Searcharxiv⌕ Search

arXiv subjects

Andrew Gould

Publications and source records attributed to Andrew Gould.

At least 397 records · Page 22Linked to original sources

Statistical Determination of the MACHO Mass Spectrum

The mass function of 51 Massive Compact Objects (MACHOs) detected toward the Galactic bulge is statistically estimated from Einstein ring crossing times $t_{\rm e}$. For a Gaussian mass function, the best fitting parameters are $\langle \log (m/M_{\odot}) \rangle = -1.12$ and $σ_{\log (m/M_{\odot})} = 0.57$. If the mass function follows a power-law distribution, the best fitting mass cut-off and power are $m_{\rm cut} = 0.04\ M_{\odot}$ and $p = -2.1$. Both mass spectra indicate that a significant fraction of events are caused by MACHOs in substellar mass range. Both best determined mass functions are compared with that obtained from Hubble Space Telescope (HST) observations. The power law is marginally favored ($2.3σ$) over the Gaussian mass function, and significantly over HST mass function ($5.5σ$), indicating that MACHOs have a different mass function from stars in the solar neighborhood. In addition, the fact that all the models have very poor fits to the longest four events with $t_{e} \geq 70\ {\rm days}$ remains a puzzle.

astro-ph↗

Imaging the Lyman-alpha Forest

We show that it is now possible to image optically thick $\lya$ clouds in fluorescent $\lya$ emission with a relatively long ($\sim 20 $hr) integration on a large ($\sim 10 $m) telescope. For a broad range of column densities ($N\gsim 10^{18.5} \cm^{-2}$), the flux of $\lya$ photons from recombination cascades is equal to $\sim 0.6$ times the flux of ionizing photons, independent of the geometry of the cloud. Additional $\lya$ photons are produced by collisional excitations when these are the cloud's primary cooling mechanism. For typical physical conditions expected in optically thick clouds, these mechanisms together lead to a $\lya$ emission flux that is $\sim (2/3) \VEVν/ν_0$ times the flux of ionizing photons, where $\VEVν$ is the mean frequency of ionizing background photons and $ν_0$ is the Lyman limit frequency. Hence, measurement of the surface brightness from an optically thick cloud (known to exist, e.g., from a quasar absorption line) gives a direct measure of the energy in the ionizing radiation background. Moreover, in the same long slit spectrum one could hope to detect emission from $\sim 200$ other $\lya$ systems. Such detections would allow one to make a 2-dimensional map of the $\lya$ forest. By taking a series of such spectra, one could map the forest in three dimensions, revealing structure in the high-redshift universe.

astro-ph↗

Limits on the Gravity Wave Background From Microlensed Quasars

The paper previously submitted under this title is incorrect in that it drastically overestimates the cumulative deflection due to a gravitational wave (GW) background. Avi Loeb gives a simple argument that there can be no $(Dω)^{1/2}$ enhancement: since the problem is linear in $h$, one can decompose the GWs into plane waves and for each of these there is no enhancement.

astro-ph↗

Monte Carlo Simulations of Macho Parallaxes From a Satellite

Three ongoing microlensing experiments have found more candidate events than expected from the known stars. These experiments measure only one parameter of the massive compact halo objects (machos), the magnification time scale of the events. More information is required to understand the nature of the machos. A satellite experiment has been proposed to measure their projected transverse speed $\tilde{v} = v/(1-z)$, where $v$ is the macho transverse speed and $z$ its distance divided by the distance of the source. Measurement of $\tilde{v}$ would determine whether the machos were in the Galactic disk, Galactic halo, or in the Large Magellanic Cloud (LMC). We simulate events observed toward the LMC by the Earth and by a satellite in an Earth like heliocentric orbit. To leading order, such an experiment determines $\tilde{v}$ up to a two fold degeneracy. More precise measurements break the degeneracy. We show that with photometric precisions of 3\% to 4\% and approximately 1 observation per day, $\tilde{v}$ can be measured with a maximum error of 20\% for 70\% to 90\% of events similar to the ones reported by the EROS and MACHO collaborations. The projected transverse velocity is known with the same maximum error for 60\% to 75\% of these events. This 20\% maximum error is not a 1 $σ$ error but is mostly due to degeneracy between two possible solutions, each one being localized to much better than 20\%. These results are obtained with an Earth-satellite separation of 1 AU, and are improved by a larger separation.

astro-ph↗

Macho Proper Motions From Optical/Infrared Photometry

Optical/infrared photometry can double the number of proper motion measurements of Massive Compact Objects (MACHOs) relative to single band photometry. The proper motion of a MACHO can be measured by finding the ratio $q$ of the (known) radius of the source star to the Einstein radius of the MACHO, $q=θ_s/θ_e$. A classic method for doing this is to look for the effect on the light curve of the finite size of the source. A modification of this method proposed by Witt (1995) is to look for color changes in the light curve due to the fact that the limb darkening of the source is different in different bands. We demonstrate that the ``classical'' method is not feasible unless the MACHO actually transits the source: if the MACHO passes at say 1.5 source radii, there is still a sizable $\sim 5\%$ effect, but the light curve cannot be distinguished from point-source light curves with different parameters. However, color measurements in $V$ $(0.55\,μ$m) and $H$ $(1.65\,μ$m) reduce the errors by a factor $\sim 120$ and permit proper motion measurements at impact parameters of up to 2 source radii. Color maps in $V-H$ are also useful in the detection of planetary systems. Giant stars have a ``red ring'' in such maps. A planet which transits this ring gives rise to a distinctive signature which can help in the measurement of the planetary system's proper motion.

astro-ph↗

Lunar Occultation of MACHOs

Lunar occultation can be used to measure the proper motions of some of the long time scale microlensing events, $t_{e} \gsim 70$ days, now being detected toward the Galactic bulge. The long events are difficult to explain within the context of standard models of the mass distribution and dynamics of the Galaxy. Han & Gould (1995b) have suggested that they may be due to a kinematically cold population near the Sun. To resolve the mass, distance, and velocity of individual events and so to determine their nature, one must measure parallaxes and proper motions. For long events, parallaxes can be often obtained from ground-based measurements, but proper motions can only rarely be determined using conventional methods. Lunar occultations are therefore key to the understanding of the long events. We carry out realistic simulations to estimate the uncertainty of these measurements and show that proper motions could be measured for about one long event per year.

astro-ph↗

Disk M Dwarf Luminosity Function From HST Star Counts

We study a sample of 257 Galactic disk M dwarfs (8 12 (M<0.25 \ms), decreasing by a factor \gsim 3 by M_V~14 (M~0.14\ms). This decrease in the LF is in good agreement with the ground-based photometric study of nearby stars by Stobie et al. (1989), and in mild conflict with the most recent LF measurements based on local parallax stars by Reid et al. (1995). The local LF of the faint Galactic disk stars can be transformed into a local mass function using an empirical mass-M_V relation. The mass function can be represented analytically over the mass range 0.1\ms<M<1.6\ms by \log(ϕ)=-1.35-1.34\log(M/\ms)-1.85 [\log(M/\ms)]^2 where ϕis the number density per logarithmic unit of mass. The total column density of M stars is only Σ_M=11.8\pm 1.8\ms\pc^{-2}, implying a total `observed' disk column density of Σ_\obs~=39\ms\pc^{-2}, lower than previously believed, and also lower than all estimates with which we are familiar of the dynamically inferred mass of the disk. The measured scale length for the M-star disk is 3.0\pm 0.4 kpc. The optical depth to microlensing toward the LMC by the observed stars in the Milky Way disk is τ~1x10^{-8}, compared to the observed optical depth found in ongoing experiments τ_\obs~ 10^{-7}. The M-stars show evidence for a population with characteristics intermediate between thin disk and spheroid populations. Approximating what may be a continuum of populations by two separate component, we find characteristic exponential scale heights of ~210 pc and ~740 pc.

astro-ph↗

Probing For Machos of Mass $10^{-15}M_\odot$-$10^{-7}M_\odot$ with Gamma-Ray Burst Parallax Spacecraft

Two spacecraft separated by $\sim 1\,\au$ and equipped with gamma-ray burst (GRB) detectors could detect or rule out a cosmological density of Massive Compact Halo Objects (MACHOs) in the mass range $10^{-15} M_{\odot}\lsim M \lsim 10^{-7} M_{\odot}$ provided that GRBs prove to be cosmological. Previously devised methods for detecting MACHOs have spanned the mass range $10^{-16} M_{\odot}\lsim M \lsim 10^{7} M_{\odot}$, but with a gap of several orders of magnitude near $10^{-9} M_{\odot}$. For MACHOs and sources both at a cosmological distance, the Einstein radius is $\sim 1\,\au\,(M/10^{-7} M_\odot)^{1/2}$. Hence, if a GRB lies within the Einstein ring of a MACHO of mass $M\lsim 10^{-7}M_\odot$ as seen by one detector, it will not lie in the Einstein ring as seen by a second detector $\sim 1\,\au$ away. This implies that if GRBs are measured to have significantly different fluxes by the two detectors, this would signal the presence of a MACHO $\lsim 10^{-7}M_\odot$. By the same token, if the two detectors measured similar fluxes for several hundred events a cosmological abundance of such low-mass MACHOs would be ruled out. The lower limit of sensitivity, $M\lsim 10^{-15}M_\odot$ is set by the finite size of the source. If low-mass MACHOs are detected, there are tests which can discriminate among events generated by MACHOs in the three mass ranges $M\lsim 10^{-12}\,M_\odot$, $10^{-12}\,M_\odot\lsim M\lsim 10^{-7}\,M_\odot$, and $M\gsim 10^{-7}\ M_\odot$. Further experiments would then be required to make more accurate mass measurements.

astro-ph↗

Microlensing Search of $10^6$ Quasars

By monitoring $10^6$ quasars one could search for lensing by stars and Massive Compact Halo Objects (Machos) out to redshifts $z\sim 4$. If Machos have a present cosmological density $Ω_{L,0}=1\%$, then the expected event rate is $Γ\sim 200\,\yr^{-1}$. The expected event rate for known stars in galaxies is $Γ\sim 20\,\yr^{-1}$ assuming that their present cosmological density is $Ω_{L,0}=0.3\%$. Typical event times are $t_e\sim 3\,\yr$ for Machos and $t_e\sim 10\,\yr$ for stars. By comparing the optical depths to quasars at different redshifts, one could measure the star-formation and Macho-formation history of the universe. By comparing the time scales of events found parallel and perpendicular to the Sun's motion relative to the microwave background, one could measure or constrain the characteristic scale of large scale motions. The lensing events themselves would help probe the structure of quasars on scales of 50--1500 AU. The monitoring program could be carried out with a single dedicated 1 m telescope with a 4 deg${}^2$ camera. Quasar lensing events can be unambiguously distinguished from quasar variability because in the former case the broad lines are unaffected while in the latter they respond to the variation in the continuum on times scales $\sim 1\,\yr$.

astro-ph↗

Search for Intra-Cluster Machos by Pixel Lensing of M87

Intra-Cluster Machos (ICMs) are a plausible candidate for at least some of the dark matter in clusters of galaxies. ICMs can be detected by searching toward M87 for ``pixel lensing'', gravitational microlensing of unresolved stars. Dedicated observations by the Wide Field and Planetary Camera on the {\it Hubble Space Telescope} would discover lenses at a rate $Γ\sim 7 f\,\rm day^{-1}$ where $f$ is the fraction of the Virgo cluster composed of Machos (Massive Compact Objects). The only important background is lensing by stars in M87 itself. Such self-lensing should be detected at a rate $Γ\sim 1\,\rm day^{-1}$. However, most of these background events should be recognizable from their angular distribution and/or measurement of their proper motions (which are smaller by a factor $\sim 2$ than those of ICMs). The observations should be carried out principally in $I$, but with $V$ measurements made $\sim 20\%$ of the time in order to determine the colors of the lensed stars. It is also possible to detect ``Class II'' lensing events which are longer and hence for which lower signal to noise is balanced by a much larger number of observations per event. There are $\sim 5$ times more of these than the higher quality events. However, it is difficult to distinguish Class II events generated by ICMs from those generated by M87 stars.

astro-ph↗

K Band Microlensing of the Inner Galaxy

Microlensing searches toward the inner galaxy $(|l|,|b|\leq 22.\hskip-2pt'5)$ have several major advantages. First, the event rate is strongly dominated by bulge-bulge lensing events where both the source and lens lie in the bulge. Second, these bulge-bulge events have very short time scales $t_e\sim 2\,$days and are therefore easily distinguished from the less frequent and much longer bulge-disk and disk-disk events. Third, since the optical depth is similar to that at higher impact parameters, while the events are shorter, the event rate is high $Γ\sim 3\times 10^{-7}\rm day^{-1}$. Fourth, because the Einstein rings are small, $r_e\sim 5\times 10^{12}\,$cm, and the source stars are large $r_s\gsim 10^{12}\,$cm, the lens will transit the face of the source for a significant fraction $(\sim 20\%)$ of events. For these transit events it will often be possible to measure a second lens parameter, the angular Einstein radius (or proper motion). In addition to the bulge-bulge events, the optical depth of the disk is $\sim 7$ times larger toward the inner Galaxy than toward Baade's Window. A microlensing search toward the inner Galaxy can be carried out by making frequent $(\sim4\,\rm day^{-1})$ K band images of a large area $\sim 0.5\,\rm deg^2$ to a depth of $K\sim 16$, and hence requires either a $1024^2$ infrared array on a dedicated 2m telescope or four such arrays on a 1m telescope.

astro-ph↗

Statistics Of Microlensing Optical Depth

We show that the statistical error, $σ_τ$, in estimating the optical depth, $τ$, due to microlensing is substantially higher than the naive Poisson estimate: $σ_τ = \sqrt{η/ N} τ$, where $N$ is the number of detections and $1.65 \leq η\lsim 3$. We construct a ``microlensing map'' using an observationally well-constrained barred bulge model and compare this with a map based on a spherical bulge model. The result indicates a significant differences in the distribution of optical depths. If the optical depths are determined over a wide range of positions around the Galactic center, one could make a rudimentary determination of the geometry of the mass distribution of the Galactic bulge by observing $\sim 200$ events. About 800 events are required to determine the relative contributions to the event rate by bulge and disk Machos.

astro-ph↗

A New Macho Search Strategy

I propose a radical revision in the search strategy for Massive Compact Objects (Machos) toward the Galactic bulge: monitor the entire $\sim 300$ square deg of the bulge and tune the search primarily to microlensing events of bright $M_V\lsim 2$ stars. By employing a small second telescope to follow up the events detected with low dimensionless impact parameter $β\lsim 0.2$, one could measure the proper motion $Ω$ of $\sim 10\,\rm yr^{-1}$ events and significantly constrain $Ω$ for a similar number. If, in addition, the events were followed from a Macho parallax satellite, it would be possible to measure individual masses, distances, and transverse velocities for the events with proper motions. If a fraction $ξ$ of the bulge is composed of low-mass objects in the range $10^{-3}\lsim M_{\rm low}/M_\odot\lsim 10^{-1}$, then mass measurements could be made for $\sim 10(ξ/0.1)(0.01 M_\odot \VEV{M_{\rm low}^{-1}})\,\rm yr^{-1}$ of these, thus allowing a direct measurement of the mass function in the sub-stellar range. In the absence of a parallax satellite, ground-based observations could significantly constrain (but not measure) Macho parallaxes. These constraints, when combined with the proper-motion measurements, would in turn constrain the mass, distance, and transverse speed of the Machos. The proposed strategy should therefore be adopted even before a parallax satellite is launched.

astro-ph↗

Self-Lensing By Binaries

Lensing of one member of a binary by its companion is studied for several classes of binaries. For binaries in which at least one member is an ordinary (non-compact) star, the optical depths to lensing is extremely low, $τ\lsim v_\esc^4/c^4\sim 10^{-11}$. Here $v_\esc$ is the escape velocity from the component with the weaker potential. Binaries composed of two compact objects can have higher optical depths. About 0.2\% of millisecond pulsar binaries will be self-lensing. Such objects can be used to probe the pulsar-emission mechanism.

astro-ph↗

The Mass Spectrum Of Machos From Parallax Measurements

We demonstrate that by making satellite-based parallax measurements of Macho events, it is possible to distinguish disk from bulge Machos and to determine individual masses of disk Machos to an accuracy of $0.2$ in the log. This is at least a 3-fold improvement over what can be done without parallaxes, i.e., just from the observed time scale of the events. In addition, we show that the physical distribution of disk Machos can be found from `reduced' velocities determined by parallax measurements.

astro-ph↗

Transverse Velocities of Galaxies From Microlens Parallaxes

The transverse velocity of a spiral galaxy can be measured to an accuracy $\sim 60\,\kms$ by making parallax observations of quasars being microlensed by stars in the disk of the galaxy. To make the measurement, a quasar must be located behind the disk of the galaxy between about 1 and 2 scale lengths from the center. The quasar must then be monitored for microlensing events and the events followed simultaneously from the Earth and a satellite in solar orbit, preferably at $\sim 30\,$AU. A systematic search in a volume within $21,000\,\kms$ could locate quasars with $B<23$ behind a total of $\sim 1900$ galaxies. The rate of lensing events (and hence galaxy velocity measurements) would be $\sim 3\,{\rm yr}^{-1}$. The events would have typical characteristic times $ω^{-1}\sim 3\,$yr. Under the assumption that the mass spectrum of lensing objects is the same in other spiral galaxies as in our own, the observations could be used to measure the Hubble parameter to an accuracy of 5\%.

astro-ph↗

Microlensing Events: End of the Dark Halo?

I obtain an upper limit for the optical depth to microlensing toward Baade's Window of $3\times 10^{-6}$ by assuming that all of the mass of the Galaxy interior to the Sun (and not in the bulge) is in a disk. The exponential scale height of the disk is left as a completely arbitrary function of radius and is varied to maximize the optical depth. I take account of the relatively small corrections induced by the fact that the bulge is not axisymmetric. If initial estimates by the OGLE collaboration of an observed optical depth $τ\sim 3.3\times 10^{-6}$ are confirmed, then essentially all of the dark matter interior to the Sun must be in a disk with a scale height of a few hundred parsecs.

astro-ph↗

Macho Parallaxes From A Single Satellite

Massive Compact Objects (Machos) are currently being discovered at substantially higher rates than would be expected from standard models of known stellar populations. To determine whether they are due to non-standard distri- butions of known populations or to a heretofore unknown (`dark') population, one must acquire more information about the individual events. Space-based parallaxes are potentially the best tool for extracting additional information. To leading order, parallax measurements by a single satellite result in a four- fold degeneracy: two possible values of $\tilde v$ and two possible signs for the component of motion perpendicular to the projected satellite-Earth vector. It had been believed that a second satellite would be required to break this degeneracy. I show that the velocity difference between the satellite and the Earth allows one to partially or totally break the degeneracy using a single satellite. For most Macho events it is possible to measure $\tilde v$ and $\tilde r_e$. For some it is also possible to measure $Φ$. The proposed Space Infrared Telescope could measure $\sim 100$ parallaxes per year by applying $\sim 400\,$hr of telescope time.

astro-ph↗