Keith Horne
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Keith Horne has a total of 319 co-authors. Here are the top co-authors by number of shared publications:
- David P. Bennett
NASA Goddard Space Flight Center
6 shared publications - Andrew Gould
Ohio State University
5 shared publications - Byeong-Gon Park
Korea Astronomy and Space Science Institute (KASI)
5 shared publications - Colin Snodgrass
Institute for Astronomy, University of Edinburgh, Royal Observatory
5 shared publications - G. W. Christie
Dept. of Physics University of Notre Dame 46556 Notre Dame IN USA
5 shared publications - Jennie McCormick
Nagano National College of Technology 381-8550 Nagano Japan M
5 shared publications
Latest Publications
- 21 December 2022
- Published by Oxford University Press (OUP) in Monthly Notices of the Royal Astronomical Society
- Vol. 519 (2) , 2754-2768
Rimmed and rippled accretion disc models to explain AGN continuum lags
We propose a solution to the problem of accretion disc sizes in active galactic nuclei being larger when measured by reverberation mapping than predicted by theory. Considering the disc’s exposed-surface thickness profile H(r), our solution invokes a steep rim or rippled structures irradiated by the central lamp-post. We model the continuum lags and the faint and bright disc spectral energy distribution (SED) in the best-studied case NGC 5548 (black hole mass M• = 7 × 107 M⊙, disc inclination i = 45○). With the lamp-post off, the faint-disc SED fixes a low accretion rate |$\dot{M}\simeq 0.0014~\mathrm{M}_\odot \, {\rm yr}^{-1}$| and high prograde black hole spin a• ≃ 0.93, for which rin = 2 G M•/c2 and |$L_{\rm disc}=0.25\, \dot{M}\, c^2$|. The bright-disc SED then requires a lamp-post luminosity |${L_{\rm LP}}\simeq 5\, \dot{M}\, c^2/(1-A)$| for disc albedo A. Reprocessing on the thin disc with T∝r−3/4 gives time lags τ∝λ4/3 but 3 times smaller than observed. Introducing a steep H(r) rim, or multiple crests, near r ∼ 5 light days, reprocessing on the steep centre-facing slope increases temperatures from ∼1500 K to ∼6000 K, and this increases optical lags to match the lag data. Most of the disc surface maintains the cooler T∝r−3/4 profile that matches the SED. The bright lamp-post may be powered by magnetic links tapping the black hole spin. The steep rim occurs near the disc’s dust sublimation radius as in the ‘failed disc wind model for broad-line clouds’. Lens-Thirring torques aligning the disc and black hole spin may also raise a warp and associated waves. In both scenarios, the small density scale height implied by the inferred value of H(r) suggests possible marginal gravitational instability in the disc.
- 21 December 2022
- Published by Oxford University Press (OUP) in Monthly Notices of the Royal Astronomical Society
- Vol. 519 (2) , 2754-2768
Rimmed and rippled accretion disc models to explain AGN continuum lags
We propose a solution to the problem of accretion disc sizes in active galactic nuclei being larger when measured by reverberation mapping than predicted by theory. Considering the disc’s exposed-surface thickness profile H(r), our solution invokes a steep rim or rippled structures irradiated by the central lamp-post. We model the continuum lags and the faint and bright disc spectral energy distribution (SED) in the best-studied case NGC 5548 (black hole mass M• = 7 × 107 M⊙, disc inclination i = 45○). With the lamp-post off, the faint-disc SED fixes a low accretion rate |$\dot{M}\simeq 0.0014~\mathrm{M}_\odot \, {\rm yr}^{-1}$| and high prograde black hole spin a• ≃ 0.93, for which rin = 2 G M•/c2 and |$L_{\rm disc}=0.25\, \dot{M}\, c^2$|. The bright-disc SED then requires a lamp-post luminosity |${L_{\rm LP}}\simeq 5\, \dot{M}\, c^2/(1-A)$| for disc albedo A. Reprocessing on the thin disc with T∝r−3/4 gives time lags τ∝λ4/3 but 3 times smaller than observed. Introducing a steep H(r) rim, or multiple crests, near r ∼ 5 light days, reprocessing on the steep centre-facing slope increases temperatures from ∼1500 K to ∼6000 K, and this increases optical lags to match the lag data. Most of the disc surface maintains the cooler T∝r−3/4 profile that matches the SED. The bright lamp-post may be powered by magnetic links tapping the black hole spin. The steep rim occurs near the disc’s dust sublimation radius as in the ‘failed disc wind model for broad-line clouds’. Lens-Thirring torques aligning the disc and black hole spin may also raise a warp and associated waves. In both scenarios, the small density scale height implied by the inferred value of H(r) suggests possible marginal gravitational instability in the disc.
- Open Access
Rimmed and Rippled Accretion Disc Models to Explain AGN Continuum Lags
We propose a solution to the problem of accretion disc sizes in active galactic nuclei being larger when measured by reverberation mapping than predicted by theory. Considering blackbody reprocessing on a disc with thickness profile $H(r)$, our solution invokes a steep rim or rippled structures irradiated by the central lamp-post. We model the continuum lags and the faint and bright disc spectral energy distribution (SED) in the best-studied case NGC 5548 (black hole mass $M = 7\times10^{7} M_\odot$, disc inclination $i=45^\circ$). With the lamp-post off, the observed disc SED requires a low accretion rate ($\dot{M} \sim 0.0014 M_\odot$/yr) and high prograde black hole spin ($a \sim 0.93$). Reprocessing on the thin disc gives time lags increasing with wavelength but 3 times smaller than observed. Introducing a steep $H(r)$ rim, or multiple crests, near $r = 5$ light days, reprocessing on their steep centre-facing slopes increases temperatures from $\sim1500$ K to $\sim6000$ K and this increases optical lags to match the lag data. Most of the disc surface maintains the cooler $T\propto r^{-3/4}$ temperature profile that matches the SED. The bright lamp-post may be powered by magnetic links tapping the black hole spin. The steep rim occurs near the sublimation radius for dust in the disc, as in the "failed disc wind model" for broad-line clouds. Lens-Thirring torques aligning the disc and black hole spin may also raise a warp and associated waves. In both scenarios, the small density scale height implied by the inferred value of $H(r)$ suggests possible marginal gravitational instability in the disc.Open AccessRimmed and Rippled Accretion Disc Models to Explain AGN Continuum Lags
We propose a solution to the problem of accretion disc sizes in active galactic nuclei being larger when measured by reverberation mapping than predicted by theory. Considering blackbody reprocessing on a disc with thickness profile $H(r)$, our solution invokes a steep rim or rippled structures irradiated by the central lamp-post. We model the continuum lags and the faint and bright disc spectral energy distribution (SED) in the best-studied case NGC 5548 (black hole mass $M = 7\times10^{7} M_\odot$, disc inclination $i=45^\circ$). With the lamp-post off, the observed disc SED requires a low accretion rate ($\dot{M} \sim 0.0014 M_\odot$/yr) and high prograde black hole spin ($a \sim 0.93$). Reprocessing on the thin disc gives time lags increasing with wavelength but 3 times smaller than observed. Introducing a steep $H(r)$ rim, or multiple crests, near $r = 5$ light days, reprocessing on their steep centre-facing slopes increases temperatures from $\sim1500$ K to $\sim6000$ K and this increases optical lags to match the lag data. Most of the disc surface maintains the cooler $T\propto r^{-3/4}$ temperature profile that matches the SED. The bright lamp-post may be powered by magnetic links tapping the black hole spin. The steep rim occurs near the sublimation radius for dust in the disc, as in the "failed disc wind model" for broad-line clouds. Lens-Thirring torques aligning the disc and black hole spin may also raise a warp and associated waves. In both scenarios, the small density scale height implied by the inferred value of $H(r)$ suggests possible marginal gravitational instability in the disc. - Open Access
Accretion Disc Time Lag Distributions: Applying CREAM to Simulated AGN Light Curves
Active Galactic Nuclei (AGN) vary in their brightness across all wavelengths. Moreover, longer wavelength ultraviolet - optical continuum light curves appear to be delayed with respect to shorter wavelength light curves. A simple way to model these delays is by assuming thermal reprocessing of a variable point source (a lamp post) by a blackbody accretion disc. We introduce a new method, CREAM (\textbf{C}ontinuum \textbf{RE}processed \textbf{A}GN \textbf{M}arkov Chain Monte Carlo), that models continuum variations using this lamp post model. The disc light curves lag the lamp post emission with a time delay distribution sensitive to the disc temperature-radius profile and inclination. We test CREAM's ability to recover both inclination and product of black hole mass and accretion rate $\mmdot$, and show that the code is also able to infer the shape of the driving light curve. CREAM is applied to synthetic light curves expected from 1000 second exposures of a 17th magnitude AGN with a 2m telescope in Sloan g and i bands with signal to noise of 500 - 900 depending on the filter and lunar phase. We also tests CREAM on poorer quality g and i light curves with SNR = 100. We find in the high SNR case that CREAM can recover the accretion disc inclination to within an uncertainty of 5 degrees and an $\mmdot$ to within 0.04 dex.Open AccessAccretion Disc Time Lag Distributions: Applying CREAM to Simulated AGN Light Curves
Active Galactic Nuclei (AGN) vary in their brightness across all wavelengths. Moreover, longer wavelength ultraviolet - optical continuum light curves appear to be delayed with respect to shorter wavelength light curves. A simple way to model these delays is by assuming thermal reprocessing of a variable point source (a lamp post) by a blackbody accretion disc. We introduce a new method, CREAM (\textbf{C}ontinuum \textbf{RE}processed \textbf{A}GN \textbf{M}arkov Chain Monte Carlo), that models continuum variations using this lamp post model. The disc light curves lag the lamp post emission with a time delay distribution sensitive to the disc temperature-radius profile and inclination. We test CREAM's ability to recover both inclination and product of black hole mass and accretion rate $\mmdot$, and show that the code is also able to infer the shape of the driving light curve. CREAM is applied to synthetic light curves expected from 1000 second exposures of a 17th magnitude AGN with a 2m telescope in Sloan g and i bands with signal to noise of 500 - 900 depending on the filter and lunar phase. We also tests CREAM on poorer quality g and i light curves with SNR = 100. We find in the high SNR case that CREAM can recover the accretion disc inclination to within an uncertainty of 5 degrees and an $\mmdot$ to within 0.04 dex. - Open Access
Kepler-like Multi-Plexing for Mass Production of Microlens Parallaxes
We show that a wide-field Kepler-like satellite in Solar orbit could obtain microlens parallaxes for several thousand events per year that are identified from the ground, yielding masses and distances for several dozen planetary events. This is roughly an order of magnitude larger than previously-considered narrow-angle designs. Such a satellite would, in addition, roughly double the number of planet detections (and mass/distance determinations). It would also yield a trove of brown-dwarf binaries with masses, and distances and (frequently) full orbits, enable new probes of the stellar mass function, identify isolated black-hole candidates. We show that the actual Kepler satellite, even with degraded pointing, can demonstrate these capabilities and make substantial initial inroads into the science potential. We discuss several "Deltas" to the Kepler satellite aimed at optimizing microlens parallax capabilities. Most of these would reduce costs. The wide-angle approach advocated here has only recently become superior to the old narrow-angle approach, due to the much larger number of ground-based microlensing events now being discovered.
Open AccessKepler-like Multi-Plexing for Mass Production of Microlens Parallaxes
We show that a wide-field Kepler-like satellite in Solar orbit could obtain microlens parallaxes for several thousand events per year that are identified from the ground, yielding masses and distances for several dozen planetary events. This is roughly an order of magnitude larger than previously-considered narrow-angle designs. Such a satellite would, in addition, roughly double the number of planet detections (and mass/distance determinations). It would also yield a trove of brown-dwarf binaries with masses, and distances and (frequently) full orbits, enable new probes of the stellar mass function, identify isolated black-hole candidates. We show that the actual Kepler satellite, even with degraded pointing, can demonstrate these capabilities and make substantial initial inroads into the science potential. We discuss several "Deltas" to the Kepler satellite aimed at optimizing microlens parallax capabilities. Most of these would reduce costs. The wide-angle approach advocated here has only recently become superior to the old narrow-angle approach, due to the much larger number of ground-based microlensing events now being discovered.
- Open Access
- 9 September 2013
- Published by Oxford University Press (OUP) in Monthly Notices of the Royal Astronomical Society
- Vol. 435 (4) , 3122-3124
A test of the failed disc wind scenario for the origin of the broad-line region in active galactic nuclei
It has been recently proposed that the broad-line region in active galactic nuclei originates from dusty clouds driven from the accretion disc by radiation pressure, at a distance from the black hole where the disc is cooler than the dust sublimation temperature. We test this scenario by checking the consistency of independent broad-line region and accretion disc reverberation measurements, for a sample of 11 well-studied active galactic nuclei. We show that independent disc and broad-line region reverberation mapping measurements are compatible with a universal disc temperature at the Hβ radius of T(RHβ) ≈ 1670 ± 231 K which is close to typical dust sublimation temperatures.
- 9 September 2013
- Published by Oxford University Press (OUP) in Monthly Notices of the Royal Astronomical Society
- Vol. 435 (4) , 3122-3124
Open AccessA test of the failed disc wind scenario for the origin of the broad-line region in active galactic nuclei
It has been recently proposed that the broad-line region in active galactic nuclei originates from dusty clouds driven from the accretion disc by radiation pressure, at a distance from the black hole where the disc is cooler than the dust sublimation temperature. We test this scenario by checking the consistency of independent broad-line region and accretion disc reverberation measurements, for a sample of 11 well-studied active galactic nuclei. We show that independent disc and broad-line region reverberation mapping measurements are compatible with a universal disc temperature at the Hβ radius of T(RHβ) ≈ 1670 ± 231 K which is close to typical dust sublimation temperatures.
- Open Access
A test of the failed disk wind scenario for the origin of the broad line region in active galactic nuclei
It has been recently proposed that the broad line region in active galactic nuclei originates from dusty clouds driven from the accretion disk by radiation pressure, at a distance from the black hole where the disk is cooler than the dust sublimation temperature. We test this scenario by checking the consistency of independent broad line region and accretion disk reverberation measurements, for a sample of 11 well studied active galactic nuclei. We show that independent disk and broad line region reverberation mapping measurements are compatible with a universal disk temperature at the H{\beta} radius of T[R(H{\beta})]=1670(231) K which is close to typical dust sublimation temperatures.
Open AccessA test of the failed disk wind scenario for the origin of the broad line region in active galactic nuclei
It has been recently proposed that the broad line region in active galactic nuclei originates from dusty clouds driven from the accretion disk by radiation pressure, at a distance from the black hole where the disk is cooler than the dust sublimation temperature. We test this scenario by checking the consistency of independent broad line region and accretion disk reverberation measurements, for a sample of 11 well studied active galactic nuclei. We show that independent disk and broad line region reverberation mapping measurements are compatible with a universal disk temperature at the H{\beta} radius of T[R(H{\beta})]=1670(231) K which is close to typical dust sublimation temperatures.
- Open Access
Kepler Microlens Planets and Parallaxes
Kepler's quest for other Earths need not end just yet: it remains capable of characterizing cool Earth-mass planets by microlensing, even given its degraded pointing control. If Kepler were pointed at the Galactic bulge, it could conduct a search for microlensing planets that would be virtually non-overlapping with ground-based surveys. More important, by combining Kepler observations with current ground-based surveys, one could measure the "microlens parallax" \pi_E for a large fraction of the known microlensing events. Such parallax measurements would yield mass and distance determinations for the great majority of microlensing planets, enabling much more precise study of the planet distributions as functions of planet and host mass, planet-host separation, and Galactic position (particularly bulge vs. disk). In addition, rare systems (such as planets orbiting brown dwarfs or black holes) that are presently lost in the noise would be clearly identified. In contrast to Kepler's current primary hunting ground of close-in planets, its microlensing planets would be in the cool outer parts of solar systems, generally beyond the snow line. The same survey would yield a spectacular catalog of brown-dwarf binaries, probe the stellar mass function in a unique way, and still have plenty of time available for asteroseismology targets.
Open AccessKepler Microlens Planets and Parallaxes
Kepler's quest for other Earths need not end just yet: it remains capable of characterizing cool Earth-mass planets by microlensing, even given its degraded pointing control. If Kepler were pointed at the Galactic bulge, it could conduct a search for microlensing planets that would be virtually non-overlapping with ground-based surveys. More important, by combining Kepler observations with current ground-based surveys, one could measure the "microlens parallax" \pi_E for a large fraction of the known microlensing events. Such parallax measurements would yield mass and distance determinations for the great majority of microlensing planets, enabling much more precise study of the planet distributions as functions of planet and host mass, planet-host separation, and Galactic position (particularly bulge vs. disk). In addition, rare systems (such as planets orbiting brown dwarfs or black holes) that are presently lost in the noise would be clearly identified. In contrast to Kepler's current primary hunting ground of close-in planets, its microlensing planets would be in the cool outer parts of solar systems, generally beyond the snow line. The same survey would yield a spectacular catalog of brown-dwarf binaries, probe the stellar mass function in a unique way, and still have plenty of time available for asteroseismology targets.
- Open Access
- 12 April 2011
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 529, A102
MOA-2009-BLG-387Lb: a massive planet orbiting an M dwarf
Aims. We report the discovery of a planet with a high planet-to-star mass ratio in the microlensing event MOA-2009-BLG-387, which exhibited pronounced deviations over a 12-day interval, one of the longest for any planetary event. The host is an M dwarf, with a mass in the range 0.07 M ⊙ < M host < 0.49 M ⊙ at 90% confidence. The planet-star mass ratio q = 0.0132 ± 0.003 has been measured extremely well, so at the best-estimated host mass, the planet mass is m p = 2.6 Jupiter masses for the median host mass, M = 0.19 M ⊙.Methods. The host mass is determined from two “higher order” microlensing parameters. One of these, the angular Einstein radius θ E = 0.31 ± 0.03 mas has been accurately measured, but the other (the microlens parallax π E, which is due to the Earth’s orbital motion) is highly degenerate with the orbital motion of the planet. We statistically resolve the degeneracy between Earth and planet orbital effects by imposing priors from a Galactic model that specifies the positions and velocities of lenses and sources and a Kepler model of orbits.Results. The 90% confidence intervals for the distance, semi-major axis, and period of the planet are 3.5 kpc < D L < 7.9 kpc, 1.1 AU < a < 2.7 AU, and 3.8 yr < P < 7.6 yr, respectively.
- 12 April 2011
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 529, A102
Open AccessMOA-2009-BLG-387Lb: a massive planet orbiting an M dwarf
Aims. We report the discovery of a planet with a high planet-to-star mass ratio in the microlensing event MOA-2009-BLG-387, which exhibited pronounced deviations over a 12-day interval, one of the longest for any planetary event. The host is an M dwarf, with a mass in the range 0.07 M ⊙ < M host < 0.49 M ⊙ at 90% confidence. The planet-star mass ratio q = 0.0132 ± 0.003 has been measured extremely well, so at the best-estimated host mass, the planet mass is m p = 2.6 Jupiter masses for the median host mass, M = 0.19 M ⊙.Methods. The host mass is determined from two “higher order” microlensing parameters. One of these, the angular Einstein radius θ E = 0.31 ± 0.03 mas has been accurately measured, but the other (the microlens parallax π E, which is due to the Earth’s orbital motion) is highly degenerate with the orbital motion of the planet. We statistically resolve the degeneracy between Earth and planet orbital effects by imposing priors from a Galactic model that specifies the positions and velocities of lenses and sources and a Kepler model of orbits.Results. The 90% confidence intervals for the distance, semi-major axis, and period of the planet are 3.5 kpc < D L < 7.9 kpc, 1.1 AU < a < 2.7 AU, and 3.8 yr < P < 7.6 yr, respectively.
- Open Access
Limits on the Orbits and Masses of Moons around Currently-Known Transiting Exoplanets
Aims. Current and upcoming space missions may be able to detect moons of transiting extra-solar planets. In this context it is important to understand if exomoons are expected to exist and what their possible properties are. Methods. Using estimates for the stability of exomoon orbits from numerical studies, a list of 87 known transiting exoplanets is tested for the potential to host large exomoons. Results. For 92% of the sample, moons larger than Luna can be excluded on prograde orbits, unless the parent exoplanet's internal structure is very different from the gas-giants of the solar system. Only WASP-24b, OGLE2-TR-L9, CoRoT-3b and CoRoT-9b could have moons above 0.4 m\oplus, which is within the likely detection capabilities of current observational facilities. Additionally, the range of possible orbital radii of exomoons of the known transiting exoplanets, with two exceptions, is below 8 Jupiter-radii and therefore rather small.
Open AccessLimits on the Orbits and Masses of Moons around Currently-Known Transiting Exoplanets
Aims. Current and upcoming space missions may be able to detect moons of transiting extra-solar planets. In this context it is important to understand if exomoons are expected to exist and what their possible properties are. Methods. Using estimates for the stability of exomoon orbits from numerical studies, a list of 87 known transiting exoplanets is tested for the potential to host large exomoons. Results. For 92% of the sample, moons larger than Luna can be excluded on prograde orbits, unless the parent exoplanet's internal structure is very different from the gas-giants of the solar system. Only WASP-24b, OGLE2-TR-L9, CoRoT-3b and CoRoT-9b could have moons above 0.4 m\oplus, which is within the likely detection capabilities of current observational facilities. Additionally, the range of possible orbital radii of exomoons of the known transiting exoplanets, with two exceptions, is below 8 Jupiter-radii and therefore rather small.
- 2 February 2010
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 710 (2) , 1641-1653
A COLD NEPTUNE-MASS PLANET OGLE-2007-BLG-368Lb: Cold neptunes are common
We present the discovery of a Neptune-mass planet OGLE-2007-BLG-368Lb with a planet-star mass ratio of q=[9.5 +/- 2.1] x 10^{-5} via gravitational microlensing. The planetary deviation was detected in real-time thanks to the high cadence of the MOA survey, real-time light curve monitoring and intensive follow-up observations. A Bayesian analysis returns the stellar mass and distance at M_l = 0.64_{-0.26}^{+0.21} M_\sun and D_l = 5.9_{-1.4}^{+0.9} kpc, respectively, so the mass and separation of the planet are M_p = 20_{-8}^{+7} M_\oplus and a = 3.3_{-0.8}^{+1.4} AU, respectively. This discovery adds another cold Neptune-mass planet to the planetary sample discovered by microlensing, which now comprise four cold Neptune/Super-Earths, five gas giant planets, and another sub-Saturn mass planet whose nature is unclear. The discovery of these ten cold exoplanets by the microlensing method implies that the mass ratio function of cold exoplanets scales as dN_{\rm pl}/d\log q \propto q^{-0.7 +/- 0.2} with a 95% confidence level upper limit of n < -0.35 (where dN_{\rm pl}/d\log q \propto q^n). As microlensing is most sensitive to planets beyond the snow-line, this implies that Neptune-mass planets are at least three times more common than Jupiters in this region at the 95% confidence level.
- 2 February 2010
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 710 (2) , 1641-1653
A COLD NEPTUNE-MASS PLANET OGLE-2007-BLG-368Lb: Cold neptunes are common
We present the discovery of a Neptune-mass planet OGLE-2007-BLG-368Lb with a planet-star mass ratio of q=[9.5 +/- 2.1] x 10^{-5} via gravitational microlensing. The planetary deviation was detected in real-time thanks to the high cadence of the MOA survey, real-time light curve monitoring and intensive follow-up observations. A Bayesian analysis returns the stellar mass and distance at M_l = 0.64_{-0.26}^{+0.21} M_\sun and D_l = 5.9_{-1.4}^{+0.9} kpc, respectively, so the mass and separation of the planet are M_p = 20_{-8}^{+7} M_\oplus and a = 3.3_{-0.8}^{+1.4} AU, respectively. This discovery adds another cold Neptune-mass planet to the planetary sample discovered by microlensing, which now comprise four cold Neptune/Super-Earths, five gas giant planets, and another sub-Saturn mass planet whose nature is unclear. The discovery of these ten cold exoplanets by the microlensing method implies that the mass ratio function of cold exoplanets scales as dN_{\rm pl}/d\log q \propto q^{-0.7 +/- 0.2} with a 95% confidence level upper limit of n < -0.35 (where dN_{\rm pl}/d\log q \propto q^n). As microlensing is most sensitive to planets beyond the snow-line, this implies that Neptune-mass planets are at least three times more common than Jupiters in this region at the 95% confidence level.