Research Project Objectives (scientific problem aimed to be solved by the proposed project, project’s research hypotheses)

The aim of the project is to use a novel tool – active galaxies – in order to obtain constraints for the expansion rate of the Universe, and thus on the properties of the dark energy. The existence of the dark energy is a challenge for physics. Current microphysics – models of the particle interaction – do

not require its existence. In particular, there is no space for the dark energy in the Standard Model of particle physics. However, observations of the Cosmic Microwave Background (CMB), Supernovae Ia (SN Ia), and the growth rate of the Large Structure show convincingly that the expansion of the Universe accelerates during the last 7 Gyr instead of being decelerated by the gravity. Usually, gravity causes gravitational attraction, so the medium responsible for accelerated expansion of the Universe must have strange properties. There is a way to get anty-gravity effect: Einstein introduced a mathematical term – cosmological constant – to his equations, and if this constant is interpreted as a material source, it has positive energy density and negative pressure, and this negative pressure term wins in the global energy budget. However, the fact that there is something like that, and actually dominates now the dynamics of the Universe, came as a surprise. The name – dark energy – for this medium helps us to see the challenge we face in astronomy and physics.

Current observational are simply consistent with the Einstein’s cosmological constant, i.e. the dark energy seems to be uniform in space and time. Numerous models of possible additional forces or fields are being developed by theoreticians, but there would be a way to differentiate between various options if astronomers can reliably measure a departure from a single universal constant. With increasing accuracy, some tension between the measurements of the Universe parameters starts to appear, either real or just due to underestimation of the systematic errors. For example, the local measurements of the Hubble constant (e.g. H0 = 73.24±1.74, Riess et al. 2016) are not quite consistent with the constraints from the microwave background (e.g. H0 = 66.93±0.62 km/s/Mps from Adam et al. 2016). Thus the known methods are being continuously refined, and new probes are being searched for, since the most important issue at this stage are systematic errors, the most difficult to estimate in any single method.

We postulate that active galaxies can be used as a new cosmological tool. They are numerous, they cover very broad redshift range from nearby sources at a distance of 10 Mpc to quasars at redshift 6. The use of active galactic nuclei (AGN), on the other hand, is more complex, there are actually several independent ways how they can be used for cosmology (see e.g. Czerny et al. 2017a), and the accuracy of those methods has still to be determined or improved. The basic idea behind is that we have to measure a specific property of a given AGN which allows (indirectly) to determine its absolute luminosity. This is later combined with the measurement of the source redshift and the observed flux, which are relatively easy to measure. In this way we construct the Hubble diagram for our sources, and we can trace the expansion rate of the Universe as a function of redshift (or, equivalently, time).

In this project we will concentrate on the following methods:

(A) spectroscopic line reverberation

(B) photometric line reverberation

(C) single spectra methods based on selection of sources radiating very close to the Eddington ratio (D) continuum reverberation.

Significance of the project (state of the art, justification for tackling specific scientific problems by the proposed project, pioneering nature of the project, the impact of the project results on the development of the research field and scientific discipline, economic and societal impact)

The need for the cosmological constant has already been postulated in early 80′ as a result of studies of the evolution of the large scale structure in the Universe (e.g. Peebles 1984). Stronger evidence came from the combination of the large scale structure growth with the fluctuations of the Cosmic Microwave Background which gave an estimate of the cosmological constant about 0.3 (Kofman et al. 1993). The next major step came from the SN Ia studies which gave much more precise measurements of the cosmological constant of about 0.7 (Riess et al. 1998, Perlmutter et al. 1998).

Since the physical interpretation of this medium is unclear, the first step is to confirm the presence of this medium with several independent methods, and, if possible, to detect observationally any departure from a simple constant. Thus, a number of independent methods were developed (for a recent review, see Huterer & Shafer 2017). New methods, now considered as secondary, are also currently being developed, like those based on SN II, strong gravitational lensing, gravitational waves, galaxy clusters, gamma-ray bursts, and active galaxies. Those methods are still in their infancy (see Czerny et al. 2017 for a review).

Active galaxies have some advantages over SN Ia as cosmological probes. The brightest of them, known as quasars, are the most luminous persistent sources in the Universe. The nucleus contains a supermassive black hole, surrounded by an accretion disk and clouds. Much of the radiation generated by the central engine is absorbed and re–processed by clouds. So active galaxies are complex, but their variable continuum emission with strong emission lines offers multiple diagnostics of their properties, including the absolute luminosity of each source. So active galaxies are not standard (or even standardizable) candles, but knowing the absolute luminosity of each source we can place them on the Hubble diagram and use for cosmology.


Fig. 1. Schematic view of an active nucleus. Variable continuum emission predominantly comes from the innermost part of accretion disk at 10 gravitational radii while the BLR is at a distance 100 times larger (pink dots), so lines coming from reprocessing respond with a delay of order of light travel time (courtesy of K. Hryniewicz).

Three of the methods used in the current project (A, B and C) will be based on the emission lines properties. Variable emission in the innermost part of the disk irradiates the clouds located in the Broad Line Region (BLR). Measured delay gives the light travel time to BLR. Studies of nearby active galaxies showed that this delay depends only on the absolute luminosity of the source measured at a specific wavelength (Kaspi et al. 2000, Betz et al. 2013). This scaling had initially no interpretation but in Czerny & Hryniewicz (2011) we showed that this relation can be simply interpreted within the frame of the model where the creation of the BLR clouds is caused by the dusty (failed) wind. This scenario, which we call FRADO – Failed Radiatively Accelerated Dusty Outflow – provides explanation why and where the clouds can appear high above the disk. The model was already tested in Czerny et al. (2015), Czerny et al. (2016) and Czerny et al. (2017b).

Watson et al. (2011) proposed to use this delay – luminosity relation to construct Hubble diagram in the same way as it is done for SN Ia. Our model justifies the use of this scaling also for distant quasars.

It was also suggested that simple multiband photometric monitoring can be used to determine the line delay (Haas et al. 2011; our method B). This is technically challenging but on the other hand, this type of data will be available for thousands of quasars when Large Synoptic Survey Telescope (LSST) starts to be operational.

Another method was proposed by Sulentic et al. (2014). It is based on a single spectrum, instead of time consuming reverberation method discussed above. In this case determination of the absolute luminosity has to be done at the basis of emission line ratios and line shapes (our method C). This method has extreme statistical potential since quasars with well measured emission line properties come in hundreds of thousands, and a considerable fraction of quasars will fall into this category.

Finally, a method based on the continuum variability in AGN has been proposed for determination of the Hubble constant even much earlier than the reverberation method (Collier et al. 1999, Cackett et al. 2007). However, first attempts failed, and actually there is a long-standing problem behind it. Accretion disk sizes determined either from continuum variability or from the microlensing are by a factor of 1.6 too large in comparison with the theory. In order to use this method (method D on our list) we have first to fix the underlying problem.

Work plan (outline of the work plan, critical paths, state of preliminary and initial research indicating feasibility of research objectives)

Our hypothesis is that active galaxies can be made an efficient direct tool in cosmology, allowing to trace the expansion rate of the Universe in a broad range of redshifts, from relatively small distances of several Mpc to redshifts of 6.

Observational samples

Cosmological constraints require broad coverage of the redshift. Low redshift coverage will come from published results and from our collaborators from China. To obtain high quality intermediate redshift data and to test the systematic errors we will continue our spectroscopic monitoring with 11-m Southern African Large Telescope (SALT). These extremely high quality data from allows to disentangle very precisely the continuum and the Mg II emission line (e.g. Modzelewska et al. 2014, Hryniewicz et al. 2014, Średzińska et al. 2017). Additional measurements will be coming from the SDSS group (preliminary results are in Shen et al. 2016, and the program is extended). Up to a hundred of measurements will come in a year or two from the Oz-DES quasar survey (King et al. 2015). Small sample of very high redshift sources is monitored in CIV line, one tentative measurement was obtained by Kaspi et al. (2007), and some firm time delay measurements will come soon (Paulina Lira, private communication). Finally, LSST will will bring detection of more than 10 million of quasars with redshifts up to about 7 (Ivezic et al. 2014), and for a fraction of them photometric lightcurves in six spectral bands will become available. Agnieszka Pollo, a Co-I of this project, is organizing Polish participation in LSST, and I am a Management Committee member of the COST action Big Data Era in Sky and Earth (TD1403), partially devoted to the most efficient use of this survey. This part will be done in collaboration with Zelijko Ivezic.

For single-spectrum methods we will simply use the SDSS catalog. The newest edition (DR12Q; Paris et al. 2017) contains 297 301 quasars, out of this number 184 101 have redshifts above 2.15. The catalog contains approximate results for the spectroscopic line measurements which simplyfies the selection of the required sub-classes of objects.

Method (A): Line time delay measurements from spectroscopy

The technical methods for time development are continuously being developed. Initially most astronomers used interpolated cross-correlation functions (ICCF) (see Gaskell & Peterson 1987), and some variants of this method. Later much attention got the methods based on random walk (e.g. code JAVELIN developed by Zu et al. 2011 is frequently used). In Czerny et al. (2013) we favored chi2 method. There is also a new method, recently advertised by Chelouche et al. (2017), based on randomness measurement. We thus plan to compare all available methods using both our SALT lightcurves and simulated lightcurves, in order to determine the most reliable method.

The study of the BLR structure will allow us to determine corrections due to the effects not included in the existing measurement of the time delay, and then to combine more reliably the existing measurements. As a pilot study, we show below the cosmological constraints based just on the quasar CTS C30.10 monitored by us with SALT, and on the combined time delay measurements from Du et al. (2015,2016), without correcting the values derived by the corresponding authors.

Fig. 2. Contour error (90% confidence level) for the determination of the cosmological parameters Omega_m and Omega_Lambda from a single quasar delay (CTS C30.10, z = 0.90052, delay 550 days in the rest frame) measured by us with SALT telescope (left panel), and the combination of all time delay measurements from a compilation from Du et al. (2016) (right panel). The use 67 measurements did not decrease the errors significantly since the measurements show large dispersion due to heterogeneous approach to the delay measurements by various groups and the negligence of the correcting factors which will be worked out in the current project.

We expect that the new methodology and new measurements included in the sample will decrease the errors.

Method(B): Photometric Line Reverberation

We plan to determine time delays using the LSST which will start its operation in 2020. The measurement of the time delay when only a few photometric band measurement with, and without line contamination, are available, is more complex than in the case of spectroscopy. We plan to try several methods and to include many effects, for example the change of the continuum slope with the source brightens. Statistical errors will be very small, but we plan to study the potential bias due to expected trends with black hole mass, Fe II contamination, and, most probably, with viewing angle which, on average, may increase as a function of redshift due to likely decrease of the dusty torus opening angle. Performing the simulations for the LSST operation in the context of AGN time delay will also help to adjust the cadence better for our purpose.

Method (C): Development of the single spectrum methods

High number of already available spectra of active galaxies makes this method extremely attractive but selection of sources radiating close to the Eddington ratio in a way which is not biased with increasing redshift is a challenge. We plan to develop further our model of the BLR in order to determine the trends with the Eddington ratio, but also with the black hole mass, spin, and the viewing angle. The model will form a base for selection of objects close to the Eddington ratio. Armed with the selection criterion, we will select suitable sources from SDSS catalog and calculate the constraints for the cosmological parameters.

Method (D): Development of the continuum reverberation method

We propose that the apparent large size of the disk is caused by scattering of the fraction of disk photons in extended hot fully ionized medium. We plan to perform the computations of the disk spectrum allowing for disk local variability in thermal timescale and including the outflowing medium. We will compare the amount of the hot outflow (column density, spacial distribution) with the polarization measurements of the spectra since scattering would induce some level of polarization.

Research Methodology (underlying scientific methodology, data reduction and treatment schemes, type and degree of access to the equipment to be used in the proposed research)

The planned research will be based on the observational data described above. The theoretical aspect of the projects will be done predominantly with the use of the software either already developed, or to be developed within the frame of the project.

Performance of the project will require:

– collection and systematic tests of various methods of the time delay measurement

– collecting available lightcurves and their re-analysis using the optimum methodology

– development of the BLR model, assignment of the corrections due to small but likely trends with the black hole mass and accretion rate in the position of the effective radius

– computations of the line profiles with the aim to select close to Eddington ratio sources. Most likely, such line shapes should contain a factor due to large turbulent component but strong asymmetry connected with outflow should not be present yet. However, quantitative criteria are needed, and those proposed by Sulentic et al. (2014) are not accurate enough for cosmological applications

– modelling the time-dependent disk spectra with the scattering in the circumnuclear region

– simulations of the LSST performance for the measurements of the line delays in LSST, with the aim to discuss the planned cadence of the visiting different sky regions. This requires combining the available code with the AGN luminosity function and predicted photometric accuracies, and folding the whole simulation with LSST cadence software.

Literature references

Abramowicz, M. A., Czerny, B., Lasota, J.P., Szuszkiewicz, E., et al. 1988, ApJ, 332, 646

Adam, R. et al., 2016, A&A, 596, A107

Ade, P.A.R. et al., 2016a, A&A, 594, A13

Ade, P.A.R. et al., 2016c, A&A, 594, A15

Ade, P.A.R. et al., 2016b, A&A, 594, A24

Adison, G.E. et al., 2017, ArXiv170706547

Alam, S. et al., 2017, MNRAS, 470, 2617

Anand, S. et al., 2017, ArXiv170807030

Bahcall, J. N., Kozlovsky, B.-Z., & Salpeter, E. E. 1972, ApJ, 171, 467

Barth, A.J. et al, 2015, ApJS, 217, 26

Bentz, M.C. et al., 2013, ApJ, 767, 149

Bentz, M.C. et al, 2016a, ApJ, 831, 2

Bentz, M.C. et al, 2016b, ApJ, 830, 136

Betoule, M. et al. 2014, A&A, 568, A22

Bonvin, V., 2017, MNRAS, 465, 4914

Cackett, E.M. et al, 2007, MNRAS, 380, 669

Chelouche, D., Pozo-Nunez, F., Zucker, S., 2017, ApJ, 844, 146

Chelouche, D. et al., 2014a, ApJ, 785, 140

Chelouche, D. et al., 2014b, ApJ, 783, L34

Cherepashchuk, A. M., & Lyutyi, V. M. 1973, Astrophys. Lett.,13, 165

Collier, S. et al., 1999, MNRAS, 302, L24

Collin, S., et al., 2006, A&A, 456, 75

Collin-Souffrin, S., 1986, A&A, 166, 115

Collin-Souffrin, S. et al., 1982, A&A, 106, 362

Czerny, B. et al., 2015, Adv. Space Res., 55, 1806

Czerny, B. et al., 2016, ApJ, 832, 15

Czerny, B. et al., 2017a, Space Science Review (submitted)

Czerny, B. et al., 2017b, ApJ, 846, 154

Czerny, B., Lehto, H., 1997, MNRAS, 285, 365

Czerny, B., Hryniewicz, K., 2011, A&A, 525, L8

Czerny, B., Hryniewicz, K., Maity, I., et al. 2013, A&A, 556, A97

Czerny, B., Schwarzenberg-Czerny, A., Loska, Z., 1999, MNRAS, 303, 148

Delubac, T. et al., BOSS collaboration, 2015, A&A, 574, A59

Demianski, M. et al., 2017, A&A, 598, A112

Dumont, A.-M. et al., 1998, A&A, 331, 11

Di Valentino, E., 2017, Nat. Astron., 1, 569,

Du, P., Lu, K. Zhang, Z. et al., 2016, ApJ, 825, 126

Du, P. et al., 2015, ApJ, 806, 22

Elvis, M., 2000, ApJ, 545, 63

Elvis, M. 2012, ASP Conference Series, Vol. 460, p. 186

Farooq, O. et al., 2017,ApJ, 835, 26

Fausnaugh, M.M. et al, 2016, MNRAS, 450, 3597

Gall, E.E.E. et al., 2016, A&A, 592, A129

Gallerani, S. et al., 2017, MNRAS, 467, 3590

Gaskell, C. M., & Peterson, B. M. 1987, ApJS, 65, 1

Hamman, F., Ferland, G., 1999, ARA&A, 37. 487

Haas, M. et al., 2011, A&A, 535, A73

Harris, D.W. et al., 2016, AJ, 151, 155

Heymans,et al., 2013, MNRAS, 432, 2433

Hill, A.R. et al., 2014, MNRAS, 438, 2317

Hryniewicz, K. et al., 2014, 562, A34

Hu, C. et al., 2015, ApJ, 804, 138

Huterer, D., Shafer, D.L., 2017, Rep. on Prog. Phys., arXiv170901091H

Ilic, D. et al, 2015, Journal of Astrophysics and Astronomy, 36, 433

Ivezic, Z. et al., 2014, in Proceedings IAU Symposium No. 304, 2014, A. Mickaelian, F. Aharonian & D. Sanders, eds.

Juarez, Y. et al., et al. 2009, A&A, 494, L25

Kaspi, S., Smith, P. S., Netzer, H., et al. 2000, ApJ, 533, 631

Kelly, B. C., Bechtold, J., & Siemiginowska, A. 2009, ApJ, 698, 895

King, A.L. et al., 2015, MNRAS, 453, 1701

Kofman, L.A., Gnedin, N.Y., Bahcall, N.A., 1993, ApJ, 413, 1

Lira, P. et al., 2016, in Active Galactic Nuclei: what’s in a name? Proceedings of a conference held 27 June – 1 July, 2016 in Garching, id.10

Lusso, E., Risaliti, G., 2016, ApJ, 819, 154

Modzelewska, J. et al., 2014, A&A, 570, A53

Murray, N., Chiang, J., 1998, ApJ, 494, 125

Mushotzky, R. F., Edelson, R., Baumgartner, W., & Gandhi, P., 2011, ApJ, 743, L12

Paris, I. et al., 2017, A&A, 597, A79

Peebles, P.J.E., 1984, ApJ, 284, 439

Perlmutter, S. et al., 1998, Natur., 391, 51

Peterson, B.M. 1997, An introduction to active galactic nuclei, Cambridge University Press

Peterson, B. M., Ferrarese, L., Gilbert, K. M., et al. 2004, ApJ, 613, 682

Proga, D. et al., 2000, ApJ, 543, 686

Proga, D. et al. 2004,

Rest, A. et al., 2014, ApJ, 795, 44

Riess, A.G. et al., 1998, AJ, 116, 1009

Riess, A.G. et al., 2016, ApJ, 826, 56

Risaliti, G. Lusso, E ., 2015 Astrophys. J. 815

Ross, A.J., Samushia, L, Howlett, C. et al., 2015, MNRAS, 449, 835

Sameshima. H., Yoshii, Y., Kawara, K., 2017, ApJ, 834, 203

Shakura, N.I., Sunyaev, R.A., 1973, A&A, 24, 337

Shariff, H., Jiao, X., Trotta, R., van Dyk, D.A., 2016, ApJ, 827, 1

Shen, Y. et al., 2016, ApJ, 818, 30

Średzińska, J. et al., 2017, 601, A32

Sulentic, J.W. et al., 2014, Journal of Physics: Conference Series, 565, article id. 012018

Timmer, J., Konig, M., 1995, A&A, 300, 707

Wandel, A. et al., 1999, 526, 579

Wang, J.-M. et al., 2014, ApJ, 793, 108

Waters, T. et al. 2016, ApJ, 827, 53

Watson, D. et al., 2011, ApJ, 740, L49

Zhao, G.-B. Et al., 2017, Nat. Astron., 1, 627

Zu, Y., Kochanek, C. S., & Peterson, B. M. 2011, ApJ, 735, 80