Back
The following two figures show time-averaged spectra of HD 173511 and HR 6348, the two standards we have observed the most regularly, plotted in Rayleigh-Jeans units to emphasize the variations. The spectra are grouped in segments of five campaigns each. The spectra are usually normalized to one of the LL segments.
HD 173511 holds approximately steady through Campaign 9, then decreases through Campaign 19. Since then, the time-averaged spectra have remained roughly constant. The total decrease, as measured in LL1, is slightly more than 5%.
HR 6348 shows a similar decrease with time, although the details differ. The main point is that both HR 6348 and HD 173511 are decreasing together, making it more likely that this is an instrumental effect and not due to intrinsic stellar variability. The overall variation is about 6%, but the first group is high primarily because of spectra obtained in Campaign P. Ignoring these initial spectra would tighten the distribution.
We have good, but less complete coverage of three additional standards, HD 166780, del UMi, and alp Lac. These sources show qualitatively similar behavior, all showing decreasing signal in the most recent observations.
PhotometryAll IRS observations of HR 6348 were made using Red self Peak-Up (PU), which gives us 22 µm photometry associated with each spectrum. Initial observations of HD 173511 used offset peak-up sources, but beginning with Campaign 5, we also used Red self-PU. Up until Campaign 20, we also observed the three other sources mentioned above as frequently as possible. The combined data set gives us five targets observed many times, allowing a detailed analysis of how their apparent flux, as measured on the Red PU sub-array, has changed with time.
In the figure above, multiple observations of a given target have been combined together in each campaign and plotted along with the standard deviation of the distribution. The horizontal lines are plotted just to guide the eye. A break in the responsivity of the red PU sub-array is apparent in all five standards between Campaigns 14 and 15. The average of all data before this break is roughly 2% higher than after the break.
The only reasonable conclusion is that the responsivity of the module has changed during the mission. Combined with the spectroscopy above, we have evidence of temporal variations in two of the four modules, although it is important to note that the measured effect differs by a factor of more than two between SL and LL.
All PU observations of the 5 standards
The above figure shows the data used to generate the preceding figure. Several outliers above and below the main set of data were rejected before generating averages for each campaign. They probably arise from cosmic ray hits or latent images.
Last modified 9 May, 2006. © The IRS Team.