Improvements to Calibrations of Astronomical Spectra. (2015)

Undergraduates: Jesus Meza, Erik D. Dennihy, Josh T. Fuchs Bart Dunlap

Faculty Advisor: Christopher Clemens
Department: Physics & Astronomy

---

Astronomical spectra are imperative to the study of a body¿¿¿s composition, rotation, temperature, and gravity. Before these measurements can be extracted however, the raw data must be reduced to a usable form. This involves Bias Subtraction, Flat Fielding, Wavelength Calibration, and Flux Calibration. The project¿¿¿s focus is on improving Wavelength Calibration processes. In Wavelength Calibration, the stellar spectra are compared to the spectra of lamps with known wavelengths. To accomplish this comparison, the known wavelength features in the lamps are matched to fits of the measured spectral line centers. Then typically a polynomial function is used to interpolate wavelengths in between the matched lines. This process is often long and tedious, and incorporates functions unrelated to the physics of dispersion by the diffraction grating used to produce spectra. This costs valuable time and introduces significant and unnecessary error. To improve the process a computer model based on light diffraction is introduced in order to quickly calculate where wavelengths will be positioned within a stellar spectrum. By using the correct functional from, the comparison to reference lamp lines can be automated, and the wavelength error decreased. A careful comparison of spectra calibrated in the conventional method and with the new model is presented.