The conference poster includes a very apt phrase that describes a primary motivation for this conference: Time discovers truth. This aphorism, attributed to Seneca, was certainly affirmed by the many exciting talks and discussions at this conference, in both formal and informal settings.

19 Workshops were held during IAU S285. 15 submitted reports of the discussions that took place, while for the remaining 4 we have reproduced the summaries that were available on our wiki prior to the Symposium.

Space-based observations of variable stars have revolutionized the field of variability studies. Dedicated satellites such as the CoRoT and Kepler missions have duty cycles which are unachievable from the ground, and effectively solve many of the aliasing problems prevalent in ground-based observation campaigns. Moreover, the location above the Earth's atmosphere eliminates a major source of scatter prevalent in observations from the ground. These two major improvements in instrumentation have triggered significant increases in our knowledge of the stars, but in order to reap the full benefits they are also obliging the community to adopt more efficient techniques for handling, analysing and interpreting the vast amounts of new, high-precision data in an effective yet comprehensive manner. This workshop heard an outline of the history and development of asteroseismology, and descriptions of the two space missions (CoRoT and Kepler) which have been foremost in accelerating those recent developments. Informal discussions on numerous points peppered the proceedings, and involved the whole audience at times. The conclusions which the workshop reached have been distilled into a list of seven recommendations (Section 5) for the asteroseismology community to study and absorb. In fact, while addressing activities (such as stellar classification or analysing and modelling light curves) that could be regarded as specific to the community in question, the recommendations include advice on matters such as improving communication, incorporating trans-disciplinary knowledge and involving the non-scientific public that are broad enough to serve as guidelines for the astrophysical community at large.

A focus session was held for those wanting to familiarise themselves with the Swift satellite and to consider its exploitation for specific scientific goals. An overview was presented, with questions throughout. Proposal preparation and and the automated science products from the X-ray Telescope were discussed. This account summarises the information given in the presentation and in the answers to the questions which were raised.

A workshop on Optical & Near Infrared Transients took place during the first afternoon of the Symposium. It ran for two sessions. The first was given over to talks about various current optical and near-infrared transient surveys, focussing on the Vista surveys, the Catalina Real-Time Transient Survey, Pan-STARRS, Gaia, TAOS and TAOS2. The second session was a panel-led discussion about coordinating multi-wavelength surveys and associated follow-ups.

The gravitational-wave window onto the universe will open in roughly five years, when Advanced LIGO and Virgo achieve the first detections of high-frequency gravitational waves, most likely coming from compact binary mergers. Electromagnetic follow-up of these triggers, using radio, optical, and high energy telescopes, promises exciting opportunities in multi-messenger time-domain astronomy. In the decade, space-based observations of low-frequency gravitational waves from massive black hole mergers, and their electromagnetic counterparts, will open up further vistas for discovery. This two-part workshop featured brief presentations and stimulating discussions on the challenges and opportunities presented by gravitational-wave astronomy. Highlights from the workshop, with the emphasis on strategies for electromagnetic follow-up, are presented in this report.

Modern X-ray observatories yield unique insight into the astrophysical time domain. Each X-ray photon can be assigned an arrival time, an energy and a sky position, yielding sensitive, energy-dependent light curves and enabling time-resolved spectra down to millisecond time-scales. Combining those with multiple views of the same patch of sky (e.g., in the Chandra and XMM-Newton deep fields) so as to extend variability studies over longer baselines, the spectral timing capacity of X-ray observatories then stretch over 10 orders of magnitude at spatial resolutions of arcseconds, and 13 orders of magnitude at spatial resolutions of a degree. A wealth of high-energy time-domain data already exists, and indicates variability on timescales ranging from microseconds to years in a wide variety of objects, including numerous classes of AGN, high-energy phenomena at the Galactic centre, Galactic and extra-Galactic X-ray binaries, supernovæ, gamma-ray bursts, stellar flares, tidal disruption flares, and as-yet unknown X-ray variables. This workshop explored the potential of strategic X-ray surveys to probe a broad range of astrophysical sources and phenomena. Here we present the highlights, with an emphasis on the science topics and mission designs that will drive future discovery in the X-ray time domain.

Gravitational microlensing is a well established and unique field of time-domain astrophysics. For two decades microlensing surveys have been regularly observing millions of stars to detect elusive events that follow a characteristic Paczyński lightcurve. This workshop reviewed the current state of the field, and covered the major topics related to microlensing: searches for extrasolar planets, and studies of dark matter. There were also discussions of issues relating to the organisation of follow-up observations for microlensing, as well as serendipitous scientific outcomes resulting from extensive microlensing data.

The first “light echo”—scattered light from a stellar outburst arriving at the Earth months or years after the direct light from the event—was detected more than 100 years ago, around Nova Persei 1901. Renewed interest in light echoes has come from the spectacular echo around V838 Monocerotis, and from discoveries of light echoes from historical and prehistorical supernovæ in the Milky Way and Large Magellanic Cloud as well as from the 19th-century Great Eruption of η Carinae. A related technique is reverberation mapping of active galactic nuclei. This report of a workshop on Light Echoes gives an introduction to light echoes, and summarizes presentations on discoveries of light echoes from historical and prehistorical events, light and shadow echoes around R CrB stars, and reverberation mapping.

Just as the astronomical “Time Domain” is a catch-phrase for a diverse group of different science objectives involving time-varying phenomena in all astrophysical régimes from the solar system to cosmological scales, so the “Virtual Observatory” is a complex set of community-wide activities from archives to astroinformatics. This workshop touched on some aspects of adapting and developing those semantic and network technologies in order to address transient and time-domain research challenges. It discussed the VOEvent format for representing alerts and reports on celestial transient events, the SkyAlert and ATELstream facilities for distributing these alerts, and the IVOA time-series protocol and time-series tools provided by the VAO. Those tools and infrastructure are available today to address the real-world needs of astronomers.

Astrotomography refers to a suite of indirect imaging techniques that achieve micro-arcsecond angular resolution by measuring projections obtained from time-resolved spectroscopic observations. The projections arise from Doppler shifts, eclipses or time delays, combined with rotation of the star or binary system being imaged. At our workshop we reviewed and discussed state-of-the-art methods for imaging the surfaces and magnetic fields of rapidly rotating stars, the accretion flows in compact binary star systems and the broad emission-line regions in active galactic nuclei.

We have entered an era in time-domain astronomy in which the detected rate of explosive transients and important ephemeral states in persistent objects threatens to overwhelm the world's supply of traditional follow-up telescopes. As new, comprehensive time-domain surveys become operational and wide-field multi-messenger observatories come on-line, that problem will become more acute. The goal of this workshop was to foster discussion about how autonomous robotic telescopes and small-aperture conventional telescopes can be employed in the most effective ways to help deal with the coming deluge of scientifically interesting follow-up opportunities. Discussion topics included the role of event brokers, automated event triage, the establishment of cooperative global telescope networks, and real-time coordination of observations at geographically diverse sites. It therefore included brief overviews of the current diverse landscape of telescopes and their interactions, and also considered planned and potential new facilities and operating models.

Models of stellar evolution constitute an extremely powerful, and for the most part apparently very successful, tool for understanding the progression of a star through its lifetime as a fairly compact entity of incandescent gas. That success has led to stellar evolution theory becoming a crutch when an observer is faced with objects whose provenance or current state are in some way puzzling, but how safe a crutch? The validity of the theory is best checked by examining binary systems whose component parameters have been determined with high precision, but it can be (and needs to be) honed through the many challenges which non-conformist single stars and triple systems also present. Unfortunately the lever of observational parameters to constrain or challenge stellar evolution theory is not as powerful as it could be, because not all determinations of stellar parameters for the same systems agree to within the precisions claimed by their respective authors. What are the sources of bias—the data, the instrument or the techniques? The workshop was invited to discuss particularly challenging cases, and to attempt to identify how and where progress might be pursued.

The workshop on Historical Time-Domain Astronomy (TDA) was attended by a near-capacity gathering of ~30 people. From information provided in turn by those present, an up-to-date overview was created of available plate archives, progress in their digitization, the extent of actual processing of those data, and plans for data distribution. Several recommendations were made for prioritising the processing and distribution of historical TDA data.

The workshop on Data Management issues for Time-Domain Astronomy was conceived as a forward-looking discussion of the primary issues that need to be addressed for science in the time domain. The very broad diversity of the science areas presented in the main Symposium made it clear that most of the general issues for astronomy data management—for example, large data volumes, the need for timely processing and network performance—would be pertinent in the time domain. In addition, there might be other tight time constraints on data processing when the output was required to trigger rapid follow-up observations, while science based on very long time-baselines might require careful consideration of long-term data preservation and availability issues. But broadly speaking, data management challenges in the time domain are not at variance to any significant degree with those for astronomy or data-intensive research in general. The workshop framed and debated a number of questions: What is the biggest challenge faced by future projects? How do grid and cloud computing figure in data management plans? Is the Virtual Observatory important to future projects? How are the issues of data life cycle being addressed?

Citizen Science is the act of collecting or analyzing data by enlisting the help of volunteers who may have no specific scientific training. The workshop discussed how “Citizen Science” fits into time-domain astronomy, what the roles of such volunteers might be, and how amateur astronomers can help in the new era of surveys.

The past year has seen major advances in the observational status of Stellar Tidal Disruption, with the discovery of two strong optical candidates in archived SDSS data and the real-time X-ray detection of Swift J1644+57, plus rapid radio and optical follow-up establishing it as a probable Tidal Disruption Flare (TDF) in “blazar mode”. These observations motivated a workshop devoted to discussion of such events and of the theory of their emission and flare rate. Observational contributions included a presentation of Swift J2058+05 (a possible second example of a TDF in blazar mode), reports on the late-time evolution and X-ray variability of the two Swift events, and a proposal that additional candidates may be evidenced by spectral signatures in SDSS. Theory presentations included models of radio emission, theory of light curves and the proposal that GRB101225A may be the Galactic tidal disruption of a neutron star, an interpretation of Swift J1644+57 as due to the disruption of a white dwarf instead of main-sequence star, calculation of the dependence of the TDF rate on the spin of the black hole, and analysis of the SDSS events, fitting their SEDs to profiles of thoretical emission from accretion disks and showing that their luminosity and rate are consistent with the proposal that TDEs can be responsible for UHECR acceleration.

In the past five years systematic searches, serendipitous discoveries and archival searches have yielded over a dozen transients that are brighter than novæ but fainter than supernovæ. The observed properties of these “gap” transients tend to place them in distinct classes, some being about 100 times brighter than novæ and durations of nearly 100 days (e.g., M85 OT, PTF10fqs), while others (such as SN2002bj and PTF10bhp) nearly reach supernovæ luminosities but fade in five days. The state of theoretical understanding varies substantially across the class of objects, and is ripe for progress.

Never before has there been such a wealth of versatile ground- and space-based facilities with which to detect variable emission across the electromagnetic spectrum and beyond, to non-EM signals such as neutrinos and gravitational waves, to probe the most extreme phenomena in the Universe. The variable sky is already providing a wealth of new and surprising observations of phenomena such as GRBs, SNe and AGN that are pushing current theories beyond the state of the art. Multi-messenger follow-up will soon become de rigeur, and upcoming radio and optical all-sky transient surveys will revolutionise the study of the transient Universe. In addition to the technical and data challenges presented by such surveys, a major new challenge will be the interpretation of the wealth of available data and the identification of the underlying physics of new classes of variable (and potentially exotic) objects. Theoretical predictions will be vital for interpreting these future transient discoveries.

The goal of this workshop was to bring together theorists and observers in order to identify unexplored synergies across three main research areas of extreme physics: gamma-ray bursts, supernovæ and, more generically, relativistic jets. It aimed to discuss key outstanding questions in these rapidly moving fields, such as the composition and acceleration of GRB and AGN jets, GRB progenitors and central engines, the origin of the wide range of observed variability time-scales in GRB prompt and after-glow light curves and related cosmological applications, the physics of the newly-discovered ultra-luminous SN-like optical transients—as well as to speculate on what we might hope to discover with future technology.

The workshop absorbed two 90-minute sessions, selecting 3 main science topics (Relativistic Jets, GRBs and SNe) which it organised as structured discussions driven by a series of short but provocative questions. The final session featured a panel-led debate but with full audience participation.

This Workshop covered the four major subjects listed below in two 90-minute sessions. Each talk or tutorial allowed questions, and concluded with a discussion.

Classification: Automatic classification using machine-learning methods is becoming a standard in surveys that generate large datasets. Ashish Mahabal (Caltech) reviewed various methods, and presented examples of several applications.

Time-Series Modelling: Suzanne Aigrain (Oxford University) discussed autoregressive models and multivariate approaches such as Gaussian Processes.

Meta-classification/mixture of expert models: Karim Pichara (Pontificia Universidad Católica, Chile) described the substantial promise which machine-learning classification methods are now showing in automatic classification, and discussed how the various methods can be combined together.

Event Detection: Pavlos Protopapas (Harvard) addressed methods of fast identification of events with low signal-to-noise ratios, enlarging on the characterization and statistical issues of low signal-to-noise ratios and rare events.