Search

We present the analysis of the first circumbinary planet microlensing event, OGLE-2007-BLG-349. This event has a strong planetary signal that is best fit with a mass ratio of $q \approx 3.4\times10^{-4}$, but there is an additional signal due to an additional lens mass, either another planet or another star. We find acceptable light curve fits with two classes of models: 2-planet models (with a single host star) and circumbinary planet models. The light curve also reveals a significant microlensing parallax effect, which constrains the mass of the lens system to be $M_L \approx 0.7 M_\odot$. Hubble Space Telescope images resolve the lens and source stars from their neighbors and indicate excess flux due to the star(s) in the lens system. This is consistent with the predicted flux from the circumbinary models, where the lens mass is shared between two stars, but there is not enough flux to be consistent with the 2-planet, 1-star models. So, only the circumbinary models are consistent with the HST data. They indicate a planet of mass $m_c = 80\pm 13\,M_\oplus$, orbiting a pair of M-dwarfs with masses of $M_A = 0.41\pm 0.07 M_\odot$ and $M_B = 0.30\pm 0.07 M_\oplus$, which makes this the lowest mass circumbinary planet system known. The ratio of the separation between the planet and the center-of-mass to the separations of the two stars is $\sim 40$, so unlike most of the circumbinary planets found by Kepler, the planet does not orbit near the stability limit., Comment: 34 pages, with 9 figures. Published in the Astronomical Journal

We present a new analysis of the Jupiter+Saturn analog system, OGLE-2006-BLG-109Lb,c, which was the first double planet system discovered with the gravitational microlensing method. This is the only multi-planet system discovered by any method with measured masses for the star and both planets. In addition to the signatures of two planets, this event also exhibits a microlensing parallax signature and finite source effects that provide a direct measure of the masses of the star and planets, and the expected brightness of the host star is confirmed by Keck AO imaging, yielding masses of M_* = 0.51(+0.05-0.04) M_sun, M_b = 231+-19 M_earth, M_c = 86+-7 M_earth. The Saturn-analog planet in this system had a planetary light curve deviation that lasted for 11 days, and as a result, the effects of the orbital motion are visible in the microlensing light curve. We find that four of the six orbital parameters are tightly constrained and that a fifth parameter, the orbital acceleration, is weakly constrained. No orbital information is available for the Jupiter-analog planet, but its presence helps to constrain the orbital motion of the Saturn-analog planet. Assuming co-planar orbits, we find an orbital eccentricity of eccentricity = 0.15 (+0.17-0.10) and an orbital inclination of i = 64 (+4-7) deg. The 95% confidence level lower limit on the inclination of i > 49 deg. implies that this planetary system can be detected and studied via radial velocity measurements using a telescope of >30m aperture., Comment: 48 pages including 10 figures, to be published in ApJ

We report a measurement of the shape of the source star in microlensing event MOA 2002-BLG-33. The lens for this event was a close binary whose centre-of-mass passed almost directly in front of the source star. At this time, the source star was closely bounded on all sides by a caustic of the lens. This allowed the oblateness of the source star to be constrained. We found that a/b = 1.02^{+0.04}_{-0.02} where a and b are its semi-major and semi-minor axes respectively. The angular resolution of this measurement is approximately 0.04 microarcsec. We also report HST images of the event that confirm a previous identification of the source star as an F8-G2 turn-off main-sequence star., Comment: 8 pages, 5 figures, Accepted by A&A

We report a measurement of limb darkening of a solar-like star in the very high magnification microlensing event MOA 2002-BLG-33. A 15 hour deviation from the light curve profile expected for a single lens was monitored intensively in V and I passbands by five telescopes spanning the globe. Our modelling of the light curve showed the lens to be a close binary system whose centre-of-mass passed almost directly in front of the source star. The source star was identified as an F8-G2 main sequence turn-off star. The measured stellar profiles agree with current stellar atmosphere theory to within ~4% in two passbands. The effective angular resolution of the measurements is <1 micro-arcsec. These are the first limb darkening measurements obtained by microlensing for a Solar-like star., Comment: Accepted for publication in A&A Letters. 5 pages, 2 embedded colour ps figures plus 1 jpg figure. Version with all figures embedded available from: http://www.roe.ac.uk/~iab/moa33paper/

The Galactic Exoplanet Survey Telescope (GEST) will observe a 2 square degree field in the Galactic bulge to search for extra-solar planets using a gravitational lensing technique. This gravitational lensing technique is the only method employing currently available technology that can detect Earth-mass planets at high signal-to-noise, and can measure the frequency of terrestrial planets as a function of Galactic position. GEST's sensitivity extends down to the mass of Mars, and it can detect hundreds of terrestrial planets with semi-major axes ranging from 0.7 AU to infinity. GEST will be the first truly comprehensive survey of the Galaxy for planets like those in our own Solar System., Comment: 17 pages with 13 figures, to be published in Proc. SPIE vol 4854, "Future EUV-UV and Visible Space Astrophysics Missions and Instrumentation"

We present an re-analysis of the longest timescale gravitational microlensing event discovered to date: MACHO-99-BLG-22/OGLE-1999-BUL-32, which was discovered by both the MACHO and OGLE microlensing alert systems. Our analysis of this microlensing parallax event includes a likelihood analysis of the lens position based upon a standard model of the Galactic velocity distribution, and this implies that the lens could be a black hole of ~100 M_solar at a distance of a few hundred parsecs in the Galactic disk or a massive stellar remnant (black hole or neutron star) in the Galactic bulge. Our new analysis includes data from the MACHO, GMAN, and MPS collaborations in addition to the OGLE data used in a previous analysis by Mao et al (2002). The crucial feature that distinguishes our analysis from that of Mao et al is an accurate constraint on the direction of lens motion and an analysis of the implications of this direction., Comment: 11 pages with 2 color Figures included

We present an analysis of the longest timescale microlensing events discovered by the MACHO Collaboration during a 7 year survey of the Galactic bulge. We find 6 events that exhibit very strong microlensing parallax signals due, in part, to accurate photometric data from the GMAN and MPS collaborations. The microlensing parallax fit parameters are used in a likelihood analysis, which is able to estimate the distance and masses of the lens objects based upon a standard model of the Galactic velocity distribution. This analysis indicates that the most likely masses of 5 of the 6 lenses are > 1 Msun, which suggests that a substantial fraction of the Galactic lenses may be massive stellar remnants. This could explain the observed excess of long timescale microlensing events. The lenses for events MACHO-96-BLG-5 and MACHO-98-BLG-6 are the most massive, with mass estimates of M/Msun = 6 +10/-3 and M/Msun = 6 +7/-3, respectively. The observed upper limits on the absolute brightness of main sequence stars for these lenses are < 1 Lsun, so both lenses are black hole candidates. The black hole interpretation is also favored by a likelihood analysis with a Bayesian prior using a conventional model for the lens mass function. We consider the possibility that the source stars for some of these 6 events may lie in the foreground or background of the Galactic bulge, but we find that this is unlikely. Future HST observations of these events can either confirm the black hole lens hypothesis or detect the lens stars and provide a direct measurement of their masses. Future observations of similar events by SIM or the Keck or VLTI interferometers will allow direct measurements of the lens masses for stellar remnant lenses as well., Comment: 47 pages, with 26 included postscript figures. Includes a new likelihood analysis with a mass function prior

Planetary systems toward the Galactic Bulge can be detected through microlensing measurements. The microlensing planet search technique has some unique merits: low-mass planets can be detected from the ground; the Galactic family of planetary systems can be sampled unbiased; the time scale for the completion of each event is relatively short. The unambiguity of the underlying low-multiplicity point lenses is an indispensable element that allows robust interpretaion of the events. We have found the first circumbinary planet in the microlensing event MACHO-97-BLG-41. We have found evidence of a low mass planet (few Earths to Neptune mass) in a very high magnification event MACHO-98-BLG-35. We emphasize the necessity for coherent searches of microlensing planets with a network of instruments in the southern hemisphere. The network can also be vitally instrumental for follow-up observations of SN1a's and GRB's which share with microlensing planets the transiency as well as the clues to our quest of the origins., Comment: Proceedings, "Gravitational Lensing: Recent Progress and Future Goals", 25-30 July 1999, Boston University

The study of extra-solar planetary systems has emerged as a new discipline of observational astronomy in the past few years with the discovery of a number of extra-solar planets. The properties of most of these extra-solar planets were not anticipated by theoretical work on the formation of planetary systems. Here we report observations and light curve modeling of gravitational microlensing event MACHO-97-BLG-41, which indicates that the lens system consists of a planet orbiting a binary star system. According to this model, the mass ratio of the binary star system is 3.8:1 and the stars are most likely to be a late K dwarf and an M dwarf with a separation of about 1.8 AU. A planet of about 3 Jupiter masses orbits this system at a distance of about 7 AU. If our interpretation of this light curve is correct, it represents the first discovery of a planet orbiting a binary star system and the first detection of a Jovian planet via the gravitational microlensing technique. It suggests that giant planets may be common in short period binary star systems., Comment: 11 pages, with 1 color and 2 b/w Figures included (published version)

We fit the data for the binary-lens microlensing event MACHO 98-SMC-1 from 5 different microlensing collaborations and find two distinct solutions characterized by binary separation d and mass ratio q: (d,q)=(0.54,0.50) and (d,q)=(3.65,0.36), where d is in units of the Einstein radius. However, the relative proper motion of the lens is very similar in the two solutions, 1.30 km/s/kpc and 1.48 km/s/kpc, thus confirming that the lens is in the Small Magellanic Cloud. The close binary can be either rotating or approximately static but the wide binary must be rotating at close its maximum allowed rate to be consistent with all the data. We measure limb-darkening coefficients for five bands ranging from I to V. As expected, these progressively decrease with rising wavelength. This is the first measurement of limb darkening for a metal-poor A star., Comment: 29 pages + 9 figures + 2 tables, submitted to ApJ

We present observations of microlensing event MACHO-98-BLG-35 which reached a peak magnification factor of almost 80. These observations by the Microlensing Planet Search (MPS) and the MOA Collaborations place strong constraints on the possible planetary system of the lens star and show intriguing evidence for a low mass planet with a mass fraction $4\times 10^{-5} \leq \epsilon \leq 2\times 10^{-4}$. A giant planet with $\epsilon = 10^{-3}$ is excluded from 95% of the region between 0.4 and 2.5 $R_E$ from the lens star, where $R_E$ is the Einstein ring radius of the lens. This exclusion region is more extensive than the generic "lensing zone" which is $0.6 - 1.6 R_E$. For smaller mass planets, we can exclude 57% of the "lensing zone" for $\epsilon = 10^{-4}$ and 14% of the lensing zone for $\epsilon = 10^{-5}$. The mass fraction $\epsilon = 10^{-5}$ corresponds to an Earth mass planet for a lensing star of mass $\sim 0.3 \msun$. A number of similar events will provide statistically significant constraints on the prevalence of Earth mass planets. In order to put our limits in more familiar terms, we have compared our results to those expected for a Solar System clone averaging over possible lens system distances and orientations. We find that such a system is ruled out at the 90% confidence level. A copy of the Solar System with Jupiter replaced by a second Saturn mass planet can be ruled out at 70% confidence. Our low mass planetary signal (few Earth masses to Neptune mass) is significant at the $4.5\sigma$ confidence level. If this planetary interpretation is correct, the MACHO-98-BLG-35 lens system constitutes the first detection of a low mass planet orbiting an ordinary star without gas giant planets., Comment: ApJ, April 1, 2000; 27 pages including 8 color postscript figures

We present the observations of the binary lensing event MACHO-98-SMC-1 conducted at the Mt.~Stromlo 74" telescope by the Microlensing Planet Search (MPS) collaboration. The MPS data constrain the first caustic crossing to have occurred after 1998 June 5.55 UT and thus directly rule out one of the two fits presented by the PLANET collaboration (model II). This substantially reduces the uncertainty in the the relative proper motion estimations of the lens object. We perform joint binary microlensing fits of the MPS data together with the publicly available data from the EROS, MACHO/GMAN and OGLE collaborations. We also study the binary lens fit parameters previously published by the PLANET and MACHO/GMAN collaborations by using them as initial values for $\chi^2$ minimization. Fits based on the PLANET model I appear to be in conflict with the GMAN-CTIO data. From our best fit, we find that the lens system has a proper motion of $\mu = 1.3\pm 0.2 \kmsk$ with respect to the source, which implies that the lens system is most likely to be located in the Small Magellanic Cloud strengthening the conclusion of previous reports., Comment: 20 pages, 4 color figures

We present photometric observations and analysis of the second microlensing event detected towards the Small Magellanic Cloud (SMC), MACHO Alert 98-SMC-1. This event was detected early enough to allow intensive observation of the lightcurve. These observations revealed 98-SMC-1 to be the first caustic crossing, binary microlensing event towards the Magellanic Clouds to be discovered in progress. Frequent coverage of the evolving lightcurve allowed an accurate prediction for the date of the source crossing out of the lens caustic structure. The caustic crossing temporal width, along with the angular size of the source star, measures the proper motion of the lens with respect to the source, and thus allows an estimate of the location of the lens. Lenses located in the Galactic halo would have a velocity projected to the SMC of v^hat ~1500 km/s, while an SMC lens would typically have v^hat ~60 km/s. We have performed a joint fit to the MACHO/GMAN data presented here, including recent EROS data of this event. These joint data are sufficient to constrain the time for the lens to move an angle equal to the source angular radius; 0.116 +/- 0.010 days. We estimate a radius for the lensed source of 1.4 +/- 0.1 R_sun. This yields a projected velocity of v^hat = 84 +/- 9 km/s. Only 0.15% of halo lenses would be expected to have a v^hat value at least as small as this, while 31% of SMC lenses would be expected to have v^hat as large as this. This implies that the lensing system is more likely to reside in the SMC than in the Galactic halo., Comment: 16 pages, including 3 tables and 3 figures; submitted to The Astrophysical Journal

We present the lightcurves of two microlensing events from the MACHO Project data that are likely to be due to lenses with masses similar to Jupiter's mass. Although the MACHO Project survey data are not sufficient to definitively establish the identification of planetary mass lenses in these cases, observations by microlensing follow-up networks such as GMAN and PLANET should be able to definitively determine the planetary nature of similar events which may occur in the near future., Comment: 5 pages, Latex, 5 postscript figures included, To appear in the Proceedings of "Planets Beyond the Solar System and the Next Generation of Space Missions", held at STScI, October 16-18, 1996

Gravitational microlensing is known for baryoninc dark matter searches. Here we show that microlensing also provides a unique tool for the detection of low mass planets (such as earths and neptunes) from the ground. A planetary system forms a binary lens (or, a multi-point lens), and we can determine the mass ratio of the planet with respect to the star and relative distance ($=$ separation/Einstein ring radius) between the star and planet. Such a microlenisng planet search project requires a $\approx 2$ m survey telescope, and a network of $1.5-2$ m follow-up telescopes capable of monitoring stars in the Bulge on a 24-hour basis. During the off-season of the Galactic bulge, this network can be used for dark matter search by monitoring the stars in the LMC and SMC., Comment: 5 pages, submitted to the Proceedings of the Second International Symposium on Sources and Detection of Dark Matter in the Universe

The MACHO collaboration has recently analyzed 2.1 years of photometric data for about 8.5 million stars in the Large Magellanic Cloud (LMC). This analysis has revealed 8 candidate microlensing events and a total microlensing optical depth of $\tau_{meas} = 2.9 +1.4/-0.9 \times 10^{-7}$. This significantly exceeds the number of events (1.1) and the microlensing optical depth predicted from known stellar populations: $\tau_{back} = 5.4\times 10^{-8}$, but it is consistent with models in which about half of the standard dark halo mass is composed of Machos of mass $\sim 0.5 \msun$. One of these 8 events appears to be a binary lensing event with a caustic crossing that is partially resolved which allows us to estimate the distance to the lenses. If the source star is not a short period binary star, then we show that the lens system is very likely to reside in the LMC. However, if we assume that the optical depth for LMC-LMC lensing is large enough to account for our entire lensing signal, then the binary event does not appear to be consistent with lensing of a single LMC source star by a binary residing in the LMC. Thus, while the binary lens may indeed reside in the LMC, there is no indication that most of the lenses reside in the LMC., Comment: 5 pages, 3 postscript figures included; To appear in the Proceedings of the Dark Matter '96 Conference held in Santa Monica, CA, Feb., 1996