2018ApJ...853..116Z

We modified the broadband photometric reverberation mapping (PRM) code, JAVELIN, and tested the availability to get broad-line region time delays that are consistent with the spectroscopic reverberation mapping (SRM) project SDSS-RM. The broadband light curves of SDSS-RM quasars produced by convolution with the system transmission curves were used in the test. We found that under similar sampling conditions (evenly and frequently sampled), the key factor determining whether the broadband PRM code can yield lags consistent with the SRM project is the flux ratio of the broad emission line to the reference continuum, which is in line with the previous findings. We further found a critical line-to-continuum flux ratio, about 6%, above which the mean of the ratios between the lags from PRM and SRM becomes closer to unity, and the scatter is pronouncedly reduced. We also tested our code on a subset of SDSS Stripe 82 quasars, and found that our program tends to give biased lag estimations due to the observation gaps when the R-L relation prior in Markov Chain Monte Carlo is discarded. The performance of the damped random walk (DRW) model and the power-law (PL) structure function model on broadband PRM were compared. We found that given both SDSS-RM-like or Stripe 82-like light curves, the DRW model performs better in carrying out broadband PRM than the PL model.