A Test of Possible NICMOS Non-linearity

Observations of standard stars with NICMOS grism (G096), when compared with the same standards taken by STIS (G750L), show a disagreement over the wavelength range 0.8 to 1 micron. Further comparison with ACS F850LP data confirms thisnon-linearity to be due to NICMOS in the sense that fainter stars in NICMOS appear too faint relative to a brighter normalisation. We investigate this over a wide magnitude range by compiling samples of stars (over the range 8 < J < 17) and galaxies (spanning the range 17 < J < 25), with available data with both NICMOS (F110W and F160W) and ground-based JHbands. We find no difference between the NICMOS F160W and ground-based H-band magnitudes. However, in the case of NICMOS F110W and J-band filters, allowing for differences in the filter response functions, we find a trend at the bright-end (J < 17), confirming a non-linearity but smaller than that was seen in Bohlin effect by a factor of 2-3. No significant slope is seen at the faint-end. This non-linearity is interpreted as being due to trapping and the difference in exposure times between the bright and faint objects (Riess 2005). We conclude that most of the observations will not be strongly affected by this effect as the source count rates are sufficiently high so that the fraction of charge lost to traps is negligible. The most affected observations are for faint sources with relatively