Previous: Flatfielding
Up: Data Reduction
Next: Flux calibration
Previous Page: Flatfielding
Next Page: Flux calibration

Spectrum extraction and Sky subtraction.

Normally this is the only part of the online reduction package used by observers. The heart of the FOS reduction software is contained in these routines. Each spectral order presents a curvature which depends primarily on the FOS optics but also on the atmospheric dispersion when the slit is near the parallactic angle. The curvature can be mapped at any z.d. as a high order polynomial fit to an observation of a bright point source. using the routine FOSCOFFS. The typical observer, working at reasonable z.d. will not have to worry about this and can use the default calibration.

The first step is to straighten the separate orders of the spectrum, one at a time. TRIM prompts you for the order you want to straighten, and then using the lexidata joystick to delimit this area, straightens the spectrum using the coefficients stored in frstcofs.dat and scndcofs.dat depending on which order you are working on. Straightening the data reconfigures the pixels according to the curvature coefficients, but there is no resampling. If the observer displays the TRIMmed image each new y value represents a single spatial point along the slit.

The next step is to carry out the sky subtraction. This is done using the routine SPECTRUM. You enter 4 cursor positions using the Lexidata joystick and control buttons. Depending on how you select your sky area(s), SPECTRUM will either interpolate under the object or calculate a straight mean. If you don't select any sky areas, the object area will not be sky subtracted.

The resulting spectra from TRIM and SPECTRUM are of a standard form - a spectrum pair consisting of the sky subtracted object signal, and the sky spectrum along the slit. The wavelength to channel number relationship is very nearly linear, and so there is no need to resample the reduced spectra.



Previous: Flatfielding
Up: Data Reduction
Next: Flux calibration
Previous Page: Flatfielding
Next Page: Flux calibration


Tue Apr 12 04:13:28 BST 1994