GNIRS contains a "flip-in" mirror that bypasses the dispersive optics (i.e., gratings and prisms) and provides a long and somewhat narrow acquisition field. This allows precise positioning of an object in a slit without moving the grating, prism, or camera. The choice of acquisition filter is independent of the wavelength of the science observations. 

Target acquisition is currently done in the following manner:

Using the acquisition mirror an image of the requested slit and decker position is obtained.

The slit and decker are removed and the acquisition field is imaged.

The object is identified and the telescope is moved to center it in the slit at the desired position. Nominally, this requires 2 images (to measure initial position and to confirm offset) for objects visible above the background; 3 images if a sky exposure is needed for background subtraction.

The slit and decker are reinserted to image the object through the slit and make final pointing adjustments if needed.

The acquisition mirror is removed and spectroscopy is started.

This technique, involving moving the slit out and back in, is being used because it is not easy to determine whether sources are properly centered in GNIRS' narrow slits. The positional repeatability of the slit has been measured to be considerably better than 1 pixel.

Acquisition observations must be defined by the PI in the Phase II definition.  PIs should specify target brightness (and finding charts if needed) in the filter specified for acquisition, if possible. It is also useful to add notes with acquisition instructions if any special conditions apply (e.g., "center on fainter source W of bright nucleus","acquisition in K filter to match finding chart", etc). See the OT details for more information on defining GNIRS acquisition observations.

Note that there are two tables, one for science involving spectroscopy in the 1-2.5µm range with one of the blue cameras, and the other for science involving spectroscopy in the 3-5µm range and use of one of the red cameras. For example, for 1-2.5µm spectroscopy with the short blue camera the normal choices are the H2+ND100X filters for targets brighter than K~7, the H2 filter for K magnitudes between ~7 and ~11.5, and the J or H filter for fainter targets.

Recommended exposure times for "comfortable" (i.e. high S/N) acquisitions are given in the tables below. Note that when using one of the long cameras (0.05 arcsec/pixel) without AO there is a 2.5 magnitude offset from the values in the table, which apply to the short cameras (0.15 arc sec/pixel). For example for an H~K~7.5 mag star to be acquired and observed with a long camera and no AO, use the "coadds x exp time" listed for H~K~10 mag.

In all situations (even very bright stars) we recommend total exposure time (no. of coadds x exposure time) of at least 5 seconds for accurate offsetting accuracy.

For targets fainter than J~H~16.5 or K~15.5 (J~H~14.5 or K~13.5 for the long camera without AO) acquisition should include obtaining a "sky" frame with the telescope offset at least a few arcsec from the nominal coordinates, for the purpose of sky subtraction.

For science using the long cameras and narrow slits do not use the ND100X filter during acquisition (unless the ND filter is also being used when obtaining spectra), because the field is offset perpendicular to the slit by ~0.10 arcsec compared to when it is not in the light path. It is also advisable not to use the ND filter with the short cameras and the narrowest slits (0.30 and 0.45 arc sec) for the same reason, if accurate centering is required.

Objects fainter than J~20.5 or H~K~19.5 require "blind" acquisition. The PI must provide IR-visible reference star(s), one or two, that can be acquired with the same guide star as the science target, and with the same coordinate system as the science target.  See "Blind or Offset Acquisitions."

Individual exposure times greater than 10 seconds at J, 3 seconds at H, and 2 seconds at K with the short camera (0.15"/pix) can lead to persistence in the form of a faint image of the acquisition field on the array in the initial few spectral frames. For that reason the recommended individual exposure times in the table below are short, and for faint targets numerous coadds are employed instead of longer exposures and fewer coadds. For the long camera (with AO) fewer, longer exposures should be used for faint targets, as indicated in the table.

Recommended GNIRS Acquisition Exposure Times - Science at 1.0-2.5 Microns
(applies to non-AO+short blue camera and to AO+long blue camera except where noted;
for long blue camera without AO subtract 2.5 mag from values in column 1)

mag	Filter	coadds x exp. timea	total time (sec) a	Read Mode
K=2.0b	H2+ND100Xc	25 x 0.2 sec	5	Very Bright
K=3.0b	H2+ND100Xc	10 x 0.5 sec	5	Very Bright
K=4.0b	H2+ND100Xc	5 x 1 sec	5	Bright
K=4.0b	H2+ND100Xc	5 x 1 sec	5	Bright
K=5.0b	H2+ND100Xc	3 x 2 sec	6	Bright
K=6.0b	H2+ND100Xc	2 x 4 sec	8	Bright
K=7.0b	H2	25 x 0.2 sec	5	Very Bright
K=8.0	H2	10 x 0.5 sec	5	Very Bright
K=9.0	H2	5 x 1 sec	5	Bright
K=10.0	H2	3 x 2 sec	6	Bright
K=11.0	H2	2 x 4 sec	8	Bright
K=12.0	H2	1 x 10 sec	10	Bright
H=6	H+ND100Xc	25 x 0.2 sec	5	Very Bright
H=7	H+ND100Xc	16 x 0.3 sec	5	Very Bright
H=8	H+ND100Xc	8 x 0.6 sec	5	Bright
H=9	H+ND100Xc	5 x 1 sec	5	Bright
H=10	H+ND100Xc	4 x 2 sec	8	Bright
H=11	H+ND100Xc	4 x 3 sec	12	Bright
H=11.5	H-MK	25 x 0.2 sec	5	Very Bright
H=12.0	H-MK	16 x 0.3 sec	5	Very Bright
H=13.0	H-MK	8 x 0.6 sec	5	Bright
H=14.0	H-MK	5 x 1 sec	5	Bright
H=15.0	H-MK	4 x 2 sec	8	Bright
H=16.0	H-MK	4 x 3 (AO 1 x 12) sec	12	Bright
H=17.0d	H-MK	8 x 3 (AO 1 x 24) sec	24	Bright
H=18.0d	H-MK	20 x 3 (AO 2 x 30) sec	60	Bright
J=11.0	J (order-blocking)	25 x 0.2 sec	5	Very Bright
J=12.0	J (order-blocking)	10 x 0.5 sec	5	Very Bright
J=13.0	J (order-blocking)	5 x 1 sec	5	Bright
J=14.0	J (order-blocking)	3 x 2 sec	6	Bright
J=15.0	J (order-blocking)	1 x 6 sec	6	Bright
J=16.0	J (order-blocking)	1 x 8 sec	8	Bright
J=17.0d	J (order-blocking)	1 x 10 sec	10	Bright
J=18.0d	J (order-blocking)	2 x 10 sec	20	Bright
J=19.0d	J (order-blocking)	6 x 10 (AO 1 x 60) sec	80	Bright
K=10.0	K (order-blocking)	25 x 0.2 sec	5	Very Bright
K=11.0	K (order-blocking)	10 x 0.5 sec	5	Very Bright
K=12.0	K (order-blocking)	5 x 1 sec	5	Bright
K=13.0	K (order-blocking)	3 x 2 sec	6	Bright
K=14.0	K (order-blocking)	4 x 2 sec	8	Bright
K=15.0	K (order-blocking)	6 x 2 sec	12	Bright
K=16.0d	K (order-blocking)	8 x 2 (AO 1 x 16) sec	16	Bright
K=17.0d	K (order-blocking)	15 x 2 (AO 2 x 15sec	30	Bright
K=18.0d	K (order-blocking)	45 x 2 (AO 5 x 18) sec	90	Bright

^a Values in table are for good conditions (IQ70 CC50). For worse conditions increase number of exposures. E.g., for poor seeing or for thin clouds increase number of coadds by about 50%; for bad seeing or for thick clouds roughly double the number of coadds.
^b Point sources brighter than K~7 with the short blue camera (K~4.5 when using long blue camera without AO) will saturate the array in imaging mode when the ND100X filter is not used. With the ND filter the limits are 5 magnitudes brighter, but see restrictions in footnote c on its use. Limits are about 2 magnitudes brighter for the red cameras. Saturation results in image persistence in subsequent frames for several minutes. In the OT library the nod sequences are set up so that spectra avoid the array rows where saturation might occur during acquisition. Alternatively the PIs may set up their own nod sequences to avoid saturated areas.
^c Do not use the ND100X filter with long camera and narrow slits or with the short cameras and the 0.30 and 0.45 arc sec wide slits (unless the ND filter is also being used when obtaining spectra); the position of a target is shifted perpendicular to the slit by ~0.10 arc-seconds when it is used.
^d In order to acquire objects fainter than J=H~16.5 or K~15.5 (J=H~14.5 or K~13.5 with long camera without AO) sky subtraction is usually needed (see above text above the table). In that case the total time (for a pair of frames) is double the value in column 4 above.

Recommended GNIRS Acquisition Exposure Times - Science at 3-5 Microns
(applies to non-AO+short red camera;
for long red camera subtract 2.5 mag from values in column 1)

/tr> /tr> /tr> /tr> /tr>

mag	Filter	coadds x exp. timea	total time (sec)a	Read Mode
K=0.0b	H2+ND100Xc	25 x 0.2 sec	5	Very Bright
K=1.0b	H2+ND100Xc	10 x 0.5 sec	5	Very Bright
K=2.0b	H2+ND100Xc	5 x 1 sec	5	Bright
K=3.0b	H2+ND100Xc	3 x 2 sec	6	Bright
K=4.0b	H2+ND100Xc	2 x 4 sec	8	Bright
K=5.0b	H2	25 x 0.2 sec	5	Very Bright
K=6.0	H2	10 x 0.5 sec	5	Very Bright
K=7.0	H2	5 x 1 sec	5	Bright
K=8.0	H2	3 x 2 sec	6	Bright
K=9.0	H2	2 x 4 sec	8	Bright
K=10.0	H2	1 x 10 sec	10	Bright
H=5	H+ND100Xc	25 x 0.2 sec	5	Very Bright
H=6	H+ND100Xc	10 x 0.5 sec	5	Very Bright
H=7	H+ND100Xc	5 x 1 sec	5	Bright
H=8	H+ND100Xc	5 x 2 sec	10	Bright
H=9	H+ND100Xc	5 x 3 sec	15	Bright
H=10	H-MK	25 x 0.2 sec	5	Very Bright
H=11	H-MK	10 x 0.5 sec	5	Very Bright
H=12.0	H-MK	5 x 1 sec	5	Bright
H=13.0	H-MK	5 x 2 sec	10	Bright
H=14.0d	H-MK	5 x 3 sec	15	Bright
H=15.0d	H-MK	7 x 3 sec	21	Bright
H=16.0d	H-MK	10 x 3 sec	30	Bright
H=17.0d	H-MK	15 x 3sec	45	Bright
L=6e	PAH	25 x 0.2 sec	5	Very Bright
L=7	PAH	10 x 0.5 sec	5	Very Bright
L=8f	PAH	10 x 0.8 sec	8	Bright
L=9f	PAH	15 x 0.8 sec	12	Bright
L=10f	PAH	20 x 0.8 (2 x 7) secg	16	Bright
L=11f	PAH	25 x 0.8 (3 x 7) secg	20	Bright
L=12f	PAH	50 x 0.8 (7 x 7) secg	40	Bright

^a Values in table are for good conditions (IQ70 CC50). For worse conditions increase number of exposures. E.g., for poor seeing or for thin clouds increase number of coadds by about 50%; for bad seeing or for thick clouds roughly double the number of coadds.
^b Point sources brighter than K~5 with the short blue camera (K~2.5 when using long blue camera without AO) will saturate the array in imaging mode when the ND100X filter is not used. With the ND filter the limits are 5 magnitudes brighter, but see restrictions in footnote c on its use. Limits are about 2 magnitudes brighter for the red cameras. Saturation will result in image persistence in subsequent frames for several minutes. In the OT library the nod sequences are set up so that spectra avoid the array rows where saturation might occur. Alternatively the PIs may set up their own nod sequences to avoid saturated areas.
^c Do not use the ND100X filter with long camera and narrow slits or with the short camera and the 0.30 and 0.45 arc sec wide slits (unless the ND filter is also being used when obtaining spectra); the position of a target perpendicular to the slit is shifted by ~0.10 arc-seconds when it is used.
^d Acquisition should include obtaining a "sky" frame with the telescope offset at least a few arcsec from the nominal coordinates, for the purpose of sky subtraction.
^e Point sources brighter than L=6 will saturate the array in imaging mode.
^f In order to acquire objects fainter than L=7 (with either pixel scale) sky subtraction is needed. See text above the tables. In that case the total time (for a pair of frames) is double the value in column 4.
^g Do not use individual exposures longer than 0.8 seconds (7 seconds with the long red camera) or the sky+telescope background will saturate the array.