Search

The exoplanet archive is an incredible resource of information on the properties of discovered extrasolar planets, but statistical analysis has been limited by the number of missing values. One of the most informative bulk properties is planet mass, which is particularly challenging to measure with more than 70\% of discovered planets with no measured value. We compare the capabilities of five different machine learning algorithms that can utilize multidimensional incomplete datasets to estimate missing properties for imputing planet mass. The results are compared when using a partial subset of the archive with a complete set of six planet properties, and where all planet discoveries are leveraged in an incomplete set of six and eight planet properties. We find that imputation results improve with more data even when the additional data is incomplete, and allows a mass prediction for any planet regardless of which properties are known. Our favored algorithm is the newly developed $k$NN$\times$KDE, which can return a probability distribution for the imputed properties. The shape of this distribution can indicate the algorithm's level of confidence, and also inform on the underlying demographics of the exoplanet population. We demonstrate how the distributions can be interpreted with a series of examples for planets where the discovery was made with either the transit method, or radial velocity method. Finally, we test the generative capability of the $k$NN$\times$KDE to create a large synthetic population of planets based on the archive, and identify potential categories of planets from groups of properties in the multidimensional space. All codes are Open Source., Comment: 30 pages, 14 figures, 1 table. Accepted for publication in the Open Journal of Astrophysics

The search for exoplanetary life must encompass the complex geological processes reflected in an exoplanet's atmosphere, or we risk reporting false positive and false negative detections. To do this, we must nurture the nascent discipline of "exogeoscience" to fully integrate astronomers, astrophysicists, geoscientists, oceanographers, atmospheric chemists and biologists. Increased funding for interdisciplinary research programs, supporting existing and future multidisciplinary research nodes, and developing research incubators is key to transforming true exogeoscience from an aspiration to a reality., Comment: Submitted as white paper to 2023-2033 Planetary Science and Astrobiology Decadal Survey; Updated to include all co-authors

Earth-Like is an interactive website and twitter bot that allows users to explore changes in the average global surface temperature of an Earth-like planet due to variations in the surface oceans and emerged land coverage, rate of volcanism (degassing), and the level of the received solar radiation. The temperature is calculated using a simple carbon-silicate cycle model to change the level of $\rm CO_2$ in the atmosphere based on the chosen parameters. The model can achieve a temperature range exceeding $-100^\circ$C to $100^\circ$C by varying all three parameters, including freeze-thaw cycles for a planet with our present-day volcanism rate and emerged land fraction situated at the outer edge of the habitable zone. To increase engagement, the planet is visualised by using a neural network to render an animated globe, based on the calculated average surface temperature and chosen values for land fraction and volcanism. The website and bot can be found at earthlike.world and on twitter as @earthlikeworld. Initial feedback via a user survey suggested that Earth-Like is effective at demonstrating that minor changes in planetary properties can strongly impact the surface environment. The goal of the project is to increase understanding of the challenges we face in finding another habitable planet due to the likely diversity of conditions on rocky worlds within our Galaxy., Comment: Accepted for publication in the International Journal of Astrobiology (IJA)

While thousands of exoplanets have been confirmed, the known properties about individual discoveries remain sparse and depend on detection technique. To utilize more than a small section of the exoplanet dataset, tools need to be developed to estimate missing values based on the known measurements. Here, we demonstrate the use of a neural network that models the density of planets in a space of six properties that is then used to impute a probability distribution for missing values. Our results focus on planetary mass which neither the radial velocity nor transit techniques for planet identification can provide alone. The neural network can impute mass across the four orders of magnitude in the exoplanet archive, and return a distribution of masses for each planet that can inform about trends in the underlying dataset. The average error on this mass estimate from a radial velocity detection is a factor of 1.5 of the observed value, and 2.7 for a transit observation. The mass of Proxima Centauri b found by this method is $1.6^{\rm +0.46}_{\rm -0.36}$ M$_\oplus$, where the upper and lower bounds are derived from the root mean square deviation from the log mass probability distribution. The network can similarly impute the other potentially missing properties, and we use this to predict planet radius for radial velocity measurements, with an average error of a factor 1.4 of the observed value. The ability of neural networks to search for patterns in multidimensional data means that such techniques have the potential to greatly expand the use of the exoplanet catalogue., Comment: 17 page, 15 figures, accepted for publication in AJ

The habitable zone (HZ) is the region around a star(s) where standing bodies of water could exist on the surface of a rocky planet. The classical HZ definition makes a number of assumptions common to the Earth, including assuming that the most important greenhouse gases for habitable planets are CO2 and H2O, habitable planets orbit main-sequence stars, and that the carbonate-silicate cycle is a universal process on potentially habitable planets. Here, we discuss these and other predictions for the habitable zone and the observations that are needed to test them. We also, for the first time, argue why A-stars may be interesting HZ prospects. Instead of relying on unverified extrapolations from our Earth, we argue that future habitability studies require first principles approaches where temporal, spatial, physical, chemical, and biological systems are dynamically coupled. We also suggest that next-generation missions are only the beginning of a much more data-filled era in the not-too-distant future, when possibly hundreds to thousands of HZ planets will yield the statistical data we need to go beyond just finding habitable zone planets to actually determining which ones are most likely to exhibit life., Comment: White paper submitted to the NAS Astro 2020 decadal survey(5 pages, 1 figure + cover page, 13 co-authors + 10 co-signers). First author name added to description. In this arxiv version, one more co-signer was added after the NAS version was submitted. Link: http://surveygizmoresponseuploads.s3.amazonaws.com/fileuploads/623127/4458621/64-2dc3fcc201a3075e3da6233f1a91c668_RamirezRamsesM.pdf

With the advent of modern observational efforts providing extensive giant molecular cloud catalogues, understanding the evolution of such clouds in a galactic context is of prime importance. While numerous previous numerical and theoretical works have focused on the cloud properties in isolated discs, few have looked into the cloud population in an interacting disc system. We present results of the first study investigating the evolution of the cloud population in galaxy experiencing an M51-like tidal fly-by using numerical simulations including star formation, interstellar medium cooling and stellar feedback. We see the cloud population shift to large unbound clouds in the wake of the companion passage, with the largest clouds appearing as fleeting short-lived agglomerations of smaller clouds within the tidal spiral arms, brought together by large scale streaming motions. These are then sheared apart as they leave the protection of the spiral arms. Clouds appear to lead diverse lives, even within similar environments, with some being born from gas shocked by filaments streaming into the spiral arms, and others from effectively isolated smaller colliding pairs. Overall this cloud population produces a shallower mass function than the disc in isolation, especially in the arms compared to the inter-arm regions. Direct comparisons to M51 observations show similarities between cloud populations, though models tailored to the mass and orbital models of M51 appear necessary to precisely reproduce the cloud population., Comment: 21 pages, 25 figures, accepted for publication in MNRAS July 25 2018

We explore the effect of different galactic disc environments on the properties of star-forming clouds through variations in the background potential in a set of isolated galaxy simulations. Rising, falling and flat rotation curves expected in halo dominated, disc dominated and Milky Way-like galaxies were considered, with and without an additional two-arm spiral potential. The evolution of each disc displayed notable variations that are attributed to different regimes of stability, determined by shear and gravitational collapse. The properties of a typical cloud were largely unaffected by the changes in rotation curve, but the production of small and large cloud associations was strongly dependent on this environment. This suggests that while differing rotation curves can influence where clouds are initially formed, the average bulk properties are effectively independent of the global environment. The addition of a spiral perturbation made the greatest difference to cloud properties, successfully sweeping the gas into larger, seemingly unbound, extended structures and creating large arm-interarm contrasts., Comment: Accepted to MNRAS on 3rd December, 2017

We investigate star formation occurring in idealised giant molecular clouds, comparing structures that evolve in isolation versus those undergoing a collision. Two different collision speeds are investigated and the impact of photoionising radiation from the stars is determined. We find that a colliding system leads to more massive star formation both with and without the addition of feedback, raising overall star formation efficiencies (SFE) by a factor of 10 and steepening the high-mass end of the stellar mass function. This rise in SFE is due to increased turbulent compression during the cloud collision. While feedback can both promote and hinder star formation in the isolated system, it increases the SFE by approximately 1.5 times in the colliding case when the thermal speed of the resulting HII regions matches the shock propagation speed in the collision., Comment: 19 pages, 7 figures, 1 table, accepted for publication in PASJ

We have found many Earth-sized worlds but we have no way of determining if their surfaces are Earth-like. This makes it impossible to quantitatively compare habitability, and pretending we can risks damaging the field., Comment: Published in Nature Astronomy, Comment, 02/2017

We performed sub-parsec ($\sim$ 0.014 pc) scale simulations of cloud-cloud collisions of two idealized turbulent molecular clouds (MCs) with different masses in the range of $0.76 - 2.67 \times 10^4$M$_{\odot}$ and with collision speeds of 5 $-$ 30 km/s. Those parameters are larger than Takahira, Tasker and Habe (2014) (paper I) in which the colliding system showed a partial gaseous arc morphology that supports the NANTEN observations of objects indicated to be colliding MCs by numerical simulations. Gas clumps with density greater than $10^{-20}$ g cm$^{-3}$ were identified as pre-stellar cores and tracked through the simulation to investigate the effect of mass of colliding clouds and collision speeds on the resulting core population. Our results demonstrate that smaller cloud property is more important for results of cloud cloud collisions. The mass function of formed cores can be approximated by a power law relation with index $\gamma$ = -1.6 in slower cloud cloud collisions ($v \sim 5 $ km/s), in good agreement with observation of MCs. A faster relative velocity increases the number of cores formed in the early stage of collisions and shortens gas accretion phase of cores in the shocked region, leading to suppression of core growth. The bending point appears in the high mass part of the core mass function and the bending point mass decreases with increasing of the collision velocity for the same combination of colliding clouds. The high mass part of the core mass function than the bending point mass can be approximated by a power law with $\gamma$ = -2.3 that is very similar to the power index of the massive part of the observed initial stellar mass function. We discuss implication of our results for the massive star formation in our Galaxy., Comment: 15 pages, 15 figures, PASJ in press

Tidal encounters are believed to be one of the key drivers of galactic spiral structure in the Universe. Such spirals are expected to produce different morphological and kinematic features compared to density wave and dynamic spiral arms. In this work we present high resolution simulations of a tidal encounter of a small mass companion with a disc galaxy. Included are the effects of gas cooling and heating, star formation and stellar feedback. The structure of the perturbed disc differs greatly from the isolated galaxy, showing clear spiral features that act as sites of new star formation, and displaying interarm spurs. The two arms of the galaxy, the bridge and tail, appear to behave differently; with different star formation histories and structure. Specific attention is focused on offsets between gas and stellar spiral features which can be directly compared to observations. We find some offsets do exist between different media, with gaseous arms appearing mostly on the convex side of the stellar arms, though the exact locations appear highly time dependent. These results further highlight the differences between tidal spirals and other theories of arm structure., Comment: 17 pages, 19 colour figures, accepted for publication in MNRAS

We determine the physical properties and turbulence driving mode of molecular clouds formed in numerical simulations of a Milky Way-type disc galaxy with parsec-scale resolution. The clouds form through gravitational fragmentation of the gas, leading to average values for mass, radii and velocity dispersion in good agreement with observations of Milky Way clouds. The driving parameter (b) for the turbulence within each cloud is characterised by the ratio of the density contrast (sigma_rho) to the average Mach number (Mach) within the cloud, b = sigma_rho/Mach. As shown in previous works, b ~ 1/3 indicates solenoidal (divergence-free) driving and b ~ 1 indicates compressive (curl-free) driving. We find that the average b value of all the clouds formed in the simulations has a lower limit of b > 0.2. Importantly, we find that b has a broad distribution, covering values from purely solenoidal to purely compressive driving. Tracking the evolution of individual clouds reveals that the b value for each cloud does not vary significantly over their lifetime. Finally, we perform a resolution study with minimum cell sizes of 8, 4, 2 and 1 pc and find that the average b value increases with increasing resolution. Therefore, we conclude that our measured b values are strictly lower limits and that a resolution better than 1 pc is required for convergence. However, regardless of the resolution, we find that b varies by factors of a few in all cases, which means that the effective driving mode alters significantly from cloud to cloud., Comment: 12 pages, 11 figures, accepted for publication in MNRAS, more info: https://www.mso.anu.edu.au/~chfeder/pubs/turb_driv_gal/turb_driv_gal.html

We investigated the effect of photoionising feedback inside turbulent star-forming clouds, comparing the resultant star formation in both idealised profiles and more realistic cloud structures drawn from a global galaxy simulation. We performed a series of numerical simulations which compared the effect of star formation alone, photoionisation and photoionisation plus supernovae feedback. In the idealised cloud, photoionisation suppresses gas fragmentation at early times, resulting in the formation of more massive stars and an increase in the star formation efficiency. At later times, the dispersal of the dense gas causes the radiative feedback effect to switch from positive to negative as the star formation efficiency drops. In the cloud extracted from the global simulation, the initial cloud is heavily fragmented prior to the stellar feedback beginning and is largely structurally unaffected by the late injection of radiation energy. The result is a suppression of the star formation. We conclude that the efficiency of feedback is heavily dependent on the gas structure, with negative feedback dominating when the density is high., Comment: Accepted to MNRAS

We explore the impact of star formation and thermal stellar feedback on the giant molecular cloud (GMC) population forming in a M83-type barred spiral galaxy. We compare three high-resolution simulations (1.5 pc cell size) with different star formation/feedback models: one with no star formation, one with star formation but no feedback, and one with star formation and thermal energy injection. We analyze the resulting population of clouds, finding that we can identify the same population of massive, virialized clouds and transient, low-surface density clouds found in our previous work (that did not include star formation or feedback). Star formation and feedback can affect the mix of clouds we identify. In particular, star formation alone simply converts dense cloud gas into stars with only a small change to the cloud populations, principally resulting in a slight decrease in the transient population. Feedback, however, has a stronger impact: while it is not generally sufficient to entirely destroy the clouds, it does eject gas out of them, increasing the gas density in the inter-cloud region. This decreases the number of massive clouds, but substantially increases the transient cloud population. We also find that feedback tends to drive a net radial inflow of massive clouds, leading to an increase in the star formation rate in the bar region. We examine a number of possible reasons for this and conclude that it is possible that the drag force from the enhanced intercloud density could be responsible., Comment: 19 pages, 16 figures, accepted for publication in MNRAS

Tidal interactions between disc galaxies and low mass companions are an established method for generating galactic spiral features. In this work we present a study of the structure and dynamics of spiral arms driven in interactions between disc galaxies and perturbing companions in 3-D N-body/smoothed hydrodynamical numerical simulations. Our specific aims are to characterize any differences between structures formed in the gas and stars from a purely hydrodynamical and gravitational perspective, and to find a limiting case for spiral structure generation. Through analysis of a number of different interacting cases, we find that there is very little difference between arm morphology, pitch angles and pattern speeds between the two media. The main differences are a minor offset between gas and stellar arms, clear spurring features in gaseous arms, and different radial migration of material in the stronger interacting cases. We investigate the minimum mass of a companion required to drive spiral structure in a galactic disc, finding the limiting spiral generation cases with companion masses of the order $1\times10^9M_\odot$, equivalent to only 4% of the stellar disc mass, or 0.5% of the total galactic mass of a Milky Way analogue., Comment: 20 pages, 23 figures, accepted for publication by MNRAS

With ALMA making it possible to resolve giant molecular clouds (GMCs) in other galaxies, it is becoming necessary to quantify the observational bias on measured GMC properties. Using a hydrodynamical simulation of a barred spiral galaxy, we compared the physical properties of GMCs formed in position-position-position space (PPP) to the observational position-position-velocity space (PPV). We assessed the effect of disc inclination: face-on (PPV_face) and edge-on (PPV_edge), and resolution: 1.5 pc versus 24 pc, on GMC properties and the further implications of using Larson's scaling relations for mass-radius and velocity dispersion-radius. The low-resolution PPV data are generated by simulating ALMA Cycle 3 observations using the CASA package. Results show that the median properties do not differ strongly between PPP and PPV_face under both resolutions, but PPV_edge clouds deviate from these two. The differences become magnified when switching to the lower, but more realistic resolution. The discrepancy can lead to opposite results for the virial parameter's measure of gravitational binding, and therefore the dynamical state of the clouds. The power-law indices for the two Larson's scaling relations decrease going from PPP, PPV_face to PPV_edge and decrease from high to low resolutions. We conclude that the relations are not entirely driven by the underlying physical origin and therefore have to be used with caution when considering the environmental dependence, dynamical state, and the extragalactic CO-to-H2 conversion factor of GMCs., Comment: 12 pages, 5 figures, accepted for publication in MNRAS

As both simulations and observations reach the resolution of the star-forming molecular clouds, it becomes important to clarify if these two techniques are discussing the same objects in galaxies. We compare clouds formed in a high resolution galaxy simulation identified as continuous structures within a contour, in the simulator's position-position-position (PPP) co-ordinate space and the observer's position-position-velocity space (PPV). Results indicate that the properties of the cloud populations are similar in both methods and up to 70% of clouds have a single counterpart in the opposite data structure. Comparing individual clouds in a one-to-one match reveals a scatter in properties mostly within a factor of two. However, the small variations in mass, radius and velocity dispersion produce significant differences in derived quantities such as the virial parameter. This makes it difficult to determine if a structure is truely gravitationally bound. The three cloud types originally found in the simulation in Fujimoto et al. (2014) are identified in both data sets, with around 80% of the clouds retaining their type between identification methods. We also compared our results when using a peak decomposition method to identify clouds in both PPP and PPV space. The number of clouds increased with this technique, but the overall cloud properties remained similar. However, the more crowded environment lowered the ability to match clouds between techniques to 40%. The three cloud types also became harder to separate, especially in the PPV data set. The method used for cloud identification therefore plays a critical role in determining cloud properties, but both PPP and PPV can potentially identify the same structures., Comment: Accepted for publication in MNRAS

Stellar feedback, star formation and gravitational interactions are major controlling forces in the evolution of Giant Molecular Clouds (GMCs). To explore their relative roles, we examine the properties and evolution of GMCs forming in an isolated galactic disk simulation that includes both localised thermal feedback and photoelectric heating. The results are compared with the three previous simulations in this series which consists of a model with no star formation, star formation but no form of feedback and star formation with photoelectric heating in a set with steadily increasing physical effects. We find that the addition of localised thermal feedback greatly suppresses star formation but does not destroy the surrounding GMC, giving cloud properties closely resembling the run in which no stellar physics is included. The outflows from the feedback reduce the mass of the cloud but do not destroy it, allowing the cloud to survive its stellar children. This suggests that weak thermal feedback such as the lower bound expected for supernova may play a relatively minor role in the galactic structure of quiescent Milky Way-type galaxies, compared to gravitational interactions and disk shear., Comment: 15 pages, 15 figures, accepted for publication in ApJ

Cloud collision have been proposed as a way to link the small-scale star formation process with the observed global relation between the surface star formation rate and gas surface density. We suggest that this model can be improved further by allowing the productivity of such collisions to depend on the relative velocity of the two clouds. Our adjustment implements a simple step function that results in the most successful collisions being at the observed velocities for triggered star formation. By applying this to a high resolution simulation of a barred galaxy, we successfully reproduce the observational result that the star formation efficiency (SFE) in the bar is lower than that in the spiral arms. This is not possible when we use an efficiency dependent on the internal turbulence properties of the clouds. Our results suggest that high velocity collisions driven by the gravitational pull of the clouds are responsible for the low bar SFE., Comment: 6 pages, 4 figures. Accepted for publication in MNRAS Letters

We performed sub-parsec (~0.06pc) scale simulations of two idealised molecular clouds with different masses undergoing a collision. Gas clumps with density greater than 1e-20 g/cm3 (0.3e4 cm-3) were identified as pre-stellar cores and tracked through the simulation. The colliding system showed a partial gas arc morphology with core formation in the oblique shock-front at the collision interface. These characteristics support NANTEN observations of objects suspected to be colliding giant molecular clouds (GMCs). We investigated the effect of turbulence and collision speed on the resulting core population and compared the cumulative mass distribution to cores in observed GMCs. Our results suggest that a faster relative velocity increases the number of cores formed but that cores grow via accretion predominately while in the shock front, leading to a slower shock being more important for core growth. The core masses obey a power law relation with index gamma = -1.6, in good agreement with observations. This suggests that core production through collisions should follow a similar mass distribution as quiescent formation, albeit at a higher mass range. If cores can be supported against collapse during their growth, the estimated ram pressure from gas infall is of the right order to counter the radiation pressure and form a star of 100Msun., Comment: Accepted for publication in ApJ. 16 pages, 14 figures

We studied the gravitational collapse of a warm (~8000 K) primordial-gas cloud as a candidate progenitor for a supermassive star (SMS; >10^5 Msun) using a three-dimensional hydrodynamical simulation, including all the relevant cooling processes of both H_2 and H, which can potentially induce cloud fragmentation. This is the first simulation of this kind to resolve protostar formation. We find that the cloud undergoes runaway collapse without a major episode of fragmentation. Although the H_2 fraction jumps by a large factor via the three-body reaction at ~10^-13 g/cm^3, its cooling remains inefficient due to the optical thickness, and the temperature remains >3000 K. When the central core of the cloud becomes opaque to continuum radiation at ~10^-8 g/cm^3, a hydrostatic protostar with ~0.2 Msun is formed. The protostar grows to the mass ~1 Msun and the radius ~2 AU within ~1 yr via rapid accretion of dense filamentary flows. With high accretion rate ~2 Msun/yr, the protostar is expected to turn into a SMS within its lifetime, eventually collapsing to a seed for the supermassive black hole observed in the early Universe at z~7., Comment: 6 pages, 6 figures, Accepted for publication in MNRAS

We investigate the impact of galactic environment on the properties of simulated giant molecular clouds formed in a M83-type barred spiral galaxy. Our simulation uses a rotating stellar potential to create the grand design features and resolves down to 1.5 pc. From the comparison of clouds found in the bar, spiral and disc regions, we find that the typical GMC is environment independent, with a mass of 5e+5 Msun and radius 11 pc. However, the fraction of clouds in the property distribution tails varies between regions, with larger, more massive clouds with a higher velocity dispersion being found in greatest proportions in the bar, spiral and then disc. The bar clouds also show a bimodality that is not reflected in the spiral and disc clouds except in the surface density, where all three regions show two distinct peaks. We identify these features as being due to the relative proportion of three cloud types, classified via the mass-radius scaling relation, which we label A, B and C. Type A clouds have the typical values listed above and form the largest fraction in each region. Type B clouds are massive giant molecular associations while Type C clouds are unbound, transient clouds that form in dense filaments and tidal tails. The fraction of each clouds type depends on the cloud-cloud interactions, which cause mergers to build up the GMA Type Bs and tidal features in which the Type C clouds are formed. The number of cloud interactions is greatest in the bar, followed by the spiral, causing a higher fraction of both cloud types compared to the disc. While the cloud types also exist in lower resolution simulations, their identification becomes more challenging as they are not well separated populations on the mass-radius relation or distribution plots. Finally, we compare the results for three star formation models to estimate the star formation rate and efficiency in each region., Comment: 21 pages, 14 figures. Accepted for publication in MNRAS

We use simulations of isolated galaxies with a few parsec resolution to explore the connection between the small-scale star formation rate - gas density relation and the induced large-scale correlation between the star formation rate surface density and the surface density of the molecular gas (the Kennicutt-Schmidt relation). We find that, in the simulations, a power-law small-scale "star formation law" directly translates into an identical power-law Kennicutt-Schmidt relation. If this conclusion holds in the reality as well, it implies that the observed approximately linear Kennicutt-Schmidt relation must reflect the approximately linear small-scale "star formation law"., Comment: Replaced with the accepted version

We present the results of a study of simulated Giant Molecular Clouds (GMCs) formed in a Milky Way-type galactic disk with a flat rotation curve. This simulation, which does not include star formation or feedback, produces clouds with masses ranging between 10^4 Msun and 10^7 Msun. We compare our simulated cloud population to two observational surveys; The Boston University- Five College Radio Astronomy Observatory Galactic Ring Survey and the BIMA All-Disk Survey of M33. An analysis of the global cloud properties as well as a comparison of Larson's scaling relations is carried out. We find that simulated cloud properties agree well with the observed cloud properties, with the closest agreement occurring between the clouds at comparable resolution in M33. Our clouds are highly filamentary - a property that derives both from their formation due to gravitational instability in the sheared galactic environment, as well as to cloud- cloud gravitational encounters. We also find that the rate at which potentially star forming gas accumulates within dense regions - wherein n_{thresh} > 10^4 cm^{-3} - is 3% per 10 Myr, in clouds of roughly 10^6 Msun. This suggests that star formation rates in observed clouds are related to the rates at which gas can be accumulated into dense subregions within GMCs via filamentary flows. The most internally well-resolved clouds are chosen for listing in a catalogue of simulated GMCs; the first of its kind. The catalogued clouds are available as an extracted data set from the global simulation., Comment: 13 pages, 11 figures, 1 table; Accepted for publication in ApJ

This paper describes the open-source code Enzo, which uses block-structured adaptive mesh refinement to provide high spatial and temporal resolution for modeling astrophysical fluid flows. The code is Cartesian, can be run in 1, 2, and 3 dimensions, and supports a wide variety of physics including hydrodynamics, ideal and non-ideal magnetohydrodynamics, N-body dynamics (and, more broadly, self-gravity of fluids and particles), primordial gas chemistry, optically-thin radiative cooling of primordial and metal-enriched plasmas (as well as some optically-thick cooling models), radiation transport, cosmological expansion, and models for star formation and feedback in a cosmological context. In addition to explaining the algorithms implemented, we present solutions for a wide range of test problems, demonstrate the code's parallel performance, and discuss the Enzo collaboration's code development methodology., Comment: 60 pages, 22 figures, submitted to ApJS. The code is available at http://enzo-project.org

We investigate the effect of star formation and diffuse photoelectric heating on the properties of giant molecular clouds (GMCs) formed in high resolution (~< 10 pc) global (~ 20 kpc) simulations of isolated Milky Way-type galaxy disks. The clouds are formed through gravitational fragmentation and structures with densities n_H>=100cm^-3 are identified as GMCs. Between 1000-1500 clouds are created in the simulations with masses M > 10^5 Msolar and 180-240 with masses M > 10^6 Msolar in agreement with estimates of the Milky Way's population. We find that the effect of photoelectric heating is to suppress the fragmentation of the ISM, resulting in a filamentary structure in the warm gas surrounding clouds. This environment suppresses the formation of a retrograde rotating cloud population, with 88% of the clouds rotating prograde with respect to the galaxy after 300 Myr. The diffuse heating also reduces the initial star formation rate, slowing the conversation of gas into stars. We therefore conclude that the interstellar environment plays an important role in the GMCs evolution. Our clouds live between 0-20 Myr with a high infant mortality (t' < 3 Myr) due to cloud mergers and star formation. Other properties, including distributions of mass, size and surface density agree well with observations. Collisions between our clouds are common, occurring at a rate of ~1/4 of the orbital period. It is not clear whether such collisions trigger or suppress star formation at our current resolution. Our star formation rate is a factor of 10 higher than observations in local galaxies. This is likely due to the absence of localized feedback in our models., Comment: 25 pages. 18 figures. Accepted for publication in ApJ

We investigate the formation and evolution of giant molecular clouds (GMCs) in a Milky-Way-like disk galaxy with a flat rotation curve. We perform a series of 3D adaptive mesh refinement (AMR) numerical simulations that follow both the global evolution on scales of ~20kpc and resolve down to scales ~<10pc with a multiphase atomic interstellar medium (ISM). In this first study, we omit star formation and feedback, and focus on the processes of gravitational instability and cloud collisions and interactions. We define clouds as regions with n_H>=100cm^-3 and track the evolution of individual clouds as they orbit through the galaxy from their birth to their eventual destruction via merger or via destructive collision with another cloud. After ~140Myr a large fraction of the gas in the disk has fragmented into clouds with masses ~10^6 Msun and a mass spectrum similar to that of Galactic GMCs. The disk settles into a quasi steady state in which gravitational scattering of clouds keeps the disk near the threshold of global gravitational instability. The cloud collision time is found to be a small fraction, ~1/5, of the orbital time, and this is an efficient mechanism to inject turbulence into the clouds. This helps to keep clouds only moderately gravitationally bound, with virial parameters of order unity. Many other observed GMC properties, such as mass surface density, angular momentum, velocity dispersion, and vertical distribution, can be accounted for in this simple model with no stellar feedback., Comment: 21 pages ApJ format, including 16 figures, accepted to ApJ

We test four commonly used astrophysical simulation codes; Enzo, Flash, Gadget and Hydra, using a suite of numerical problems with analytic initial and final states. Situations similar to the conditions of these tests, a Sod shock, a Sedov blast and both a static and translating King sphere occur commonly in astrophysics, where the accurate treatment of shocks, sound waves, supernovae explosions and collapsed haloes is a key condition for obtaining reliable validated simulations. We demonstrate that comparable results can be obtained for Lagrangian and Eulerian codes by requiring that approximately one particle exists per grid cell in the region of interest. We conclude that adaptive Eulerian codes, with their ability to place refinements in regions of rapidly changing density, are well suited to problems where physical processes are related to such changes. Lagrangian methods, on the other hand, are well suited to problems where large density contrasts occur and the physics is related to the local density itself rather than the local density gradient., Comment: 18 pages, 15 figures. Accepted for publication in MNRAS

Until recently, simulations that modeled entire galaxies were restricted to an isothermal or fixed 2- or 3-phase interstellar medium (ISM). This obscured the full role of the ISM in shaping the observed galactic-scale star formation relations. In particular the Kennicutt relation suggests that star formation rates depend in a simple way on global galactic quantities, such as mean gas mass surface density and dynamical time. Contrary to this, observations of nearby star-forming regions, including images from the Spitzer telescope, show that all the way down to ``local-scales'' ($\sim$ parsecs) star formation is a highly clustered process with the gas existing at a wide range of densities and multiple phases, and seemingly decoupled from the larger-scale Galaxy. Many different physical processes appear to be influencing the star formation rate, including heating and cooling of the gas, ``turbulent'' energy injection from outflows, winds and supernova blastwaves, ionization, and magnetic field support. Here we present results from global simulations of disk galaxy star formation that include a fully multiphase ISM and compare to simpler isothermal models. We also investigate the effect of background heating. We discuss the process of giant molecular cloud formation as one of the possible links between the galactic and local scales of star formation., Comment: 7 pages, 5 figures: Proceedings for "The Evolving ISM in the Milky Way & Nearby Galaxies", 4th Spitzer Science Center Conference

Modelling global disc galaxies is a difficult task which has previously resulted in the small scale physics of the interstellar medium being greatly simplified. In this talk, I compare simulations of galaxies with different ISM properties to determine the importance of the ISM structure in the star formation properties of the disc., Comment: 4 pages, proceedings from the conference "Formation and Evolution of Galaxy Disks", Rome, October 1-5, 2007. To be published in the series of the Astronomical Society of the Pacific

We have carried out a hydrodynamical code comparison study of interacting multiphase fluids. The two commonly used techniques of grid and smoothed particle hydrodynamics (SPH) show striking differences in their ability to model processes that are fundamentally important across many areas of astrophysics. Whilst Eulerian grid based methods are able to resolve and treat important dynamical instabilities, such as Kelvin-Helmholtz or Rayleigh-Taylor, these processes are poorly or not at all resolved by existing SPH techniques. We show that the reason for this is that SPH, at least in its standard implementation, introduces spurious pressure forces on particles in regions where there are steep density gradients. This results in a boundary gap of the size of the SPH smoothing kernel over which information is not transferred., Comment: 15 pages, 13 figures, to be submitted to MNRAS. For high-resolution figures, please see http://www-theorie.physik.unizh.ch/~agertz/

We perform cosmological simulations of galaxies forming at z=3 using the hydrodynamics grid code, Enzo. By selecting the largest galaxies in the volume to correspond to Lyman-break galaxies, we construct observational spectra of the HI flux distribution around these objects, as well as column densities of CIV and OVI throughout a refined region. We successfully reproduce the most recent observations of the mean HI flux in the close vicinity of Lyman-break galaxies but see no evidence for the proximity effect in earlier observations. While our galaxies do return metals to the IGM, their quantity and volume appears to be somewhat less than observed. We conclude that either we do not adequately resolve galactic winds, or that at least some of the intergalactic metal enrichment is by early epoch objects whose mass is smaller than our minimum resolved halo mass., Comment: To be published in ApJL, in press. 4 pages, 4 figures; added references to work by Ferrara to sections 1 and 3.2

We use a high-resolution grid-based hydrodynamics method to simulate the multi-phase interstellar medium in a Milky Way-size quiescent disk galaxy. The models are global and three-dimensional, and include a treatment of star formation and feedback. We examine the formation of gravitational instabilities and show that a form of the Toomre instability criterion can successfully predict where star formation will occur. Two common prescriptions for star formation are investigated. The first is based on cosmological simulations and has a relatively low threshold for star formation, but also enforces a comparatively low efficiency. The second only permits star formation above a number density of 1000 cm^-3 but adopts a high efficiency. We show that both methods can reproduce the observed slope of the relationship between star formation and gas surface density (although at too high a rate for our adopted parameters). A run which includes feedback from type II supernovae is successful at driving gas out of the plane, most of which falls back onto the disk. This feedback also substantially reduces the star formation rate. Finally, we examine the density and pressure distribution of the ISM, and show that there is a rough pressure equilibrium in the disk, but with a wide range of pressures at a given location (and even wider for the case including feedback, Comment: 14 pages, 12 figures, accepted to Astrophysical Journal