Showing 1–12 of 12 results for author: Rosenberg, F

IBM Deep Learning Service

Authors: Bishwaranjan Bhattacharjee, Scott Boag, Chandani Doshi, Parijat Dube, Ben Herta, Vatche Ishakian, K. R. Jayaram, Rania Khalaf, Avesh Krishna, Yu Bo Li, Vinod Muthusamy, Ruchir Puri, Yufei Ren, Florian Rosenberg, Seetharami R. Seelam, Yandong Wang, Jian Ming Zhang, Li Zhang

Abstract: Deep learning driven by large neural network models is overtaking traditional machine learning methods for understanding unstructured and perceptual data domains such as speech, text, and vision. At the same time, the "as-a-Service"-based business model on the cloud is fundamentally transforming the information technology industry. These two trends: deep learning, and "as-a-service" are colliding… ▽ More Deep learning driven by large neural network models is overtaking traditional machine learning methods for understanding unstructured and perceptual data domains such as speech, text, and vision. At the same time, the "as-a-Service"-based business model on the cloud is fundamentally transforming the information technology industry. These two trends: deep learning, and "as-a-service" are colliding to give rise to a new business model for cognitive application delivery: deep learning as a service in the cloud. In this paper, we will discuss the details of the software architecture behind IBM's deep learning as a service (DLaaS). DLaaS provides developers the flexibility to use popular deep learning libraries such as Caffe, Torch and TensorFlow, in the cloud in a scalable and resilient manner with minimal effort. The platform uses a distribution and orchestration layer that facilitates learning from a large amount of data in a reasonable amount of time across compute nodes. A resource provisioning layer enables flexible job management on heterogeneous resources, such as graphics processing units (GPUs) and central processing units (CPUs), in an infrastructure as a service (IaaS) cloud. △ Less

Submitted 18 September, 2017; originally announced September 2017.

Constraining cloud parameters using high density gas tracers in galaxies

Authors: M. V. Kazandjian, I. Pelupessy, R. Meijerink, F. P. Israel, C. M. Coppola, M. J. F. Rosenberg, M. Spaans

Abstract: Far-infrared molecular emission is an important tool used to understand the excitation mechanisms of the gas in the inter-stellar medium of star-forming galaxies. In the present work, we model the emission from rotational transitions with critical densities n >~ 10^4 cm-3. We include 4-3 < J <= 15-14 transitions of CO and 13CO, in addition to J <= 7-6 transitions of HCN, HNC, and HCO+ on galactic… ▽ More Far-infrared molecular emission is an important tool used to understand the excitation mechanisms of the gas in the inter-stellar medium of star-forming galaxies. In the present work, we model the emission from rotational transitions with critical densities n >~ 10^4 cm-3. We include 4-3 < J <= 15-14 transitions of CO and 13CO, in addition to J <= 7-6 transitions of HCN, HNC, and HCO+ on galactic scales. We do this by re-sampling high density gas in a hydrodynamic model of a gas-rich disk galaxy, assuming that the density field of the interstellar medium of the model galaxy follows the probability density function (PDF) inferred from the resolved low density scales. We find that in a narrow gas density PDF, with a mean density of ~10 cm-3 and a dispersion σ= 2.1 in the log of the density, most of the emission of molecular lines, emanates from the 10-1000 cm-3 part of the PDF. We construct synthetic emission maps for the central 2 kpc of the galaxy and fit the line ratios of CO and 13CO up to J = 15-14, as well as HCN, HNC, and HCO+ up to J = 7-6, using one photo-dissociation region (PDR) model. We attribute the goodness of the one component fits for our model galaxy to the fact that the distribution of the luminosity, as a function of density, is peaked at gas densities between 10 and 1000 cm-3. We explore the impact of different log-normal density PDFs on the distribution of the line-luminosity as a function of density, and we show that it is necessary to have a broad dispersion, corresponding to Mach numbers >~ 30 in order to obtain significant emission from n > 10^4 cm-3 gas. Such Mach numbers are expected in star-forming galaxies, LIRGS, and ULIRGS. By fitting line ratios of HCN(1-0), HNC(1-0), and HCO+(1-0) for a sample of LIRGS and ULIRGS using mechanically heated PDRs, we constrain the Mach number of these galaxies to 29 < M < 77. △ Less

Submitted 8 September, 2016; originally announced September 2016.

Comments: 13 pages, 11 figures

Neutral carbon and CO in 76 (U)LIRGs and starburst galaxy centers A method to determine molecular gas properties in luminous galaxies

Authors: F. P. Israel, M. J. F. Rosenberg, P. van der Werf

Abstract: We present fluxes in both neutral carbon [CI] lines at the centers of 76 galaxies with FIR luminosities between 10^{9} and 10^{12} L(o) obtained with Herschel-SPIRE and with ground-based facilities, along with the J=7-6, J=4-3, J=2-1 12CO and J=2-1 13CO line fluxes. We investigate whether these lines can be used to characterize the molecular ISM of the parent galaxies in simple ways and how the mo… ▽ More We present fluxes in both neutral carbon [CI] lines at the centers of 76 galaxies with FIR luminosities between 10^{9} and 10^{12} L(o) obtained with Herschel-SPIRE and with ground-based facilities, along with the J=7-6, J=4-3, J=2-1 12CO and J=2-1 13CO line fluxes. We investigate whether these lines can be used to characterize the molecular ISM of the parent galaxies in simple ways and how the molecular gas properties define the model results. In most starburst galaxies, the [CI]/13CO flux ratio is much higher than in Galactic star-forming regions, and it is correlated to the total FIR luminosity. The [CI](1-0)/CO(4-3), the [CI](2-1) (2-1)/CO(7-6), and the [CI] (2-1)/(1-0) flux ratios are also correlated, and trace the excitation of the molecular gas. In the most luminous infrared galaxies (LIRGs), the ISM is fully dominated by dense and moderately warm gas clouds that appear to have low [C]/[CO] and [13CO]/[12CO] abundances. In less luminous galaxies, emission from gas clouds at lower densities becomes progressively more important, and a multiple-phase analysis is required to determine consistent physical characteristics. Neither the CO nor the [CI] velocity-integrated line fluxes are good predictors of H2 column densities in individual galaxies, and X(CI) conversion factors are not superior to X(CO) factors. The methods and diagnostic diagrams outlined in this paper also provide a new and relatively straightforward means of deriving the physical characteristics of molecular gas in high-redshift galaxies up to z=5, which are otherwise hard to determine. △ Less

Submitted 29 April, 2015; originally announced April 2015.

Journal ref: A&A 578, A95 (2015)

The Herschel Comprehensive (U)LIRG Emission Survey (HerCULES): CO Ladders, fine structure lines, and neutral gas cooling

Authors: M. J. F. Rosenberg, P. P. van der Werf, S. Aalto, L. Armus, V. Charmandaris, T. Díaz-Santos, A. S. Evans, J. Fischer, Y. Gao, E. González-Alfonso, T. R. Greve, A. I. Harris, C. Henkel, F. P. Israel, K. G. Isaak, C. Kramer, R. Meijerink, D. A. Naylor, D. B. Sanders, H. A. Smith, M. Spaans, L. Spinoglio, G. J. Stacey, I. Veenendaal, S. Veilleux , et al. (5 additional authors not shown)

Abstract: (Ultra) Luminous Infrared Galaxies ((U)LIRGs) are objects characterized by their extreme infrared (8-1000 $μ$m) luminosities ($L_{LIRG}>10^{11} $L$_\odot$ and $L_{ULIRG}>10^{12}$ L$_\odot$). The Herschel Comprehensive ULIRG Emission Survey (HerCULES; PI van der Werf) presents a representative flux-limited sample of 29 (U)LIRGs that spans the full luminosity range of these objects (10… ▽ More (Ultra) Luminous Infrared Galaxies ((U)LIRGs) are objects characterized by their extreme infrared (8-1000 $μ$m) luminosities ($L_{LIRG}>10^{11} $L$_\odot$ and $L_{ULIRG}>10^{12}$ L$_\odot$). The Herschel Comprehensive ULIRG Emission Survey (HerCULES; PI van der Werf) presents a representative flux-limited sample of 29 (U)LIRGs that spans the full luminosity range of these objects (10$^{11}\leq L_\odot \geq10^{13}$). With the \emph{Herschel Space Observatory}, we observe [CII] 157 $μ$m, [OI] 63 $μ$m, and [OI] 145 $μ$m line emission with PACS, CO J=4-3 through J=13-12, [CI] 370 $μ$m, and [CI] 609 $μ$m with SPIRE, and low-J CO transitions with ground-based telescopes. The CO ladders of the sample are separated into three classes based on their excitation level. In 13 of the galaxies, the [OI] 63 $μ$m emission line is self absorbed. Comparing the CO excitation to the IRAS 60/100 $μ$m ratio and to far infrared luminosity, we find that the CO excitation is more correlated to the far infrared colors. We present cooling budgets for the galaxies and find fine-structure line flux deficits in the [CII], [SiII], [OI], and [CI] lines in the objects with the highest far IR fluxes, but do not observe this for CO $4\leq J_{upp}\leq13$. In order to study the heating of the molecular gas, we present a combination of three diagnostic quantities to help determine the dominant heating source. Using the CO excitation, the CO J=1-0 linewidth, and the AGN contribution, we conclude that galaxies with large CO linewidths always have high-excitation CO ladders, and often low AGN contributions, suggesting that mechanical heating is important. △ Less

Submitted 13 January, 2015; originally announced January 2015.

Cloud Usage Patterns: A Formalism for Description of Cloud Usage Scenarios

Authors: Aleksandar Milenkoski, Alexandru Iosup, Samuel Kounev, Kai Sachs, Piotr Rygielski, Jason Ding, Walfredo Cirne, Florian Rosenberg

Abstract: Cloud computing is becoming an increasingly lucrative branch of the existing information and communication technologies (ICT). Enabling a debate about cloud usage scenarios can help with attracting new customers, sharing best-practices, and designing new cloud services. In contrast to previous approaches, which have attempted mainly to formalize the common service delivery models (i.e., Infrastruc… ▽ More Cloud computing is becoming an increasingly lucrative branch of the existing information and communication technologies (ICT). Enabling a debate about cloud usage scenarios can help with attracting new customers, sharing best-practices, and designing new cloud services. In contrast to previous approaches, which have attempted mainly to formalize the common service delivery models (i.e., Infrastructure-as-a-Service, Platform-as-a-Service, and Software-as-a-Service), in this work, we propose a formalism for describing common cloud usage scenarios referred to as cloud usage patterns. Our formalism takes a structuralist approach allowing decomposition of a cloud usage scenario into elements corresponding to the common cloud service delivery models. Furthermore, our formalism considers several cloud usage patterns that have recently emerged, such as hybrid services and value chains in which mediators are involved, also referred to as value chains with mediators. We propose a simple yet expressive textual and visual language for our formalism, and we show how it can be used in practice for describing a variety of real-world cloud usage scenarios. The scenarios for which we demonstrate our formalism include resource provisioning of global providers of infrastructure and/or platform resources, online social networking services, user-data processing services, online customer and ticketing services, online asset management and banking applications, CRM (Customer Relationship Management) applications, and online social gaming applications. △ Less

Submitted 5 October, 2014; originally announced October 2014.

Comments: SPEC (Standard Performance Evaluation Corporation) Research Group --- Cloud Working Group

Report number: SPEC-RG-2013-001

Molecular gas heating in Arp 299

Authors: M. J. F. Rosenberg, R. Meijerink, F. P. Israel, P. P. van der Werf, E. M. Xilouris, A. Weiß

Abstract: Understanding the heating and cooling mechanisms in nearby (Ultra) luminous infrared galaxies can give us insight into the driving mechanisms in their more distant counterparts. Molecular emission lines play a crucial role in cooling excited gas, and recently, with Herschel Space Observatory we have been able to observe the rich molecular spectrum. CO is the most abundant and one of the brightest… ▽ More Understanding the heating and cooling mechanisms in nearby (Ultra) luminous infrared galaxies can give us insight into the driving mechanisms in their more distant counterparts. Molecular emission lines play a crucial role in cooling excited gas, and recently, with Herschel Space Observatory we have been able to observe the rich molecular spectrum. CO is the most abundant and one of the brightest molecules in the Herschel wavelength range. CO transitions are observed with Herschel, and together, these lines trace the excitation of CO. We study Arp 299, a colliding galaxy group, with one component harboring an AGN and two more undergoing intense star formation. For Arp 299 A, we present PACS spectrometer observations of high-J CO lines up to J=20-19 and JCMT observations of $^{13}$CO and HCN to discern between UV heating and alternative heating mechanisms. There is an immediately noticeable difference in the spectra of Arp 299 A and Arp 299 B+C, with source A having brighter high-J CO transitions. This is reflected in their respective spectral energy line distributions. We find that photon-dominated regions (PDRs) are unlikely to heat all the gas since a very extreme PDR is necessary to fit the high-J CO lines. In addition, this extreme PDR does not fit the HCN observations, and the dust spectral energy distribution shows that there is not enough hot dust to match the amount expected from such an extreme PDR. Therefore, we determine that the high-J CO and HCN transitions are heated by an additional mechanism, namely cosmic ray heating, mechanical heating, or X-ray heating. We find that mechanical heating, in combination with UV heating, is the only mechanism that fits all molecular transitions. We also constrain the molecular gas mass of Arp 299 A to 3e9 Msun and find that we need 4% of the total heating to be mechanical heating, with the rest UV heating. △ Less

Submitted 8 July, 2014; originally announced July 2014.

Journal ref: A&A 568, A90 (2014)

Random mixtures of polycyclic aromatic hydrocarbon spectra match interstellar infrared emission

Authors: Marissa J. F. Rosenberg, Olivier Berné, Christiaan Boersma

Abstract: The mid-infrared (IR; 5-15~$μ$m) spectrum of a wide variety of astronomical objects exhibits a set of broad emission features at 6.2, 7.7, 8.6, 11.3 and 12.7 $μ$m. About 30 years ago it was proposed that these signatures are due to emission from a family of UV heated nanometer-sized carbonaceous molecules known as polycyclic aromatic hydrocarbons (PAHs), causing them to be referred to as aromatic… ▽ More The mid-infrared (IR; 5-15~$μ$m) spectrum of a wide variety of astronomical objects exhibits a set of broad emission features at 6.2, 7.7, 8.6, 11.3 and 12.7 $μ$m. About 30 years ago it was proposed that these signatures are due to emission from a family of UV heated nanometer-sized carbonaceous molecules known as polycyclic aromatic hydrocarbons (PAHs), causing them to be referred to as aromatic IR bands (AIBs). Today, the acceptance of the PAH model is far from settled, as the identification of a single PAH in space has not yet been successful and physically relevant theoretical models involving ``true'' PAH cross sections do not reproduce the AIBs in detail. In this paper, we use the NASA Ames PAH IR Spectroscopic Database, which contains over 500 quantum-computed spectra, in conjunction with a simple emission model, to show that the spectrum produced by any random mixture of at least 30 PAHs converges to the same 'kernel'-spectrum. This kernel-spectrum captures the essence of the PAH emission spectrum and is highly correlated with observations of AIBs, strongly supporting PAHs as their source. Also, the fact that a large number of molecules are required implies that spectroscopic signatures of the individual PAHs contributing to the AIBs spanning the visible, near-infrared, and far infrared spectral regions are weak, explaining why they have not yet been detected. An improved effort, joining laboratory, theoretical, and observational studies of the PAH emission process, will support the use of PAH features as a probe of physical and chemical conditions in the nearby and distant Universe. △ Less

Submitted 21 May, 2014; originally announced May 2014.

Radiative and mechanical feedback into the molecular gas of NGC 253

Authors: M. J. F. Rosenberg, M. V. Kazandjian, P. P. van der Werf, F. P. Israel, R. Meijerink, A. Weiß, M. A. Requena-Torres, R. Güsten

Abstract: Starburst galaxies are undergoing intense periods of star formation. Understanding the heating and cooling mechanisms in these galaxies can give us insight to the driving mechanisms that fuel the starburst. Molecular emission lines play a crucial role in the cooling of the excited gas. With SPIRE on the Herschel Space Observatory we have observed the rich molecular spectrum towards the central reg… ▽ More Starburst galaxies are undergoing intense periods of star formation. Understanding the heating and cooling mechanisms in these galaxies can give us insight to the driving mechanisms that fuel the starburst. Molecular emission lines play a crucial role in the cooling of the excited gas. With SPIRE on the Herschel Space Observatory we have observed the rich molecular spectrum towards the central region of NGC 253. CO transitions from J=4-3 to 13-12 are observed and together with low-J line fluxes from ground based observations, these lines trace the excitation of CO. By studying the CO excitation ladder and comparing the intensities to models, we investigate whether the gas is excited by UV radiation, X-rays, cosmic rays, or turbulent heating. Comparing the $^{12}$CO and $^{13}$CO observations to large velocity gradient models and PDR models we find three main ISM phases. We estimate the density, temperature,and masses of these ISM phases. By adding $^{13}$CO, HCN, and HNC line intensities, we are able to constrain these degeneracies and determine the heating sources. The first ISM phase responsible for the low-J CO lines is excited by PDRs, but the second and third phases, responsible for the mid to high-J CO transitions, require an additional heating source. We find three possible combinations of models that can reproduce our observed molecular emission. Although we cannot determine which of these are preferable, we can conclude that mechanical heating is necessary to reproduce the observed molecular emission and cosmic ray heating is a negligible heating source. We then estimate the mass of each ISM phase; $6\times 10^7$ M$_\odot$ for phase 1 (low-J CO lines), $3\times 10^7$ M$_\odot$ for phase 2 (mid-J CO lines), and $9\times 10^6$ M$_\odot$ for phase 3 (high-J CO lines) for a total system mass of $1\times10^{8}$ M$_\odot$. △ Less

Submitted 20 January, 2014; originally announced January 2014.

The excitation of near-infrared H2 emission in NGC 253

Authors: M. J. F. Rosenberg, P. P. van der Werf, F. P. Israel

Abstract: Because of its large angular size and proximity to the Milky Way, NGC 253, an archetypal starburst galaxy, provides an excellent laboratory to study the intricacies of this intense episode of star formation. We aim to characterize the excitation mechanisms driving the emission in NGC 253. Specifically we aim to distinguish between shock excitation and UV excitation as the dominant driving mechanis… ▽ More Because of its large angular size and proximity to the Milky Way, NGC 253, an archetypal starburst galaxy, provides an excellent laboratory to study the intricacies of this intense episode of star formation. We aim to characterize the excitation mechanisms driving the emission in NGC 253. Specifically we aim to distinguish between shock excitation and UV excitation as the dominant driving mechanism, using Brγ, H_2 and [FeII] as diagnostic emission line tracers. Using SINFONI observations, we create linemaps of Brγ, [FeII]_{1.64}, and all detected H_2 transitions. By using symmetry arguments of the gas and stellar gas velocity field, we find a kinematic center in agreement with previous determinations. The ratio of the 2-1 S(1) to 1-0 S(1) H_2 transitions can be used as a diagnostic to discriminate between shock and fluorescent excitation. Using the 1-0 S(1)/2-1 S(1) line ratio as well as several other H_2 line ratios and the morphological comparison between H_2 and Brγand [FeII], we find that excitation from UV photons is the dominant excitation mechanisms throughout NGC 253. We employ a diagnostic energy level diagram to quantitatively differentiate between mechanisms. We compare the observed energy level diagrams to PDR and shock models and find that in most regions and over the galaxy as a whole, fluorescent excitation is the dominant mechanism exciting the H_2 gas. We also place an upper limit of the percentage of shock excited H_2 at 29%. We find that UV radiation is the dominant excitation mechanism for the H_2 emission. The H_2 emission does not correlate well with Brγbut closely traces the PAH emission, showing that not only is H_2 fluorescently excited, but it is predominately excited by slightly lower mass stars than O stars which excite Brγ, such as B stars. △ Less

Submitted 7 December, 2012; originally announced December 2012.

[FeII] as a tracer of Supernova Rate in Near-by Starburst Galaxies

Authors: M. J. F. Rosenberg, P. P. van der Werf, F. P. Israel

Abstract: Supernovae play an integral role in the feedback of processed material into the ISM of galaxies and are responsible for much of the chemical enrichment of the universe. The rate of supernovae can also reveal the star formation histories. Supernova rates are usually measured through the non-thermal radio continuum luminosity; however, a correlation between near-infrared [FeII] emission and supernov… ▽ More Supernovae play an integral role in the feedback of processed material into the ISM of galaxies and are responsible for much of the chemical enrichment of the universe. The rate of supernovae can also reveal the star formation histories. Supernova rates are usually measured through the non-thermal radio continuum luminosity; however, a correlation between near-infrared [FeII] emission and supernova remnants has also been noted. We aim to find a quantitative relationship between the [FeII] at 1.26 um ([FeII]$_{1.26}$) luminosity and supernova rate in a sample of 11 near-by starburst galaxy centers. We perform a pixel-pixel analysis of this correlation on SINFONI data cubes. Using Br$γ$ equivalent width and luminosity as the only observational inputs into the Starburst 99 model, we derive the supernova rate at each pixel and thus create maps of supernova rates. We then compare these morphologically and quantitatively to the [FeII]$_{1.26}$ luminosity. We have found that a strong linear and morphological correlation exists between supernova rate and [FeII]$_{1.26}$ on a pixel-pixel basis: \[ log\frac{ν_{SNrate}}{yr^{-1}pc^{-2}} = 1.01 \pm 0.2\ast log\frac{[FeII]_{1.26}}{erg s^{-1}pc^{-2}} - 41.17 \pm 0.9\] This relation is valid for normal star forming galaxies but breaks down for extreme ultra luminous galaxies. The supernova rates derived from the Starburst 99 model are in good agreement with the radio-derived supernova rates, which underlines the strength of using [FeII] emission as a tracer of supernova rate. With the strong correlation found in this sample of galaxies, we conclude that [FeII]$_{1.26}$ emission can be generally used to derive accurate supernova rates on either a pixel-pixel or integrated galactic basis. △ Less

Submitted 13 February, 2012; originally announced February 2012.

[FeII] as a tracer supernova rate

Authors: M. J. F. Rosenberg, P. P. van der Werf, F. P. Israel

Abstract: Supernovae play an integral role in the feedback of processed material into the ISMof galaxies and are responsible for most of the chemical enrichment of the universe. The rate of supernovae can also reveal the star formation histories. Supernova rate is usually measured through the non-thermal radio continuum luminosity, but in this paper we establish a quantitative relationship between the [FeII… ▽ More Supernovae play an integral role in the feedback of processed material into the ISMof galaxies and are responsible for most of the chemical enrichment of the universe. The rate of supernovae can also reveal the star formation histories. Supernova rate is usually measured through the non-thermal radio continuum luminosity, but in this paper we establish a quantitative relationship between the [FeII]1.26 luminosity and supernova rate in a sample of 11 near-by starburst galaxies. SINFONI data cubes are used to perform a pixel pixel analysis of this correlation. Using Br equivalent width and luminosity as the only observational inputs into Starburst 99, the supernova rate is derived at each pixel and a map of supernova rate is created. This is then compared morphologically and quantitatively to [FeII]1.26 luminosity map. We find a strong linear and morphological correlation between supernova rate and [FeII]1.26 on a pixel-pixel basis: log SNrate yr-1 pc-2 = (1.01 \pm 0.2) \ast log[FeII]1.26 ergs-1 pc-2 - 41.17 \pm 0.9 The Starburst 99 derived supernova rates are also in good agreement with the radio derived supernova rates, which further demonstrates the strength of [FeII] as a tracer of supernova rate. With the strong correlation found in this sample of galaxies, we now qualitatively use [FeII]1.26 to derive supernova rate on either a pixel-pixel or integrated galactic basis. △ Less

Submitted 16 January, 2012; originally announced January 2012.

Comments: Conference proceedings for Astronomy at High Angular Resolution 2011: The Central Kiloparsec in Galactic Nuclei

Coupled Blind Signal Separation and Spectroscopic Database Fitting of the Mid Infrared PAH Features

Authors: M. J. F. Rosenberg, O. Berné, C. Boersma, L. J. Allamandola, A. G. G. M Tielens

Abstract: The aromatic infrared bands (AIBs) observed in the mid infrared spectrum are attributed to Polycyclic Aromatic Hydrocarbons (PAHs). We observe the NGC 7023-North West (NW) PDR in the mid-infrared (10 - 19.5 micron) using the Infrared Spectrometer (IRS), on board Spitzer. Clear variations are observed in the spectra, most notably the ratio of the 11.0 to 11.2 micron bands, the peak position of the… ▽ More The aromatic infrared bands (AIBs) observed in the mid infrared spectrum are attributed to Polycyclic Aromatic Hydrocarbons (PAHs). We observe the NGC 7023-North West (NW) PDR in the mid-infrared (10 - 19.5 micron) using the Infrared Spectrometer (IRS), on board Spitzer. Clear variations are observed in the spectra, most notably the ratio of the 11.0 to 11.2 micron bands, the peak position of the 11.2 and 12.0 micron bands, and the degree of asymmetry of the 11.2 micron band. The observed variations appear to change as a function of position within the PDR. We aim to explain these variations by a change in the abundances of the emitting components of the PDR. A Blind Signal Separation (BSS) method, i.e. a Non-Negative Matrix Factorization algorithm is applied to separate the observed spectrum into components. Using the NASA Ames PAH IR Spectroscopic Database, these extracted signals are fit. The observed signals alone were also fit using the database and these components are compared to the BSS components. Three component signals were extracted from the observation using BSS. We attribute the three signals to ionized PAHs, neutral PAHs, and Very Small Grains (VSGs). The fit of the BSS extracted spectra with the PAH database further confirms the attribution to ionized and neutral PAHs and provides confidence in both methods for producing reliable results. The 11.0 micron feature is attributed to PAH cations while the 11.2 micron band is attributed to neutral PAHs. The VSG signal shows a characteristically asymmetric broad feature at 11.3 micron with an extended red wing. By combining the NASA Ames PAH IR Spectroscopic Database fit with the BSS method, the independent results of each method can be confirmed and some limitations of each method are overcome. △ Less

Submitted 29 June, 2011; originally announced June 2011.