relativistic-reverberation-in-mcg-5-23-16-and-swift-j127.4-5654

A Hard Look At Relativistic Reverberation in MCG–5-23-16 \& SWIFT J2127.4+5654: Testing the Lamp-Post Model

This repository contains the code data associated with the article titled: A Hard Look At Relativistic Reverberation in MCG–5-23-16 \& SWIFT J2127.4+5654: Testing the Lamp-Post Model, published in the Astrophysical Journal.

Abstract

X-ray reverberation mapping has emerged as a new tool to probe accretion in AGN, providing a potentially powerful probe of accretion at the black hole scale. The lags, along with relativistic spectral signatures are often interpreted in light of the lamp-post model. Focusing specifically on testing the prediction of the relativistic reverberation model, we have targeted several of the brightest Seyfert Galaxies in X-rays with different observing programs. Here, we report the results from two large campaigns with NuSTAR targeting MCG–5-23-16 and SWIFT J2127.4+5654 to test the model predictions in the 3-50 keV band. These are two of three sources that showed indications of a delayed Compton hump in early data. With triple the previously analyzed exposures, we find no evidence for relativistic reverberation in MCG–5-23-16, and the energy-dependent lags are consistent with a log-linear continuum. In SWIFT J2127.4+5654, although a continuum-only model explains the data, the relativistic reverberation model provides a significant improvement to the energy and frequency-dependent lags, but with parameters that are not consistent with the time-averaged spectrum. This adds to mounting evidence showing that the lag data is not consistent with a static lamp-post model.

Description

The analysis is organied into several python notebooks, which sometimes call outside functions either from my toolset package aztools or the helpers.py file. The lag calculations are done using the package fqlag.

The analysis for each of the two sources is contained in the folders named mcg5 and sw2127 respectively. The helper.py script is used by both of them.

The analysis of the xmm data which is used briefly in some parts of the analysis uses individual scripts located in {data-folder}/xmm/ (where data-folder refers to either mcg5 or sw2127) and README file in each case explains what the scripts do

A quick description of the notebooks is as follows:

• mcg5/data.md and sw2127/data.md: Used for data preparation, reduction and the extraction of the spectra and light curves
• mcg5/lc_psd.md and sw2127/lc_psd.md: Used for estimating the power spectra.
• mcg5/lag_22l3.md and sw2127/lag_22l3.md: Lag calculations using the full 22 energy bins used in section 3.1 in the paper.
• mcg5/lag_22l3b.md and sw2127/lag_22l3b.md: Lag calculations at coarse energy bins (using 11 bins). This first calculates the lag at the published frequencies, then lag-vs-frequency, and lags at the new frequencies

Data Products:

The data products are available throught the Open Science Framework (links below). Only data that is not available in the heasarc public archive (or can be derived from it in a straightforward manner) is included.

There are two sets of files, one for each source, named: mcg5.tgz (1GB) and sw2127.tgz (1GB). Each one contains the following folders:

• nustar/: contains the folders {obsid}_p for each observations. Each folder contains the light curves and spectra from that observation.
• xmm/: contains codes and data products for the xmm data, described in details in xmm/README
• spec_xmm_nustar/: contains the nustar and xmm spectra and spectral fits used to obtain a time-averaged spectral fit. A discription of the fits is in spec_xmm_nustar/README
• timing/: contains the results of the psd and lag calculations, including all the modeling with kynreverb. The lag spectra are written into pha format where they can be modeled with xspec.
  • timing/*npz contains the actual lightcurve data used in the calculations.
  • timing/psd: contains the lightcurve/psd calculations and modeling.
  • timing/lag: contains the lag numbers in *npz files, and the lag spectra in the pha folder. The plots, including those in the paper are in the plots folder.