Makee: Profile Plots

The profile plots are estimated from medians along the dispersion direction. The profile is smoothed slightly. Here is an example of a "good" set of profiles: .

The first plot are the half-width of the profiles in pixels vs. order number. The dotted line is the median value for the half-widths. The total number of pixels extracted for every order is twice this median value.

The second plot is the offset between the object centroid and the "star" (trace object) centroid vs. order number. A monotically increasing or decreasing trend in the offsets are expected due to the difference in atmospheric refraction between the object and trace star exposures. Three points are excluded from a polynomial fit to the offset values (excluded points circled.) This fit determines the actual centroid used for extracting each order.

The remaining plots are smoothed profiles for each order. A symmetrical "gaussian" shape is expected. The horizontal axis is pixels relative to the trace star centroid. The vertical axis is in digital numbers (DN) for the median pixel in that order. The minimum value is always zero. The vertical dotted lines show the measured object center and the left and right bounds of the profile section extracted. This section nominally encompasses 98% of the object flux. The brackets to the left and right of the object show the start of the background used in the extraction. The text at the top of the plots are: (1) the order# (where "01" is always the first order extracted), (2) the center row ("cr=") which is the row number on the original image for the center of the object in the center column, (3) the background level ("bg=") in DN, and (4) the Full-Width-Half-Maximum of the object profile in arc-seconds. (This measured FWHM is usually slightly larger than seeing plus guiding width measured on the guide TV.)

In this case, the profile has been shifted off to the left in the bluest orders due to atmospheric refraction: . For the first several orders, only sky on the right side of the profile can be used for background subtraction. Although not optimal, this spectrum can still be extracted successfully. The zenith angle for this object was 45 degrees (airmass 1.4). For objects at large zenith angle it is important to either use a TV guide filter centered near the center of your observed wavelength range, or move the object (on the TV) slightly to one side of the slit in order to better center the orders in the blue.
What to look for:

Back to makee home page.