Background radiation in the mid-IR

All of telescopes, warm part of instruments, and sky emit infrared radiation according to their temperatures. Since mid-infrared is corresponding to the peak of blackbody emission around 300K, their mid-infrared radiation is large for the ground-based observations. In the near-infrared, relatively high background radiation is also observed and canceled by using sky median frames. However, it can be done when object brightness is not too much less than sky radiations and when flat can be measured in high accuracies. In the mid-infrared, situation is different. Sky emission is much brighter than most of the brightest object in all over the sphere. Flat accuracy is not so high because sky emission varies fastly with spatially fluctuation. Even if sky has very uniform emission across a frame, flat fielding by low-accurate flat frames lead us to artificial spatial pattern and signal object tend to be embedded in them. So a different method is required to cancel the background radiation.

What is 'chop and nod' ?

In the mid-infrared, 'chop and nod' technique is normally used to cancel the very high background radiation. Nod is beam switching by changing the telescope direction a little. Chop is tilting the telescope secondary with a little angle at a telescope position. Chop is done by around 0.1 to a few Hz, and nod is done more slowly (0.03Hz or slower).

By 'chop and nod', one take images where object is imaged on a bit different positions of the frames. For the obtained data, you first cancel the background radiation by subtracting the blank sky data in the offset beam of the chopping. This corresponding to cancelling fast varing conponents of the background emission. Then remaining residual patterns, which is caused by using different parts of the optics, is removed by make subtraction between frame sets of different nod positions.

In this method, one don't devide the high sky frames by low-accurate flat frames. One subtract the sky emission by using close sky emission which seems to be almost the same brightness. After this subtraction counts for pixels become around zero. Devision of near zero frame by low-accurate flat frames do not make large artificial fluctuations this time. It becomes easy to detect objects. Nod subtraction is done to cancel artificial structures generated by using different parts of the optics. It is not required when the object brightness (brightness you want to detect) is high enough and then nod method can be omitted.

Subaru's IRSM: chopping performance

From a point of view of purpose and performance of chop and nod technique, chopping frequency must be faster than the sky varing time and chopping amplitude must be smaller than the sky varing scale. However, object size and the secondary performance limit the amplitude to be used. Also the chopping frequency is limited by secondary performances and weather. Chopping frequency is determined through clock parameters.

Normally, main beam of chopping uses normal configuration of all optics, and offset beam uses tilted secondary configuration. In wide amplitude mode, where amplitude is larger than 30", both beam of chopping uses tilted secondary configuration and get wide amplitude.

You should use sphere synchronous mode usually, because this mode fix object positions on the images and you can use offset beam images in some cases. In Az/El synchronous mode, object in offset beam images rotate on the images because Subaru is an Az/El driving telescope.

When you decide offset beam positions of chopping and nodding, you should take care the position has no significant mid-infrared emission not to contaminate the object structure in main beam after subtractions. When object is small enough and you can image in some beams, you can use those data and improve signal to noise ratios of detections. (All of the sensitivities listed in these web pages are those for on-source integration time, which is for only one beam of the chop and nod.) This is of course limited in the case when you use sphere synchronous mode. In the spectroscopic observations, you can choose beam positions so that you object is located on different positions of the slit.

Subaru 2ndary mirror
Input Parameters	Amplitude, Position Angle, and Coordinate Base.
Amplitude	normally up to 30"
	In wide amplitude mode, from 30" up to 60"
Coordinate Base	1. Right Ascention and Declination (alpha, delta).
	2. Azimuth and Elevation (Az/El).
Frequency	up to 1Hz for COMICS practically.
This mode may be used from the end of 2001, technically.