4 Comparison MOS-IFS(LR)

Table 2 summarizes the results of the MOS-IFS comparison, at equal S/N per object, R = 150, and considering the observed mean sizes and densities of galaxies in the HDF (and from Keck data) - further details in the comments:

 

Table 2: IFS-MOS comparison. Unless explicity stated otherwise, the numbers represent ratios of MOS/IFS.

KAB	20.5	21.5	22.5	23.5	24.5	25.5	26.5	27.5	Comments
FOV	22	22	22	22	22	22	22	22	1
Crowding	0.75	0.71	0.61	0.53	0.47	0.41	0.38	0.31	2
Edge	0.86	0.86	0.86	0.87	0.87	0.87	0.875	0.875	2
Multiplex	14	13.4	11.4	10.	9.	7.9	7.	6.
Slit losses	0.29	0.34	0.4	0.48	0.55	0.63	0.84	1.0	3
Detector proj	0.88	0.79	0.66	0.54	0.44	0.40	0.33	0.37	4
tifs/tmos	0.26	0.27	0.26	0.26	0.24	0.25	0.28	0.37	5
Simult	0.5	0.5	0.5	0.5	0.5	0.5	0.5	0.5	6
Smos/Sifs	1.84	1.8	1.5	1.3	1.1	1.0	1.0	1.12

(1) From the baseline designs
(2) Assuming spectrum width/length of 10 x r_1/2/514 px. For higher spectral resolution, these factors will be more relevant.
(3) These factors correspond to N_mos/N_ifs and are directly related to the slit losses. At longer wavelength, these factors are more relevant as a consequence of the increase of the image size due to the diffraction limited PSF.
(4) 1/F. Its value is dominated by the detector pixel size, weighted by the relative importance of the different sources of noise.
(5) Applying eq. (4), considering $\epsilon_{mos}$ = $\epsilon_{ifs}$
(6) MOS requires two exposures to cover the complete 0.6-5 micron range, while IFS does it simultaneously.

In figure 6, the relative speed of the MOS and IFS designs are shown, for R=150. It is clear that the multiplexing gain of MOS is partially compensated by the larger sensitivity of IFS. For larger spectral resolution the crowding and edge factors will be more important, and the detector noise relatively more important. At longer wavelengths, the image size is going to be larger, which implies more slit losses for MOS.

At the faintest end, MOS can adjust better its pixel size to the object size, which implies an advange in terms of sensitivity. However, this occurs at K_AB> 27. (Note that at K_AB = 26.5, r_1/2 = 0.15'').

 

Figure 6: Relative 'speed' of the IFS-LR and MOS desings as a function of the magnitude, for R = 150, and equal S/N per object, not per pixel.