Antarctic pack-ice seals, a group of four species of true seals (Phocidae), play a pivotal role in the Southern Ocean foodweb as wide-ranging predators of Antarctic krill (Euphausia superba). Due to their circumpolar distribution and the remoteness and vastness of their habitat, little is known about their population sizes. Estimating pack-ice seal population sizes and trends is key to understanding how the Southern Ocean ecosystem will react to threats such as climate change driven sea ice loss and krill fishing. We present a functional pack-ice seal detection pipeline using Worldview-3 imagery and a Convolutional Neural Network that counts and locates seal centroids. We propose a new CNN architecture that detects objects by combining semantic segmentation heatmaps with binary classification and counting by regression. Our pipeline locates over 30% of seals, when compared to consensus counts from human experts, and reduces the time required for seal detection by 95% (assuming just a single GPU). While larger training sets and continued algorithm development will no doubt improve classification accuracy, our pipeline, which can be easily adapted for other large-bodied animals visible in sub-meter satellite imagery, demonstrates the potential for machine learning to vastly expand our capacity for regular pack-ice seal surveys and, in doing so, will contribute to ongoing international efforts to monitor pack-ice seals.

Spitzer & Chandra observations of the massive star forming region RCW 38 identify 624 young stellar objects (YSOs) in four subclusters surrounding the previously known cluster centred on the O5.5 binary IRS 2.

IRAS 20050+2720 is a young star forming region at a distance of 700 pc with a mass of 430 M⊙, but without apparent high mass stars. We characterize the distribution of young stellar objects (YSOs) in this region, which lacks high UV flux, and compare evolutionary trends with other young star forming clusters. We present results of our multiwavelength study of IRAS 20050+2720 which includes observations by Chandra, Spitzer, and 2MASS and UBVRI photometry. In total, about 300 YSOs in different evolutionary stages are found. We newly identify a second cluster core, which consists mostly of class II objects, about 2 pc from the center of the cloud. YSOs of earlier evolutionary stages are more clustered than more evolved objects. The X-ray luminosity function (XLF) of IRAS 20050+2720 is roughly lognormal, but steeper than the XLF of the more massive Orion nebula complex. The dust in IRAS 20050+2720 seems altered compared to the ISM.

IRAS 20050+2720 is young star-forming region at a distance of 700 pc without apparent high-mass stars. We present results of our multi-wavelength study of IRAS 20050+2720 which includes observations by Chandra and Spitzer, and Two Micron All Sky Survey and UBVRI photometry. In total, about 300 young stellar objects (YSOs) in different evolutionary stages are found. We characterize the distribution of YSOs in this region using a minimum spanning tree analysis. We newly identify a second cluster core, which consists mostly of class II objects, about 10' from the center of the cloud. YSOs of earlier evolutionary stages are more clustered than more evolved objects. The X-ray luminosity function (XLF) of IRAS 20050+2720 is roughly lognormal, but steeper than the XLF of the more massive Orion Nebula complex. IRAS 20050+2720 shows a lower N-H/A(K) ratio compared with the diffuse interstellar medium.

We report Spitzer observations of five newly identified bow shocks in the massive star-forming region RCW 38. Four are visible at Infrared Array Camera (IRAC) wavelengths, the fifth is only visible at 24 mu m. Chandra X-ray emission indicates that winds from the central O5.5 binary, IRS 2, have caused an outflow to the northeast and southwest of the central subcluster. The southern lobe of hot ionized gas is detected in X-rays; shocked gas and heated dust from the shock front are detected with Spitzer at 4.5 and 24 mu m. The northern outflow may have initiated the present generation of star formation, based on the filamentary distribution of the protostars in the central subcluster. Further, the bow-shock driving star, YSO 129, is photo-evaporating a pillar of gas and dust. No point sources are identified within this pillar at near-to mid-IR wavelengths. We also report on IRAC 3.6 and 5.8 mu m observations of the cluster DBS2003-124, northeast of RCW 38, where 33 candidate young stellar objects (YSOs) are identified. One star associated with the cluster drives a parsec-scale jet. Two Herbig-Haro objects associated with the jet are visible at IRAC and Multiband Imaging Photometer for Spitzer (MIPS) wavelengths. The jet extends over a distance of similar to 3 pc. Assuming a velocity of 100 km s(-1) for the jet material gives an age of 3 x 10(4) yr, indicating that the star (and cluster) are likely to be very young, with a similar or possibly younger age than RCW 38, and that star formation is ongoing in the extended RCW 38 region.

We present a study of the structure of the high-mass star-forming region RCW 38 and the spatial distribution of its young stellar population. Spitzer Infrared Array Camera (IRAC) photometry (3-8 mu m) is combined with Two Micron All Sky Survey near-IR data to identify young stellar objects (YSOs) by IR-excess emission from their circumstellar material. Chandra X-ray data are used to identify class III pre-main-sequence stars lacking circumstellar material. We identify 624 YSOs: 23 class 0/I and 90 flat spectrum protostars, 437 class II stars, and 74 class III stars. We also identify 29 (27 new) O star candidates over the IRAC field. Seventy-two stars exhibit IR-variability, including 7 class 0/I and 12 flat spectrum YSOs. A further 177 tentative candidates are identified by their location in the IRAC [3.6] versus [3.6]-[5.8] color-magnitude diagram. We find strong evidence of subclustering in the region. Three subclusters were identified surrounding the central cluster, with massive and variable stars in each subcluster. The central region shows evidence of distinct spatial distributions of the protostars and pre-main-sequence stars. A previously detected IR cluster, DB2001_Obj36, has been established as a subcluster of RCW 38. This suggests that star formation in RCW 38 occurs over a more extended area than previously thought. The gas-to-dust ratio is examined using the X-ray derived hydrogen column density, N-H and the K-band extinction, and found to be consistent with the diffuse interstellar medium, in contrast with Serpens and NGC 1333. We posit that the high photoionizing flux of massive stars in RCW 38 affects the agglomeration of the dust grains.

We present Chandra X-ray data of the NGC 1333 embedded cluster, combining these data with existing Chandra data, Sptizer photometry and ground based spectroscopy of both the NGC 1333 & Serpens North clusters to perform a detailed study of the X-ray properties of two of the nearest embedded clusters to the Sun. In NGC 1333, a total of 95 cluster members are detected in X-rays, of which 54 were previously identified with Spitzer. Of the Spitzer sources, we detect 23% of the Class I protostars, 53% of the Flat Spectrum sources, 52% of the Class II, and 50% of the Transition Disk YSOs. Forty-one Class III members of the cluster are identified, bringing the total identified YSO population to 178. The X-ray Luminosity Functions (XLFs) of the NGC 1333 and Serpens clusters are compared to each other and the Orion Nebula Cluster. Based on this comparison, we obtain a new distance for the Serpens cluster of 360+22/-13 pc. The X-ray luminosity was found to depend on the bolometric luminosity as in previous studies of other clusters, and that Lx depends primarily on the stellar surface area. In the NGC 1333 cluster, the Class III sources have a somewhat higher X-ray luminosity for a given surface area. We also find evidence in NGC 1333 for a jump in the X-ray luminosity between spectral types of M0 and K7, we speculate that this may result from the presence of radiative zones in the K-stars. The gas column density vs. extinction in the NGC 1333 was found to be N_H = 0.89 +/- 0.13 x 10^22 A_K, this is lower than expected of the standard ISM but similar to that found previously in the Serpens Cloud Core.

We present spectral observations of 130 young stellar objects (YSOs) in the Serpens Cloud Core and NGC 1333 embedded clusters. The observations consist of near-IR spectra in the H and K bands from SpeX on the IRTF and far-red spectra (6000–9000 Å) from Hectospec on the Multi-Mirror Telescope. These YSOs were identified in previous Spitzer and Chandra observations, and the evolutionary classes of the YSOs were determined from the Spitzer mid-IR photometry. With these spectra we search for corroborating evidence for the pre-main-sequence nature of the objects, study the properties of the detected emission lines as a function of evolutionary class, and obtain spectral types for the observed YSOs. The temperatures implied by the spectral types are combined with luminosities determined from the near-IR photometry to construct Hertzsprung–Russell (H–R) diagrams for the clusters. By comparing the positions of the YSOs in the H–R diagrams with the pre-main-sequence tracks of Baraffe (1998), we determine the ages of the embedded sources and study the relative ages of the YSOs with and without optically thick circumstellar disks. The apparent isochronal ages of the YSOs in both clusters range from less than 1 Myr to 10 Myr, with most objects below 3 Myr. The observed distributions of ages for the Class II and Class III objects are statistically indistinguishable. We examine the spatial distribution and extinction of the YSOs as a function of their isochronal ages. We find the sources <3 Myr to be concentrated in the molecular cloud gas, while the older sources are spatially dispersed and are not deeply embedded. Nonetheless, the sources with isochronal ages >3 Myr show all the characteristics of YSOs in their spectra, their IR spectral energy distributions, and their X-ray emission; we find no evidence that they are contaminating background giants or foreground dwarfs. However, we find no corresponding decrease in the fraction of sources with infrared excess with isochronal age; this suggests that the older isochronal ages may not measure the true age of the >3 Myr YSOs. Thus, the nature of the apparently older sources and their implications for cluster formation remain unresolved.

We present Spitzer and Chandra observations of the nearby ( similar to 260 pc) embedded stellar cluster in the Serpens cloud core. We observed, using Spitzer's IRAC and MIPS instruments, in six wavelength bands from 3 to 70 mu m, to detect thermal emission from circumstellar disks and protostellar envelopes and to classify stars using color-color diagrams and SEDs. These data are combined with Chandra observations to examine the effects of circumstellar disks on stellar X-ray properties. Young diskless stars were also identified from their increased X-ray emission. We have identified 138 YSOs in Serpens: 22 Class 0/I, 16 flat-spectrum, 62 Class II, 17 transition disk, and 21 Class III stars; 60 of these exhibit X-ray emission. Our primary results are the following: ( 1) 10 protostars detected previously in the submillimeter are detected at lambda < 24 mu m, seven at lambda < 8 mu m; ( 2) the protostars are more closely grouped than more evolved YSOs ( median protostar separation similar to 0.024 pc); and ( 3) the luminosity and temperature of the X-ray-emitting plasma around these YSOs do not show any significant dependence on evolutionary class. We combine the infrared-derived values of A(K) and X-ray values of N-H for eight Class III objects and find that the column density of hydrogen gas per magnitude of extinctions is less than half the standard interstellar value, for A(K) > 1. This may be the result of grain growth through coagulation and/or the accretion of volatiles in the Serpens cloud core.

We are analyzing ~200 sources found in a 40 ksec observation of the open cluster h Per. The data are being processed with the ANCHORS pipeline which provides fluxes and low resolution X-ray spectra. For the stronger sources temperatures are derived from spectral fitting; for weaker sources interpretation is through quantiles. The luminosity distribution is discussed, including the effects of shallow sampling on the source population. The distribution of these properties on optical and infrared color magnitude diagrams is investigated for the cool pre-main sequence stars in this 10 Myr cluster. Motivation: In a ROSAT PSPC image of h and χ Per, Evans and Seward (2000, ApJ, 538, 777) found that the X-ray sources were not spatially correlated with B stars in the clusters. The sources in the shallow (11 ksec) image were brighter than solar mass stars in the Orion Nebula Cluster (ONC), which is surprising since the ONC is younger (4 x 106 years) than h and χ Per (14 x 106 years). Because of the low resolution of the ROSAT image, individual optical sources could not be identified with the X-ray sources. Using the Chandra observation we can make these identifications, and see whether the X-ray luminosities are as expected at the age of the cluster, or whether, for instance, the unusually high rotation in the cluster affects the X-ray level. X-ray observations of the h Per field. The image was obtained by binning the image by 4 and applying gaussian smoothing with rkernel=3. K-band 2MASS image of the h Per region. Locations of X-ray sources are indicated as magenta circles. Optical matches have been made between optical photometry from Keller, et al. (2001, AJ, 122, 248) and also 2MASS J, H, and K photometry. The figure shown above is converted from V and B for stars within 5’ of the cluster center. Stars which are X-ray sources are marked with red diamonds. Optical photometry is only reliable to about 17th magnitude (the dotted line is at V=17.1mag, and the solid red line is at V=17.5mag). Approximately half the X-ray sources have optical matches. This is partly because we so far only have photometry as deep as F stars at the distance of h Per. We also expect that up to a quarter of the X-ray sources are background AGN. In this figure, the absolute magnitude Mv is plotted, using a distance of 2410 pc, and an E(B-V) = 0.56 mag. The (BV) color has NOT been corrected for reddening. Stellar masses at MV=2.4mag and MV=4.7mag are indicated for an age of 107 years, according to the tracks of Siess, Dufour, and Forestini (2000). Green asterisks indicate a tight sequence of X-ray sources, corresponding to F stars, with optical matches. (see discussion about those stars in the next figure on the right) Future work: We (TGB) are obtaining deeper optical photometry to determine the optical properties of X-ray sources which are cooler pre-Main Sequence stars. This will also be important in confirming sources suspected of being background AGN: those which have a very large Xray to optical ratio. MV=2.4 =>2.0M

One of the requirements of the Chandra mission is to be able to quickly examine spacecraft data which span the full ten year planned mission lifetime. It was clear from the outset that warehoused FITS files would not meet this requirement. We opted to create a set of databases which resample all engineering telemetry into a fixed five minute time grid. This allows a year's worth of data to be compressed to the scale of about 10 GB. Bi-level data were written into special databases which only report when bi-levels change. This not only results in space savings, but allows for very rapid responses to queries which are concerned with system settings. Currently we can access one year's worth of data across several tables in under 15 minutes. We discuss the design of the databases, the status of testing and implementation. Finally, we discuss ongoing database development and plans for handling image and source data.

Chandra's Monitoring and Trends Analysis (M&TA) team utilizes a series of database tables to store binned information about spacecraft configuration and condition over the life of the mission. The DATASEEKER was developed as an efficient means to specify a particular set of this data and have it conveniently downloaded in a usable format.In Chandra's brief operational life of a little over two years in orbit, there are already well over 10 GB of data stored in approximately 50 relational database tables for the Science Operation Team to sift through. DATASEEKER has made it possible to discriminately pull out small subsets of these data in a matter of minutes.

We report on results of a 96.7 ks Chandra observation of one of the youngest, most embedded and massive young stellar clusters studied to date in X-rays-RCW 38. We find a region of extended emission about 1.25 x 1.75 pc. The emission is consistent with synchrotron emission. The power index of the emission steepens toward 1.75 the cluster core, implying that the magnetic field originates there. This is the first evidence of synchrotron emission filling a region of active star formation. The source of the necessary magnetic field is unclear. We measure the hydrogen column to the emission and find that N-H increases from 7.6 x 10(21) to 1.61 x 10(22) cm(-2), increasing from the northwest to the southeast quadrants of the cluster, consistent with near-infrared extinction data.