| Abstract : | The existence of QPOs in AGNs was first determined by Papadakis[1] in 1993, in the X-ray light curve (XLC) of the Seyfert galaxy NGC 3516, revealing periodic variations with a characteristic timescale of approximately 10 hours. McHardy et al.[2] demonstrated that QPOs are not limited to individual AGNs but are a widespread phenomenon.
We present here a comparative study of different methods to detect possible (QPO) signals in the AGN XLCs. Low count rates and irregular sampling pose significant challenges to the study of XLCs. To overcome this, we shift our focus to the frequency domain. Despite random gaps in data, power spectral density (PSD) analysis using the Fast Fourier Transform (FFT) allows us to obtain an overall frequency domain picture, including identifying dominant frequencies present in the XLC. The Structure Function (SF) is an excellent complementary approach to the conventional FFT[3], especially for the identification of QPOs. The advantage of the Structure Function (SF) is that it can be applied to irregularly sampled data and is unaffected by any gaps in the XLC.
Applying this combination of methods to XLCs of x-ray identified AGNs from publicly available datasets, we obtain the FFT-PSD in each case. Fitting the PSD with a power law, we analyse the residues of the fitted PSD for peaks which lie above the local moving average in different regions of interest, with 2σ and 3σ, i.e. 95% and 99% confidence limits, respectively. We use the SF as an independent indicator of the genuineness of the oscillation frequencies obtained. The details of QPOs identified using this methodology will be presented.
References:
[1]Papadakis and Lawrence, Nature Vol.361, pg.233 (1993).
[2]McHardy et al., MNRAS Vol.392, pg.985 (2004).
[3]Agarwal et al., Galaxies Vol.9, pg.20 (2021)
|
