Flattening Scientific CCD Imaging Data with a Dome Flat Field System

TL;DR

This paper introduces a dome flat field system using LEDs for CCD imaging calibration, demonstrating it produces superior flat fields compared to traditional twilight and sky flats, especially for photometric accuracy.

Contribution

The paper presents a novel LED-based dome flat field system that improves flat field quality over traditional methods for CCD imaging calibration.

Findings

Dome flat fields outperform twilight and sky flats in photometric calibration.

LED illumination offers advantages like wavelength diversity, cost-effectiveness, and long lifespan.

Standard star tests confirm the superior reproducibility of the dome flat system.

Abstract

We describe the flattening of scientific CCD imaging data using a dome flat field system. The system uses light emitting diodes (LEDs) to illuminate a carefully constructed dome flat field screen. LEDs have several advantages over more traditional illumination sources: they are available in a wide range of output wavelengths, are inexpensive, have a very long source lifetime, and are straightforward to control digitally. The circular dome screen is made of a material with Lambertian scattering properties that efficiently reflects light of a wide range of wavelengths and incident angles. We compare flat fields obtained using this new system with two types of traditionally-constructed flat fields: twilight sky flats and nighttime sky flats. Using photometric standard stars as illumination sources, we test the quality of each flat field by applying it to a set of standard star observations. We find that the dome flat field system produces flat fields that are superior to twilight or nighttime sky flats, particularly for photometric calibration. We note that a ratio of the twilight sky flat to the nighttime sky flat is flat to within the expected uncertainty; but since both of these flat fields are inferior to the dome flat, this common test is not an appropriate metric for testing a flat field. Rather, the only feasible and correct method for determining the appropriateness of a flat field is to use standard stars to measure the reproducibility of known magnitudes across the detector.