Infrared catastrophe

[AlexGKim]

denali coadd-4-80641-20210103.fits 39633169497982658

The flux seems to run away at high wavelengths. This behavior is apparent in previous observations of the same object though not as extreme,

[julienguy]

I do not see anything wrong with the spectroscopy. I think you have found a genuine variable object :-). Could it be the superposition of 2 sources, one blue, one red, and variable?

plot_spectra -i ~/redux/denali/tiles/pernight/80641/*/coadd-4-80641-*.fits -t 39633178004032511 --rebin 4

[AlexGKim]

It may be due to inconsistent fiber placement? My knowledge of astrophysics is limited, I don't know of what could be so red.

[arjundey]

I think this is a case of the fiber being mispositioned off the star. If this was a real superposition of two sources (or even a single variable source), the EW of the absorption lines would change, and they do not seem to - there does not appear to any change in the type of the spectrum, if one corrects for the continuum shape (a bit hard to tell from the spectra, but would be easy if you just divided out the continuum and then compared). In the case of a double star system, the spectral type from the red star would be quite different, and if it was varying (i.e., a spectroscopic binary), you would see the two types clearly. If it were a single variable object, the effective temp would change and the effective spectral type would change. As a result, i think this is just a case where the fiber is off the source and depending on whether the seeing is good or bad and what airmass this is done at, we get different contributions of red and blue light from the same source. The finding chart below from the viewer shows that the SV1 target was not centered on the source. The link to the viewer is here: https://www.legacysurvey.org/viewer?ra=97.35266&dec=45.24&layer=ls-dr9&zoom=16&sources-dr9&targets-dr9-sv1-bright