,clear,collr ifs,7,pinhole,clear,collr	IFS acquisition (no disperser, with pinhole mask)	Source placement in field of view in red
ifs,14,bare,green ifs,7,bare,green	IFS spectrum (disper in place, no pinhole mask)	Spectral imaging in green
ifs,14,pinhole,green ifs,7,pinhole,green	IFS spectrum (disper and pinhole mask in place)	Spectral imaging in green
ifs,14,bare,red ifs,7,bare,green	IFS spectrum (disperser in place, no pinhole mask)	Spectral imaging in red
ifs,14,pinhole,red ifs,7,pinhole,red	IFS spectrum (disperser and pinhole mask in place)	Spectral imaging in red

2.5 Taking GIFS exposures

GIFS exposures can be specified by clicking on the Expose button at the bottom of the GIFS window. The process is similar to other instruments at the ARC 3.5m telescope. This launches the GIFSExpose window. You can specify the image type (object, flat, dark, or bias), the integration time, the number of exposures, the name and a comment to be added to the FITS header. Files with names of the form "name.####.fits," where "####" is a running number, will be created at the end of the integration in a directory named with the current UT date. Files are temporarily archived at APO and are downloaded to a path on your machine which can be selected under TUI>>Downloads>>.

2.6 Direct Imaging with GIFS

GIFS supports direct filter imaging using the filter list in Section 1.1, with up to six filters mounted at a time. If you plan to combine direct imaging with IFS use, you will need to allocate a filter slot for blocking filters for each IFS grating that you will be using. Direct images can be made in either direct imaging mode or coronagraphic imaging mode.

In general, the process is:

1. Slew to your target and request guiding.
2. Select filter and configure instrument for imaging using one of the "direct_imaging" presets.
3. Initial acquisition of target.
4. Offset to place target at preferred location in field of view (as needed).
5. Expose as required.

2.7 IFS Operations

The IFS component consists of two sections, the pre-formatter optics and the spectrograph. The pre-formatter section magnifies the image from the telescope focus onto a microlens array, so that the lenslets act as spatial samples in an image plane. This design follows the TIGER design concept (Bacon, R., et al. 1988, “The Integral Field Spectrograph TIGER”, in ESO conference “VLTs and their instrumentation”, Munich, March 1988.), similar to OSIRIS at Keck (Larkin, J.E., et al. 2003, “OSIRIS: Infrared integral field spectrograph for the Keck adaptive optics system” SPIE, 4841, 1600.54), where spectra are interleaved between concentrated image points.

The spectrograph uses existing GIFS optics such as the field lens, collimator lens, band-limiting filters in the existing filter wheel, and a camera to focus the images onto a back-illuminated EMCCD. The spectrograph portion of the IFS is located on the disperser drive behind the filter wheel in the parallel beam section of the instrument, and consists of three IFS dispersers, one for the red_ifs, one for the green_ifs and one for a higher resolution red ifs (hires), as well as supporting a clear path for direct imaging. Each IFS disperser consists of a transmissive grating, between two prisms to return the dispersed light to its original path in both rotation and translation. The gratings use Volume Phase Holograph technology for transmission and high throughput.

The lenslet arrays concentrate light into image spots that act as the input to the spectrograph. The grid of image spots can be observed without dispersion, as in the target acquisition sequence for the IFS, or dispersed such that the spectra are spatially separated on the detector. At most, we can have 32x32 spatial samples, 1024 spectra, on the detector in a single frame. The lenslets are on a regular grid, and sample the image plane like a detector. In analogy with detector pixels, the lenslet images are termed “spaxels” (i.e., spatial elements) in the rest of this document.

The observer may choose between two magnification modes: 14"x14" or 7"x7". For most purposes, the larger magnification is better, due to the typical APO seeing limit and accuracy of offset maneuvers. On a night with particularly good seeing or on targets that are bright, and not read-noise-limited, the 7"x7" may be preferable.

Integral field spectroscopy with GIFS is more complex than direct imaging. In general, the data acquisition process is:

1. Slew to your target and request guiding.
2. Select filter and configure instrument for imaging using the blocking filter that you will be using with the IFS grating.
3. Take initial direct image.
4. Use the "Offset" window to offset pointing to put the target at (x,y)=(420,496) in the direct imaging field of view. Confirm you moved to the correct location on the detector with another direct image (and the help of an observing specialist).
5. Configure GIFS for IFS acquisition with your preferred magnification (14" or 7"), choice of lenslet array (bare or pinhole), and green or red collimator.
6. Take an IFS acquisition image ("clear," no disperser in place) to locate your target in the field of view. The IFS field of view is rotated with respect to the direct imaging field by 187 degrees. Exposure durations can be ~2-4x shorter than your initial acquisition image, due to image compression by the spaxels. The orientation of the IFS on the sky is rotated with respect to the direct imaging configuration by 187°, due to a combination of the optics and the orientation of the lenslet array. If you wish, examine the image using the GIFS DRP
7. Configure the instrument for spectral imaging.
8. Take spectral images.
9. Turn on truss lamp for wavelength calibration and spectral trace data.
10. Take wavelength calibration and spectra trace data.
11. Repeat science target observations as needed.
12. Reduce IFS spectral data using the GIFS DRP.

2.8 Focusing the Telescope

The telescope is focused by inspection of images taken in direct imaging mode. Generally, focusing is most efficiently accomplished by asking the Observing Specialist to focus the telescope. The observing specialist can run the GIFS focus script to expedite the process and provide a current seeing estimate.

As with all instruments, monitoring focus is advisable. The telescope focus will change over the course of the night, especially at the beginning of the night when the telescope is changing temperature.

GIFS guiding is performed using the NA2 guide camera. You will need to request guiding.

2.9 Checking Instrument Focus

The GIFS coronagraphic wedge has a small burr on its underside (x,y)=(498,541), (Figure 3.2.1) which we use for collimator focus checks for wavelengths < 9,000 Å

The instrument's focus is wavelength dependent. As a result, the collimator must be adjusted to achieve proper focus for any given filter/lenslet/grating combination. The ICC/TUI will set the focus automatically. If a filter is loaded, and the direct image seems blurry (e.g. can't see the burr), contact Bill Ketzeback at APO. This is particularly important if you are using the filters in table 1.5, or your own filters.

2.10 Slewing and Offsetting

Slews to targets are controlled using the TUI Slew window (Figure 2.10.1), either by manually specifying target coordinates or by providing an ASCII file with target coordinates, which can be selected using the "Catalog" button at the bottom of the window. The default field rotation places E-W in the detector x-direction. This can be altered by specifying a rotation angle in the TUI Slew window.

Figure 2.10.1: Navigating to Targets in TUI

TUI Slew Window	TUI Offset Window

It is necessary to offset from the acquired coordinate location of the target to the IFS aperture “sweet spot” at (x,y)=(420,496). Use the TUI Offset window (Figure 2.10.2) to input either an "Obj Arc" or "Boresight" offset in pixel or angle increments. For conversion from pixel location to offsets in arcseconds, the pixel scale is 0.165” in direct imaging mode. The location of the IFS aperture is not immediately apparent in GIFS direct images, but is just off the tip of the coronagraphic wedge, allowing for another method of properly placing your target on the IFS when the wedge is in place. No matter the absolute coordinate of the sweet spot, the relative offset between the coronagraphic burr and the sweet spot remains constant. From the burr, the sweet spot should be at (δx,δy)=(-78,-45) in pixel coordinates.

3 Instrument Performance

3.1 Detector Summary with Leach (analog) electronics

Table 3.1: Detector summary with Leach (analog) electronics

Detector	e2v CCD201
Serial Number
Number of rows	1024
Number of columns	1024
Pixel Size	13.3μm x 13.3μm
Gain	2.3e-/ADU
Readout noise	2.88e-
Dark Current	1e-/pixel/hour
Linearity Regime	< 30,000 DN
Full Well (=34.78Ke/gain)	80,000 e-
Chip QE 4000 Å 5000 Å 6500 Å 9000 Å	52% 90% 91% 35%
Readout time	18 seconds
Default Analog Mode Timing File	fwd5_0.lod

3.2 Detector Layout/Cosmetics

The optical center of GIFS and the CCD center are not perfectly aligned. There is vignetting of the image, reaching up to 57% light loss compared to the center of the field: this can be compensated with flat field images obtained using the BrQtz lamp. There are no known bad columns or hot pixels as of May 2014. However, there are some dust specks visible. Fringing is seen with very narrow band or very red filters.

Figure 3.2.1: GIFS Detector Response

		x	y
		544	555
		462	376
		176	218
		636	265
		531	868
		731	343
		941	137
Flat field response of detector with and without coronagraphic wedge in place. Note the presence of vignetting.	Close-up of coronagraphic wedge and the burr on the side. The burr is a fortuitous defect helpful in focusing	Dust speckle locations (pixel coordinates).

Figure 3.2.3: GIFS Detector Bias Frames

Example bias frames. Note the striping (and its variation from frame to frame). A bias removal IDL-based script which uses the reference pixels is provided in the data distribution. See the GIFS DRP for more details.

Figure 3.2.2: GIFS Detector Linearity

Detector response plotted against bright quartz truss lamp integration time, As measured for a 51x51 pixel box centered on (x,y)=(425,475). To ensure linearity, exposures should be kept under 34k ADU. Standard deviations for each point plotted are smaller than the plot symbol used. To avoid saturation after lamp replacement in Feb. 2012, integration times should be 75% of the values shown above.

3.3 Instrument Sensitivity

Table 3.3: Representative Integration Times: Direct Imaging

Source Type	Integration Time (s)	Exposure Level (photometric conditions)
BD+28 4211 R=10.6, V=10.5 Red Grating Blocker	5	Peak single pixel: 15k DN over bias Integrated over point source: 1.12e7 DN
DG Tau, T Tauri star R=11.4 Hα	30	Peak single pixel: 22.2k DN over bias Integrated over point source: 9.38e5 DN
Standard Star (HZ 44) V=11.47 Green Grating Blocker	5	Peak single pixel: 6500 DN over bias Integrated over point source: 366k DN

4 Instrument Calibration

4.1 Calibration Lamps and Controls

GIFS uses the truss lamps on the 3.5m for wavelength, flat field, and spectral trace observations. Truss lamp controls are found under Misc>>Truss Lamps on the TUI menu bar (Figure 4.1.1). Individual lamps can be turned on and off independently.

Figure 4.1.1: TUI Truss Lamp Window

4.2 Direct Imaging Calibration

Observers will need bias frames, flat field frames, and observations of standard stars. Bias frames show a 10 ADU gradient top to bottom, and a 4DN gradient along rows, as well as low-level striping. Bias frames are actually 100ms darks, and can contain isolated cosmic ray events. We suggest using a median of an odd number of bias frames as a super bias, and then explore destriping algorithms. Raw analog images have 1072x1032 pixels, including 16 pixels of overscan plus 16 pixels of dark reference (i.e., aluminum covered pixels). On the right hand side of the image are another 16 pixels of overscan. The top of the image has another 8 row-wide dark reference strip. These strips can be used as alternates to dedicated bias frames or dark frames, but will not fully account for the striping. The active image area corresponds to pixels x=33-1056, y=0-1023.

Flat fields are typically obtained using the Bright Quartz truss lamp. Integration times depend on choice of blocking filter, but nominal values are given in Table 4.2.

Table 4.2: Representative Integration Times: Imaging Calibrations

	Red Blocking Filters	Green Blocking Filters
Flat Field -- Bright Quartz Lamp	2s gives 10,000 counts over bias	9s gives 10,000 counts over bias

4.3 IFS Calibration

To reduce IFS data, observers need to take the same calibration data as when performing direct imaging (bias, flat field, and standard star frames). Additionally, you must take IFS acquisition images with the bright quartz lamp to locate the spaxels, wavelength calibration data (He, Ne, and Ar lamps on, disperser in place) with each grating/filter used, and trace observations (at least 200 counts over bias, bright quartz lamp on and dispersing grating in place) with each grating used. Analog IFS observations use a combination of emission line spectra for wavelength calibration and bright quartz observations to locate individual spectra (trace).

Table 4.3: Representative Integration Times: IFS Calibrations

	Red grating	Green grating
Spectral Trace -- Bright Quartz Lamp	7.5s gives 200 counts over background	112s gives 200 counts over background
Wavelength Calibration	Ne 30s gives 2k counts over background S/N=100 in 150s	HeNe 150s gives 450 counts over bias for brightest line S/N=30 in 300s

4.4 Flux Calibration

If you wish to perform relative flux calibration for your spectra, you must observe some spectrophotometric standards. Information about standard stars can be found here.

A TUI catalog of standard stars is described on the above standards page.

Table 4.4: Nominal Integration Times for V=10.5 Continuum Source, Bare Lenslet Array, and 14" x 14" FOV

IFS Acquisition	10s to S/N=100 per spaxel
Red grating	900s to S/N=100 per dispersed pixel
Green grating	600s to S/N=100 per dispersed pixel

5 Instrument Issues

No known hot pixels or bad channels as of August 2014.

6 End of Your Run

Be sure to turn off any calibration lamps you may have used, and then hand over the instrument by simply quitting out of the GIFS instrument control window in TUI. In some circumstances, you may continue to use the instrument after your shift. For example, if you are the first half observer and the scheduled second-half observer is not using GIFS. Ask your observatory specialist for permission to do so if circumstances warrant.

7 Data Storage and Retrieval

Your data will remain on newton.apo.nmsu.edu:/export/images/your program code/UTYYMMDD/ for 9 to 12 months before being automatically deleted. Data can be accessed using scp or sftp to the institutions' observer accounts on newton (you can also use ftp and the images account and password); call APO (575-437-6822) if you are unsure of the correct login.

8 Data Reduction

Data reduction for IFS observations is complex. We have developed an IDL-based software package to handle the conversion of the observations into data cubes, and a manual for its use. Click here to go to the GIFS Data Reduction Pipeline (DRP) Manual

The IDL-based package can be requested by emailing the GIFS Team. It will be available at this site for direct download soon.

9 Appendices

A. Instrument Design

Figure A.1: Instrument Schematic

SolidWorks view of GIFS from bottom (telescope) to detector (top): Box A – (magnifier drive) houses IFS lenslets and associated optics, a coronagraphic wedge for direct imaging and a clear path for conventional filter imaging. B is the manual slider for alternate coronagraphic spots. C is the location of the internal calibration system that is not currently being utilized. Box D houses the collimating lens and its associated focus drive. E is the filter wheel. F is the disperser drive housing IFS VPH gratings, and a clear path for conventional imaging. G is the camera lens. H is the shutter assembly and camera lens mount. J is the CCD in its dewar. K is the Photon ETC controller for the CCD (not currently used), and L is the Leach controller (currently used). The instrument image planes are at the lenslet array (See Box A) for all modes and at the CCD surface. The filter wheel is located near the pupil plane.

Figure A.2: Optical Layout (IFS Mode)

Simplified optical diagram for the low-resolution red IFS mode.

B. Scripting Commands

In TUI, two scripts are available to GIFS users, both found by navigating on the menu bar to Scripts/GIFS/. Both are nodding scripts, utilizing an ABBA nod pattern. One ("Nod_Objarc") uses an "objarc" offshift, while the other ("Nod_boresight") uses a boresight offshift. Users may implement their own personal Python scripts by navigating to Scripts/Open.

C. Sample Data

Figure C.1: GIFS Data Examples

Direct Imaging	IFS Acquisition Mode ("clear")	IFS Spectral Mode ("bare")

Sample FITS Header:

SIMPLE	=	T	/	file does conform to FITS standard
BITPIX	=	16	/	number of bits per data pixel
NAXIS	=	2	/	number of data axes
NAXIS1	=	1072	/	length of data axis 1
NAXIS2	=	1032	/	length of data axis 2
EXTEND	=	T	/	FITS dataset may contain extensions
OBSERVAT	=	'APO'	/	Per the IRAF observatory list.
TELESCOP	=	'3.5m'
INSTRUME	=	'gifs'	/	Instrument name
LATITUDE	=	+3.2780361000000E+01	/	Latitude of telescope base
LONGITUD	=	-1.0582041700000E+02	/	Longitude of telescope base
TIMESYS	=	'TAI'	/	Time system for DATE-OBS
UTC-TAI	=	-35.0	/	UTC = TAI + UTC_TAI(seconds)
UT1-TAI	=	-35.28	/	UT1 = TAI + UT1_TAI(seconds)
LST	=	'12:58:07.518'	/	Local Apperent Sidereal time
OBJNAME	=	'HD189733'	/	Object name, per TCC ObjName
RADECSYS	=	'FK5'	/	Coordinate system, per TCC ObjSys
EQUINOX	=	+2.0000000000000E+03	/	Equinox, per TCC ObjSys
OBJANGLE	=	+1.0460000000000E-03	/	Angle from inst x,y to sky
RA	=	'20:00:43.68'	/	RA hours, from TCC ObjNetPos
DEC	=	'22:42:35.44'	/	Dec degrees, from TCC ObjNetPos
ARCOFFX	=	+2.5000000000000E-03	/	TCC arc offset X
ARCOFFY	=	+5.6100000000000E-04	/	TCC arc offset Y
OBJOFFX	=	+0.0000000000000E+00	/	TCC object offset X
OBJOFFY	=	+0.0000000000000E+00	/	TCC object offset Y
CALOFFX	=	+0.0000000000000E+00	/	TCC calibration offset X
CALOFFY	=	+0.0000000000000E+00	/	TCC calibration offset Y
BOREOFFX	=	+0.0000000000000E+00	/	TCC boresight offset X
BOREOFFY	=	+0.0000000000000E+00	/	TCC boresight offset Y
AIRPRESS	=	+7.3156000000000E+04	/	Air pressure, Pascals
HUMIDITY	=	+3.9000000000000E-01	/	Humidity, fraction
TELAZ	=	-2.5168000000000E+01	/	TCC AxePos azimuth
TELALT	=	+7.9071000000000E+01	/	TCC AxePos altitude
TELROT	=	+5.6264000000000E+01	/	TCC AxePos rotator
TELFOCUS	=	+5.2500000000000E+02	/	TCC SecFocus
ZD	=	+1.0929000000000E+01	/	Zenith distance
AIRMASS	=	+1.0184721546228E+00	/	1/cos(ZD)
CTYPE1	=	'RA---TAN'	/	WCS projection
CTYPE2	=	'DEC--TAN'	/	WCS projection
CRPIX1	=	+5.5090000000000E+02	/	WCS reference pixel
CRPIX2	=	+4.1490000000000E+02	/	WCS reference pixel
CRVAL1	=	+3.0018471800000E+02	/	WCS reference sky pos.
CRVAL2	=	+2.2710405000000E+01	/	WCS reference sky pos.
CD1_1	=	-4.4585731666720E-05	/	WCS (1/InstScaleX)*cos(InstAng)
CD1_2	=	+8.3068183875392E-10	/	WCS (1/InstScaleY)*sin(InstAng)
CD2_1	=	+8.1396353666715E-10	/	WCS (-1/(InstScaleX)*sin(InstAng)
CD2_2	=	+4.5501494716519E-05	/	WCS (1/InstScaleY)*cos(InstAng)
COMMENT		FITS (Flexible Image Transport System) format is defined in 'Astronomy
COMMENT		and Astrophysics', volume 376, page 359; bibcode: 2001A&A...376..359H
BZERO	=	32768	/	offset data range to that of unsigned short
BSCALE	=	1	/	default scaling factor
SECTION1	=	'///////////////// GFP/IFS astrometry information //////////////'
FIMPIX	=	'0.165 arcsec/pix'	/	Filter imaging pixel scale
IFS14SS	=	'0.0139 arcsec/pix'	/	IFS 14x14 spaxel scale
IFS7SS	=	'0.00688 arcsec/pix'	/	IFS 7x7 spaxel scale
IFS14LS	=	'0.43 arcsec/lenslet'	/	IFS 14x14 lenslet scale
IFS7LS	=	'0.215 arcsec/lenslet'	/	IFS 7x7 lenslet scale
FIMSSX1	=	'570 '	/	Filter_imaging_sweet_spot X1
FIMSSY2	=	'480 '	/	Filter imaging sweet spot Y1
IFSDR	=	'172.87 or 187.12 degrees'	/	IFS_Delta_Rotation
SECTION2	=	'//////////////// Science instrument configuration /////////////'
INST	=	'gfp-ifs '	/	Goddard Fabry-Perot-Integral Field Spectrograph
M_POS	=	8003	/	Numerical position of Magnifier
M_TYPE	=	'14x14 '	/	Magnifier preset name
L_POS	=	168980	/	Numerical position of Lenslets
L_TYPE	=	'bare_lens'	/	Lenslets preset name
D_POS	=	727420	/	Numerical position of Disperser
D_TYPE	=	'green '	/	Disperser preset name
OBS_MODE	=	'IFS '	/	Observation Mode - IFS, IFS_ACQ, F-P or CLEAR
F_POS	=	1	/	Position of filter on wheel
FILTER	=	'GREEN_IFS'	/	Filter element type
FWAVE	=	'4893 '	/	Filter center wavelength
F_FWHM	=	'317 '	/	Filter full-width at half max
CALMIRR	=	'OUT '	/	Setting of calibration mirror
SECTION3	=	'////////////// Fabry-Perot parameters ////////////////'
ETALON	=	' '	/	Fabry-Perot etalon
ETALPOS	=	' '	/	Etalon position
FPX	=	' '	/	Fabry-Perot X setting
FPY	=	' '	/	Fabry-Perot Y setting
FPZ	=	' '	/	Fabry-Perot Z setting
FPTEMP	=	' '	/	Fabry-Perot etalon temperature
AIRTEMP	=	' '	/	Fabry-Perot air temperature
FPCALZER	=	' '	/	Fabry-Perot calib zero point offset
FPCALN	=	' '	/	Fabry-Perot Y setting
FPW	=	' '	/	Fabry-Perot Z setting
GFPX0	=	0.0	/	X optical center
GFPY0	=	315.0	/	Y optical center
GFPF	=	' '	/	Gradient
LAMP	=	' '	/	Calibration lamp
XCOARSE	=	' '	/	CS-100 X course
XFINE	=	' '	/	CS-100 X fine
XRBAL	=	' '	/	CS-100 X R-balance
XGAIN	=	' '	/	CS-100 X gain
XTIME	=	' '	/	CS-100 X time constant
YCOARSE	=	' '	/	CS-100 Y course
YFINE	=	' '	/	CS-100 Y fine
YRBAL	=	' '	/	CS-100 Y R-balance
YGAIN	=	' '	/	CS-100 Y gain
YTIME	=	' '	/	CS-100 Y time constant
ZCOARSE	=	' '	/	CS-100 Z coarse
ZFINE	=	' '	/	CS-100 Z fine
ZRBAL	=	' '	/	CS-100 Z R-balance
ZGAIN	=	' '	/	CS-100 Z gain
ZTIME	=	' '	/	CS-100 Z time constant
SECTION4	=	'/////////////////// CCD information ///////////////'
DETECTOR	=	' '	/	E2V CCD 201-BI
CTRLTYPE	=	'Leach '	/	CCD controller type
CCD_MODE	=	'Analog '	/	Analog or photon counting
CCD_PARA	=	'CCD parameters for analog mode'
PIX_SIZE	=	'13.3 x 13.3 microns'	/	Pixel size
GAIN	=	'2.3 e-/ADU'	/	CCD readout gain
R_NOISE	=	'2.88 e- rms'	/	CCD readout noise
FWELL	=	'80 Ke- '	/	Full well
FWELLADU	=	'34.7826086957 Ke-/gain'	/	Full well (ADU)
BIAS	=	3000	/	Bias read noise
DCURRENT	=	'1e-/pix/hour'	/	Static dark current of detector (at 170K)
CCD_TEMP	=	155.0	/	Current CCD temperature (K)
HTR_POWR	=	51.0	/	Current CCD heater power (% of max)
SRO_RATE	=	100.0	/	Serial readout Rate (kHz)
CLK_MODE	=	' '	/	Either IMO or NIMO (currently IMO)
EMCCD	=	'Photon counting mode parameters'
HVCLK	=	' '	/	High voltage clock (Volts)
EMGAIN	=	' '	/	Electron multipying gain parameter
F_RATE	=	' '	/	CCD frame rate (frames per second)
ETH	=	' '	/	Event threshold (e-)
CIC	=	' '	/	Clock induced charge (e-/pix/frame)
SECTION5	=	'////////////////// Exposure information /////////////////'
EXP_TIME	=	180.0	/	Exposure time in seconds
TIM_FILE	=	'fw5_0.lod'	/	timing-load file for CCD exposure
DATE-OBS	=	'2014-05-24T03:55:16.877605'	/	TAI time at start of exposure
MJD-OBS	=	56805	/	Modified Julian Date (TAI) of exposure
ENDRD	=	'2014-05-24T03:55:16.877615'	/	TAI time at end of reading final frame
TIME2CLR	=	0.01	/	Estimated time (sec) to clear the chip
TIME2RD	=	18.0	/	Estimated time (sec) to read the chip
EXP_TYPE	=	'OBJECT '	/	Type of image taken
CNPIX1	=	35.5	/	Lower left X pixel pos of image (overscan)
CNPIX2	=	3.5	/	Lower left Y pixel pos of image (overscan)

D. GIFS Instrument History

The present GIFS instrument is the reincarnation of the Goddard Fabry Perot (GFP). The NASA APRA program supported the GIFS upgrade, which matures integral field spectroscopy and photon counting detector technologies. The team working on this instrument has been co-located at NASA Goddard and was led by Bruce Woodgate. The team now includes Michael McElwain (PI), Carol Grady, James Bubeck, Don Lindler, Jorge Llop, Michael Malatesta, George Hilton, Tim Norton, Ashlee Wilkins, and John Wisniewski.