An Empirical Method for Improving the Quality of RXTE HEXTE Spectra

TL;DR

This paper presents an empirical correction method for RXTE HEXTE spectra, significantly enhancing spectral quality and fit accuracy, especially for Cluster B data, by creating and applying a calibration tool based on residual analysis.

Contribution

The paper introduces a novel empirical correction tool for RXTE HEXTE spectra, improving spectral quality and fit accuracy, especially for Cluster B, using residual-based calibration.

Findings

Residual spectrum of Cluster A is free of artifacts.

Cluster B residuals show significant features near 30-50 keV.

Application of the correction tool improves fit quality for bright black hole spectra.

Abstract

We have developed a correction tool to improve the quality of RXTE HEXTE spectra by employing the same method we used earlier to improve the quality of RXTE PCA spectra. We fit all of the hundreds of HEXTE spectra of the Crab individually to a simple power-law model, some 37 million counts in total for Cluster A and 39 million counts for Cluster B, and we create for each cluster a combined spectrum of residuals. We find that the residual spectrum of Cluster A is free of instrumental artifacts while that of Cluster B contains significant features with amplitudes ~1%; the most prominent is in the energy range 30-50 keV, which coincides with the iodine K edge. Starting with the residual spectrum for Cluster B, via an iterative procedure we created the calibration tool hexBcorr for correcting any Cluster B spectrum of interest. We demonstrate the efficacy of the tool by applying it to Cluster B spectra of two bright black holes, which contain several million counts apiece. For these spectra, application of the tool significantly improves the goodness of fit, while affecting only slightly the broadband fit parameters. The tool may be important for the study of spectral features, such as cyclotron lines, a topic that is beyond the scope of this paper.