Change page style: 

Using the ITC to Determine Saturation

You are here

The upper plot in the output of the GNIRS ITC contains the spectrum of the signal and the spectrum of the square root of the background. The lower plot shows the intermediate S/N (for one exposure) and the final S/N (all exposures). The ITC is commonly run with the "Analysis Method" (located near the bottom of the form) set to "Software aperture [i.e., slit length] that gives 'optimum' S/N ratio" - which produces a spectrum summed over several rows of pixels. This mode allows the user to see the S/N that will be achieved, but not whether individual pixels are being saturated.

To determine the latter, the "Analysis Method" should be set to "Software aperture of (diameter or slit length)" The value entered in the box should be 0.15 arcsec if a short camera is being used and 0.05 arc sec if a long camera is being used.

Individual pixels in the GNIRS detector array are approaching full saturation and have non-linear responses when their wells contain more than 65,000 electrons (shallow well) or more than 130,000 electrons (deep well). These values correspond to sqrt (background) of 255e and 360e, respectively. If the signal exceeds 65,000e or 130,000e or the sqrt(background) exceeds 255e or 360e (depending on well size selected) the spectrum will be saturated at the corresponding wavelengths.

In general the user should reduce the individual exposure time until the spectrum is not saturated. In some case, however, where the background saturates only at wavelengths where the atmosphere is opaque or nearly opaque, the user may choose to use longer exposure times to improve the observing efficiency (and thereby the S/N) in the more transparent portions of the spectrum.

For observations of bright standard stars at 1.0-2.5 microns in some of the commonly used cross-dispersed and long slit modes, guideline exposure times fopr avoiding saturation are available here.