Selecting superluminous supernovae in faint galaxies from the first year of the Pan-STARRS1 Medium Deep Survey

The Pan-STARRS1 (PS1) survey has obtained imaging in five bands (grizyP1) over 10 Medium Deep Survey (MDS) fields covering a total of 70 square degrees. This paper describes the search for apparently hostless supernovae (SNe) within the first year of PS1 MDS data with an aim of discovering superluminous supernovae (SLSNe). A total of 249 hostless transients were discovered down to a limiting magnitude of MAB ∼ 23.5, of which 76 were classified as Type Ia supernovae (SNe Ia). There were 57 SNe with complete light curves that are likely core-collapse SNe (CCSNe) or type Ic SLSNe and 12 of these have had spectra taken. Of these 12 hostless, non-Type Ia SNe, 7 were SLSNe of type Ic at redshifts between 0.5 and 1.4. This illustrates that the discovery rate of type Ic SLSNe can be maximized by concentrating on hostless transients and removing normal SNe Ia. We present data for two possible SLSNe; PS1-10pm (z = 1.206) and PS1-10ahf (z = 1.1), and estimate the rate of type Ic SLSNe to be between 3+3−2×10−53+3−2×10−53+3−2×10−5 and 8+2−1×10−58+2−1×10−58+2−1×10−5 that of the CCSN rate within 0.3 ≤ z ≤ 1.4 by applying a Monte Carlo technique. The rate of slowly evolving, type Ic SLSNe (such as SN2007bi) is estimated as a factor of 10 lower than this range.

Selection of burst-like transients and stochastic variables using multi-brand image differencing in the Pan-STARRS1 Medium Deep Survey

We present a novel method for the light-curve characterization of Pan-STARRS1 Medium Deep Survey (PS1 MDS) extragalactic sources into stochastic variables (SVs) and burst-like (BL) transients, using multi-band image-differencing time-series data. We select detections in difference images associated with galaxy hosts using a star/galaxy catalog extracted from the deep PS1 MDS stacked images, and adopt a maximum a posteriori formulation to model their difference-flux time-series in four Pan-STARRS1 photometric bands gP1, rP1, iP1, and zP1. We use three deterministic light-curve models to fit BL transients; a Gaussian, a Gamma distribution, and an analytic supernova (SN) model, and one stochastic light-curve model, the Ornstein-Uhlenbeck process, in order to fit variability that is characteristic of active galactic nuclei (AGNs). We assess the quality of fit of the models band-wise and source-wise, using their estimated leave-out-one cross-validation likelihoods and corrected Akaike information criteria. We then apply a K-means clustering algorithm on these statistics, to determine the source classification in each band. The final source classification is derived as a combination of the individual filter classifications, resulting in two measures of classification quality, from the averages across the photometric filters of (1) the classifications determined from the closest K-means cluster centers, and (2) the square distances from the clustering centers in the K-means clustering spaces. For a verification set of AGNs and SNe, we show that SV and BL occupy distinct regions in the plane constituted by these measures. We use our clustering method to characterize 4361 extragalactic image difference detected sources, in the first 2.5 yr of the PS1 MDS, into 1529 BL, and 2262 SV, with a purity of 95.00% for AGNs, and 90.97% for SN based on our verification sets. We combine our light-curve classifications with their nuclear or off-nuclear host galaxy offsets, to define a robust photometric sample of 1233 AGNs and 812 SNe. With these two samples, we characterize their variability and host galaxy properties, and identify simple photometric priors that would enable their real-time identification in future wide-field synoptic surveys.

The VIMOS-VLT Deep Survey: the evolution of type-1 AGN

Quasars and AGN play an important role in many aspects of the modern cosmology. Of particular interest is the issue of the interplay between AGN activity and formation and evolution of galaxies and structures. Studies on nearby galaxies revealed that most (and possibly all) galaxy nuclei contain a super-massive black hole (SMBH) and that between a third and half of them are showing some evidence of activity (Kormendy and Richstone, 1995). The discovery of a tight relation between black holes mass and velocity dispersion of their host galaxy suggests that the evolution of the growth of SMBH and their host galaxy are linked together. In this context, studying the evolution of AGN, through the luminosity function (LF), is fundamental to constrain the theories of galaxy and SMBH formation and evolution. Recently, many theories have been developed to describe physical processes possibly responsible of a common formation scenario for galaxies and their central black hole (Volonteri et al., 2003; Springel et al., 2005a; Vittorini et al., 2005; Hopkins et al., 2006a) and an increasing number of observations in different bands are focused on collecting larger and larger quasar samples. Many issues remain however not yet fully understood. In the context of the VVDS (VIMOS-VLT Deep Survey), we collected and studied an unbiased sample of spectroscopically selected faint type-1 AGN with a unique and straightforward selection function. Indeed, the VVDS is a large, purely magnitude limited spectroscopic survey of faint objects, free of any morphological and/or color preselection. We studied the statistical properties of this sample and its evolution up to redshift z 4. Because of the contamination of the AGN light by their host galaxies at the faint magnitudes explored by our sample, we observed that a significant fraction of AGN in our sample would be missed by the UV excess and morphological criteria usually adopted for the pre-selection of optical QSO candidates. If not properly taken into account, this failure in selecting particular sub-classes of AGN could, in principle, affect some of the conclusions drawn from samples of AGN based on these selection criteria. The absence of any pre-selection in the VVDS leads us to have a very complete sample of AGN, including also objects with unusual colors and continuum shape. The VVDS AGN sample shows in fact redder colors than those expected by comparing it, for example, with the color track derived from the SDSS composite spectrum. In particular, the faintest objects have on average redder colors than the brightest ones. This can be attributed to both a large fraction of dust-reddened objects and a significant contamination from the host galaxy. We have tested these possibilities by examining the global spectral energy distribution of each object using, in addition to the U, B, V, R and I-band magnitudes, also the UV-Galex and the IR-Spitzer bands, and fitting it with a combination of AGN and galaxy emission, allowing also for the possibility of extinction of the AGN flux. We found that for 44% of our objects the contamination from the host galaxy is not negligible and this fraction decreases to 21% if we restrict the analysis to a bright subsample (M1450 <-22.15). Our estimated integral surface density at IAB < 24.0 is 500 AGN per square degree, which represents the highest surface density of a spectroscopically confirmed sample of optically selected AGN. We derived the luminosity function in B-band for 1.0 < z < 3.6 using the 1/Vmax estimator. Our data, more than one magnitude fainter than previous optical surveys, allow us to constrain the faint part of the luminosity function up to high redshift. A comparison of our data with the 2dF sample at low redshift (1 < z < 2.1) shows that the VDDS data can not be well fitted with the pure luminosity evolution (PLE) models derived by previous optically selected samples. Qualitatively, this appears to be due to the fact that our data suggest the presence of an excess of faint objects at low redshift (1.0 < z < 1.5) with respect to these models. By combining our faint VVDS sample with the large sample of bright AGN extracted from the SDSS DR3 (Richards et al., 2006b) and testing a number of different evolutionary models, we find that the model which better represents the combined luminosity functions, over a wide range of redshift and luminosity, is a luminosity dependent density evolution (LDDE) model, similar to those derived from the major Xsurveys. Such a parameterization allows the redshift of the AGN density peak to change as a function of luminosity, thus fitting the excess of faint AGN that we find at 1.0 < z < 1.5. On the basis of this model we find, for the first time from the analysis of optically selected samples, that the peak of the AGN space density shifts significantly towards lower redshift going to lower luminosity objects. The position of this peak moves from z 2.0 for MB <-26.0 to z 0.65 for -22< MB <-20. This result, already found in a number of X-ray selected samples of AGN, is consistent with a scenario of “AGN cosmic downsizing”, in which the density of more luminous AGN, possibly associated to more massive black holes, peaks earlier in the history of the Universe (i.e. at higher redshift), than that of low luminosity ones, which reaches its maximum later (i.e. at lower redshift). This behavior has since long been claimed to be present in elliptical galaxies and it is not easy to reproduce it in the hierarchical cosmogonic scenario, where more massive Dark Matter Halos (DMH) form on average later by merging of less massive halos.

A multifrequency study of the Lockman Hole region: preparing for a LOFAR deep survey

In this thesis, we have presented two deep 1.4 GHz and 345 MHz overlapping surveys of the Lockman Hole field taken with the Westerbork Synthesis Radio Telescope. We extracted a catalogue of ~6000 radio sources from the 1.4 GHz mosaic down to a flux limit of ~55 μJy and a catalogue of 334 radio sources down to a flux limit of ~4 mJy from the inner 7 sq. degree region of the 345 MHz image. The extracted catalogues were used to derive the source number counts at 1.4 GHz and at 345 MHz. The source counts were found to be fully consistent with previous determinations. In particular the 1.4 GHz source counts derived by the present sample provide one of the most statistically robust determinations in the flux range 0.1 < S < 1 mJy. During the commissioning program of the LOFAR telescope, the Lockman Hole field was observed at 58 MHz and 150 MHz. The 150 MHz LOFAR observation is particularly relevant as it allowed us to obtain the first LOFAR flux calibrated high resolution image of a deep field. From this image we extracted a preliminary source catalogue down to a flux limit of ~15 mJy (~10σ), that can be considered complete down to 20‒30 mJy. A spectral index study of the mJy sources in the Lockman Hole region, was performed using the available catalogues ( 1.4 GHz, 345 MHz and 150 MHz) and a deep 610 MHz source catalogue available from the literature (Garn et al. 2008, 2010).

The XMM deep survey in the CDF-S. VI. Obscured AGN selected as infrared power-law galaxies

Context. Accretion onto supermassive black holes is believed to occur mostly in obscured active galactic nuclei (AGN). Such objects are proving rather elusive in surveys of distant galaxies, including those at X-ray energies. Aims. Our main goal is to determine whether the revised IRAC criteria of Donley et al. (2012, ApJ, 748, 142; objects with an infrared (IR) power-law spectral shape), are effective at selecting X-ray type-2 AGN (i.e., absorbed N_H > 10^22 cm^-2). Methods. We present the results from the X-ray spectral analysis of 147 AGN selected by cross-correlating the highest spectral quality ultra-deep XMM-Newton and the Spitzer/IRAC catalogues in the Chandra Deep Field South. Consequently it is biased towards sources with high S/N X-ray spectra. In order to measure the amount of intrinsic absorption in these sources, we adopt a simple X-ray spectral model that includes a power-law modified by intrinsic absorption at the redshift of each source and a possible soft X-ray component. Results. We find 21/147 sources to be heavily absorbed but the uncertainties in their obscuring column densities do not allow us to confirm their Compton-Thick nature without resorting to additional criteria. Although IR power-law galaxies are less numerous in our sample than IR non-power-law galaxies (60 versus 87 respectively), we find that the fraction of absorbed (N_H^intr > 10^22 cm^-2) AGN is significantly higher (at about 3 sigma level) for IR-power-law sources (similar to 2/3) than for those sources that do not meet this IR selection criteria (~1/2). This behaviour is particularly notable at low luminosities, but it appears to be present, although with a marginal significance, at all luminosities. Conclusions. We therefore conclude that the IR power-law method is efficient in finding X-ray-absorbed sources. We would then expect that the long-sought dominant population of absorbed AGN is abundant among IR power-law spectral shape sources not detected in X-rays.

Cosmological Constraints from Measurements of Type Ia Supernovae Discovered During the First 1.5 Yr of the Pan-STARRS1 Survey

We present griz_P1 light curves of 146 spectroscopically confirmed Type Ia supernovae (SNe Ia; 0.03 < z < 0.65) discovered during the first 1.5 yr of the Pan-STARRS1 Medium Deep Survey. The Pan-STARRS1 natural photometric system is determined by a combination of on-site measurements of the instrument response function and observations of spectrophotometric standard stars. We find that the systematic uncertainties in the photometric system are currently 1.2% without accounting for the uncertainty in the Hubble Space Telescope Calspec definition of the AB system. A Hubble diagram is constructed with a subset of 113 out of 146 SNe Ia that pass our light curve quality cuts. The cosmological fit to 310 SNe Ia (113 PS1 SNe Ia + 222 light curves from 197 low-z SNe Ia), using only supernovae (SNe) and assuming a constant dark energy equation of state and flatness, yields w = -1.120^+0.360_-0.206(Stat)^+0.269_0.291(Sys). When combined with BAO+CMB(Planck)+H₀, the analysis yields Ω_M = 0.280^+0.013_0.012 and w = -1.166^+0.072_-0.069 including all identified systematics. The value of w is inconsistent with the cosmological constant value of -1 at the 2.3σ level. Tension endures after removing either the baryon acoustic oscillation (BAO) or the H₀ constraint, though it is strongest when including the H₀ constraint. If we include WMAP9 cosmic microwave background (CMB) constraints instead of those from Planck, we find w = -1.124^+0.083_-0.065, which diminishes the discord to <2σ. We cannot conclude whether the tension with flat ΛCDM is a feature of dark energy, new physics, or a combination of chance and systematic errors. The full Pan-STARRS1 SN sample with ∼three times as many SNe should provide more conclusive results.

THE SLOAN DIGITAL SKY SURVEY CO-ADD: A GALAXY PHOTOMETRIC REDSHIFT CATALOG

We present and describe a catalog of galaxy photometric redshifts (photo-z) for the Sloan Digital Sky Survey (SDSS) Co-add Data. We use the artificial neural network (ANN) technique to calculate the photo-z and the nearest neighbor error method to estimate photo-z errors for similar to 13 million objects classified as galaxies in the co-add with r < 24.5. The photo-z and photo-z error estimators are trained and validated on a sample of similar to 83,000 galaxies that have SDSS photometry and spectroscopic redshifts measured by the SDSS Data Release 7 (DR7), the Canadian Network for Observational Cosmology Field Galaxy Survey, the Deep Extragalactic Evolutionary Probe Data Release 3, the VIsible imaging Multi-Object Spectrograph-Very Large Telescope Deep Survey, and the WiggleZ Dark Energy Survey. For the best ANN methods we have tried, we find that 68% of the galaxies in the validation set have a photo-z error smaller than sigma(68) = 0.031. After presenting our results and quality tests, we provide a short guide for users accessing the public data.

The 2df SDSS LRG and QSO survey: evolution of the luminosity function of luminous red galaxies to z=0.6

We present new measurements of the luminosity function (LF) of luminous red galaxies (LRGs) from the Sloan Digital Sky Survey (SDSS) and the 2dF SDSS LRG and Quasar (2SLAQ) survey. We have carefully quantified, and corrected for, uncertainties in the K and evolutionary corrections, differences in the colour selection methods, and the effects of photometric errors, thus ensuring we are studying the same galaxy population in both surveys. Using a limited subset of 6326 SDSS LRGs (with 0.17 < z < 0.24) and 1725 2SLAQ LRGs (with 0.5 < z < 0.6), for which the matching colour selection is most reliable, we find no evidence for any additional evolution in the LRG LF, over this redshift range, beyond that expected from a simple passive evolution model. This lack of additional evolution is quantified using the comoving luminosity density of SDSS and 2SLAQ LRGs, brighter than M-0.2r - 5 log h(0.7) = - 22.5, which are 2.51 +/- 0.03 x 10(-7) L circle dot Mpc(-3) and 2.44 +/- 0.15 x 10(-7) L circle dot Mpc(-3), respectively (< 10 per cent uncertainty). We compare our LFs to the COMBO-17 data and find excellent agreement over the same redshift range. Together, these surveys show no evidence for additional evolution (beyond passive) in the LF of LRGs brighter than M-0.2r - 5 log h(0.7) = - 21 ( or brighter than similar to L-*).. We test our SDSS and 2SLAQ LFs against a simple 'dry merger' model for the evolution of massive red galaxies and find that at least half of the LRGs at z similar or equal to 0.2 must already have been well assembled (with more than half their stellar mass) by z similar or equal to 0.6. This limit is barely consistent with recent results from semi-analytical models of galaxy evolution.

The 2dF-SDSS LRG and QSO survey: the star formation histories of luminous red galaxies

We present a detailed investigation into the recent star formation histories of 5697 luminous red galaxies (LRGs) based on the H delta (4101 angstrom), and [O II] (3727 angstrom) lines and the D4000 index. LRGs are luminous (L > 3L*) galaxies which have been selected to have photometric properties consistent with an old, passively evolving stellar population. For this study, we utilize LRGs from the recently completed 2dF-SDSS LRG and QSO Survey (2SLAQ). Equivalent widths of the H delta and [O II] lines are measured and used to define three spectral types, those with only strong H delta absorption (k+a), those with strong [O II] in emission (em) and those with both (em+a). All other LRGs are considered to have passive star formation histories. The vast majority of LRGs are found to be passive (similar to 80 per cent); however, significant numbers of k+a (2.7 per cent), em+a (1.2 per cent) and em LRGs (8.6 per cent) are identified. An investigation into the redshift dependence of the fractions is also performed. A sample of SDSS MAIN galaxies with colours and luminosities consistent with the 2SLAQ LRGs is selected to provide a low-redshift comparison. While the em and em+a fractions are consistent with the low-redshift SDSS sample, the fraction of k+a LRGs is found to increase significantly with redshift. This result is interpreted as an indication of an increasing amount of recent star formation activity in LRGs with redshift. By considering the expected lifetime of the k+a phase, the number of LRGs which will undergo a k+a phase can be estimated. A crude comparison of this estimate with the predictions from semi-analytic models of galaxy formation shows that the predicted level of k+a and em+a activities is not sufficient to reconcile the predicted mass growth for massive early types in a hierarchical merging scenario.

Star formation in a stellar mass-selected sample of galaxies to z=3 from the GOODS-NICMOS Survey

We present a study of the star-forming properties of a stellar mass-selected sample of galaxies in the GOODS (Great Observatories Origins Deep Survey) NICMOS Survey (GNS), based on deep Hubble Space Telescope (HST) imaging of the GOODS North and South fields. Using a stellar mass-selected sample, combined with HST/ACS and Spitzer data to measure both ultraviolet (UV) and infrared-derived star formation rates (SFRs), we investigate the star forming properties of a complete sample of ∼1300 galaxies down to log M_*= 9.5 at redshifts 1.5 < z < 3. Eight per cent of the sample is made up of massive galaxies with M_*≥ 10^11 M_⊙. We derive optical colours, dust extinctions and UV and infrared SFR to determine how the SFR changes as a function of both stellar mass and time. Our results show that SFR increases at higher stellar mass such that massive galaxies nearly double their stellar mass from star formation alone over the redshift range studied, but the average value of SFR for a given stellar mass remains constant over this ∼2 Gyr period. Furthermore, we find no strong evolution in the SFR for our sample as a function of mass over our redshift range of interest; in particular we do not find a decline in the SFR among massive galaxies, as is seen at z < 1. The most massive galaxies in our sample (log M_*≥ 11) have high average SFRs with values SFR_UV, corr= 103 ± 75 M_⊙ yr^−1, and yet exhibit red rest-frame (U−B) colours at all redshifts. We conclude that the majority of these red high-redshift massive galaxies are red due to dust extinction. We find that A_2800 increases with stellar mass, and show that between 45 and 85 per cent of massive galaxies harbour dusty star formation. These results show that even just a few Gyr after the first galaxies appear, there are strong relations between the global physical properties of galaxies, driven by stellar mass or another underlying feature of galaxies strongly related to the stellar mass.

Observational and theoretical advances in cosmological foreground emission

Observational and theoretical work towards the separation of foreground emission from the cosmic microwave background is described. The bulk of this work is in the design, construction, and commissioning of the C-Band All-Sky Survey (C-BASS), an experiment to produce a template of the Milky Way Galaxy's polarized synchrotron emission. Theoretical work is the derivation of an analytical approximation to the emission spectrum of spinning dust grains.

The performance of the C-BASS experiment is demonstrated through a preliminary, deep survey of the North Celestial Pole region. A comparison to multiwavelength data is performed, and the thermal and systematic noise properties of the experiment are explored. The systematic noise has been minimized through careful data processing algorithms, implemented both in the experiment's Field Programmable Gate Array (FPGA) based digital backend and in the data analysis pipeline. Detailed descriptions of these algorithms are presented.

The analytical function of spinning dust emission is derived through the application of careful approximations, with each step tested against numerical calculations. This work is intended for use in the parameterized separation of cosmological foreground components and as a framework for interpreting and comparing the variety of anomalous microwave emission observations.

GALEX AND PAN-STARRS1 DISCOVERY OF SN IIP 2010aq: THE FIRST FEW DAYS AFTER SHOCK BREAKOUT IN A RED SUPERGIANT STAR

We present the early UV and optical light curve of Type IIP supernova (SN) 2010aq at z = 0.0862, and compare it to analytical models for thermal emission following SN shock breakout in a red supergiant star. SN 2010aq was discovered in joint monitoring between the Galaxy Evolution Explorer (GALEX) Time Domain Survey (TDS) in the NUV and the Pan-STARRS1 Medium Deep Survey (PS1 MDS) in the g, r, i, and z bands. The GALEX and Pan-STARRS1 observations detect the SN less than 1 day after the shock breakout, measure a diluted blackbody temperature of 31,000 +/- 6000 K 1 day later, and follow the rise in the UV/optical light curve over the next 2 days caused by the expansion and cooling of the SN ejecta. The high signal-to-noise ratio of the simultaneous UV and optical photometry allows us to fit for a progenitor star radius of 700 +/- 200R(circle dot), the size of a red supergiant star. An excess in UV emission two weeks after shock breakout compared with SNe well fitted by model atmosphere-code synthetic spectra with solar metallicity is best explained by suppressed line blanketing due to a lower metallicity progenitor star in SN 2010aq. Continued monitoring of PS1 MDS fields by the GALEX TDS will increase the sample of early UV detections of Type II SNe by an order of magnitude and probe the diversity of SN progenitor star properties.

Pan-STARRS1 DISCOVERY OF TWO ULTRALUMINOUS SUPERNOVAE AT(z) approximate to 0.9

We present the discovery of two ultraluminous supernovae (SNe) at z approximate to 0.9 with the Pan-STARRS1 Medium Deep Survey. These SNe, PS1-10ky and PS1-10awh, are among the most luminous SNe ever discovered, comparable to the unusual transients SN 2005ap and SCP 06F6. Like SN 2005ap and SCP 06F6, they show characteristic high luminosities (M-bol approximate to -22.5 mag), blue spectra with a few broad absorption lines, and no evidence for H or He. We have constructed a full multi-color light curve sensitive to the peak of the spectral energy distribution in the rest-frame ultraviolet, and we have obtained time series spectroscopy for these SNe. Given the similarities between the SNe, we combine their light curves to estimate a total radiated energy over the course of explosion of (0.9-1.4) x 10(51) erg. We find photospheric velocities of 12,000-19,000 km s(-1) with no evidence for deceleration measured across similar to 3 rest-frame weeks around light curve peak, consistent with the expansion of an optically thick massive shell of material. We show that, consistent with findings for other ultraluminous SNe in this class, radioactive decay is not sufficient to power PS1-10ky, and we discuss two plausible origins for these events: the initial spin-down of a newborn magnetar in a core-collapse SN, or SN shock breakout from the dense circumstellar wind surrounding a Wolf-Rayet star.

The superluminous supernova PS1-11ap: bridging the gap between low and high redshift

We present optical photometric and spectroscopic coverage of the superluminous supernova (SLSN) PS1-11ap, discovered with the Pan-STARRS1 Medium Deep Survey at z = 0.524. This intrinsically blue transient rose slowly to reach a peak magnitude of Mu = −21.4 mag and bolometric luminosity of 8 × 1043 erg s−1 before settling on to a relatively shallow gradient of decline. The observed decline is significantly slower than those of the SLSNe-Ic which have been the focus of much recent attention. Spectroscopic similarities with the lower redshift SN2007bi and a decline rate similar to 56Co decay time-scale initially indicated that this transient could be a candidate for a pair instability supernova (PISN) explosion. Overall the transient appears quite similar to SN2007bi and the lower redshift object PTF12dam. The extensive data set, from 30 d before peak to 230 d after, allows a detailed and quantitative comparison with published models of PISN explosions. We find that the PS1-11ap data do not match these model explosion parameters well, supporting the recent claim that these SNe are not pair instability explosions. We show that PS1-11ap has many features in common with the faster declining SLSNe-Ic, and the light-curve evolution can also be quantitatively explained by the magnetar spin-down model. At a redshift of z = 0.524, the observer-frame optical coverage provides comprehensive rest-frame UV data and allows us to compare it with the SLSNe recently found at high redshifts between z = 2 and 4. While these high-z explosions are still plausible PISN candidates, they match the photometric evolution of PS1-11ap and hence could be counterparts to this lower redshift transient.

Machine learning for transient discovery in Pan-STARRS1 difference imaging

Efficient identification and follow-up of astronomical transients is hindered by the need for humans to manually select promising candidates from data streams that contain many false positives. These artefacts arise in the difference images that are produced by most major ground-based time-domain surveys with large format CCD cameras. This dependence on humans to reject bogus detections is unsustainable for next generation all-sky surveys and significant effort is now being invested to solve the problem computationally. In this paper, we explore a simple machine learning approach to real-bogus classification by constructing a training set from the image data of similar to 32 000 real astrophysical transients and bogus detections from the Pan-STARRS1 Medium Deep Survey. We derive our feature representation from the pixel intensity values of a 20 x 20 pixel stamp around the centre of the candidates. This differs from previous work in that it works directly on the pixels rather than catalogued domain knowledge for feature design or selection. Three machine learning algorithms are trained (artificial neural networks, support vector machines and random forests) and their performances are tested on a held-out subset of 25 per cent of the training data. We find the best results from the random forest classifier and demonstrate that by accepting a false positive rate of 1 per cent, the classifier initially suggests a missed detection rate of around 10 per cent. However, we also find that a combination of bright star variability, nuclear transients and uncertainty in human labelling means that our best estimate of the missed detection rate is approximately 6 per cent.