Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Cluster detection”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,513 records · Page 84
All sourcesNASA NTRSDOE OSTI
NASA NTRS2012

Obtaining Accurate Change Detection Results from High-Resolution Satellite Sensors

Multi-date acquisitions of high-resolution imaging satellites (e.g. GeoEye and WorldView), can display local changes of current economic interest. However, their large data volume precludes effective manual analysis, requiring image co-registration followed by image-to-image change detection, preferably with minimal analyst attention. We have recently developed an automatic change detection procedure that minimizes false-positives. The processing steps include: (a) Conversion of both the pre- and post- images to reflectance values (this step is of critical importance when different sensors are involved); reflectance values can be either top-of-atmosphere units or have full aerosol optical depth calibration applied using bi-directional reflectance knowledge. (b) Panchromatic band image-to-image co-registration, using an orthorectified base reference image (e.g. Digital Orthophoto Quadrangle) and a digital elevation model; this step can be improved if a stereo-pair of images have been acquired on one of the image dates. (c) Pan-sharpening of the multispectral data to assure recognition of change objects at the highest resolution. (d) Characterization of multispectral data in the post-image ( i.e. the background) using unsupervised cluster analysis. (e) Band ratio selection in the post-image to separate surface materials of interest from the background. (f) Preparing a pre-to-post change image. (g) Identifying locations where change has occurred involving materials of interest.

multispectral↗
NASA NTRS1974

Progress in gamma ray burst astronomy

Spectral observations on nine of the eleven known cosmic gamma ray bursts detected from IMP-7 from late 1972 through 1973 are reported. All nine events were found to have average spectra ranging from 100 to 1100 keV. The hypothesis of a single greater than 100 keV average event spectrum for all gamma ray bursts (inferred from IMP-6 observations) was investigated. In addition to the mutual consistency of these 14 IMP events, all independently reported average event spectra fit the IMP-7 common spectrum. This observed lack of variability of the average event spectrum suggests that the production mechanism is tied to more rigidly fixed source parameters than most celestial or solar hard X-ray phenomena. Other characteristics of gamma ray bursts events, are discussed: a high-energy spectral tail with power-law index -2.5, a rising size spectrum from the earlier event rate, and a hint of clustering of event occurrence in time.

Cline, T. L.↗
NASA NTRS1995

Spectroscopic detection of the secondaries of the Hyades interferometric spectroscopic binary theta(sup 2) Tauri and of the interferometric spectroscopic binary alpha Andromedae

We report new high-resolution, wide-wavelength-coverage CCD observations of the visual and near-infrared spectra of the interferometric spectroscopic binaries theta(sup 2) Tau (primary spectral type A 7 III) and alpha And (primary spectral type B8 IVpMnHg), which allow the first direct spectroscopic detection of their secondaries. We have measured primary and secondary radial velocities and used them to redetermine the spectroscopic orbits of the primaries, make an improved determination of K(sub 2) for theta(sup 2) Tau, and measure K(sub 2) for alpha And for the first time. The spectroscopic orbits combined with the interferometrically measured visual orbits of theta(sup 2) Tau and alpha And provide masses and distances, M(sub 1) = 2.1 +/- 0.3 solar mass, M(sub 2) = 1.6 +/- 0.2 solar mass, and d = 44.1 +/- 2.2 pc for theta(sup 2) Tau and M(sub 1) = 5.5 +/- 0.5 solar mass, M(sub 2) = 2.3 +/- 0.2 solar mass, and d = 34.0 +/- 1.3 pc for alpha And. The distance to theta(sup 2) Tau and a foreground correction of 0.07 +/- 0.05 mag set the distance to the center of the Hyades cluster at 45.5 +/- 2.5 pc, or m - M = 3.29 +/- 0.12. Indirect estimates of the mass of theta(sup 2) Tau's primary, made via the mass-luminosity relation and via the Hyades cluster turnoff mass, are consistent with these dynamical masses for its primary and secondary. But in the case of alpha And, the mass-luminosity relation suggests that our dynamical masses for its primary and secondary may be too large. For alpha And somewhat smaller dynamical masses (4.0 and 1.9 solar mass) and the correspondingly smaller distance (31 pc), required by the corresponding larger orbital parallax, provide results consistent with the mass-luminosity relation for normal stars.

Tomkin, Jocelyn↗
NASA NTRS2019

X-ray Binary Luminosity Function Scaling Relations for Local Galaxies Based on Subgalactic Modeling

We present new Chandra constraints on the X-ray luminosity functions(XLFs)of X-ray binary(XRB)populations, as well as their scaling relations, for a sample of 38 nearby galaxies(D = 3.4–29 Mpc). Our galaxy sample is drawn primarily from the Spitzer Infrared Nearby Galaxies Survey(SINGS)and contains a wealth of Chandra(5.8 Ms total)and multiwavelength data, allowing for star formation rates(SFRs)and stellar masses(M⁎)to be measured on subgalactic scales. We divided the 2478 X-ray-detected sources into 21 subsamples in bins of specific SFR(sSFR ≡ SFR/M⁎)and constructed XLFs. To model the XLF dependence on sSFR, we fit a global XLF model, containing contributions from high-mass XRBs(HMXBs), low-mass XRBs(LMXBs), and background sources from the cosmic X-ray background that respectively scale with SFR, M⁎, and sky area. We find an HMXB XLF that is more complex in shape than previously reported and an LMXB XLF that likely varies with sSFR, potentially due to an age dependence. When applying our global model to XLF data for each individual galaxy, we discover a few galaxy XLFs that significantly deviate from our model beyond statistical scatter. Most notably, relatively low-metallicity galaxies have an excess of HMXBs above≈10^(38)erg/s, and elliptical galaxies that have relatively rich populations of globular clusters(GCs)show excesses of LMXBs compared to the global model. Additional modeling of how the XRB XLF depends on stellar age, metallicity, and GC specific frequency is required to sufficiently characterize the XLFs of galaxies.

Bret D. Lehmer↗
DOE OSTI2022

A deep learning approach to real-time HIV outbreak detection using genetic data

Pathogen genomic sequence data are increasingly made available for epidemiological monitoring. A main interest is to identify and assess the potential of infectious disease outbreaks. While popular methods to analyze sequence data often involve phylogenetic tree inference, they are vulnerable to errors from recombination and impose a high computational cost, making it difficult to obtain real-time results when the number of sequences is in or above the thousands. Here, we propose an alternative strategy to outbreak detection using genomic data based on deep learning methods developed for image classification. The key idea is to use a pairwise genetic distance matrix calculated from viral sequences as an image, and develop convolutional neutral network (CNN) models to classify areas of the images that show signatures of active outbreak, leading to identification of subsets of sequences taken from an active outbreak. We showed that our method is efficient in finding HIV-1 outbreaks with R0 ≥ 2.5, and overall a specificity exceeding 98% and sensitivity better than 92%. We validated our approach using data from HIV-1 CRF01 in Europe, containing both endemic sequences and a well-known dual outbreak in intravenous drug users. Our model accurately identified known outbreak sequences in the background of slower spreading HIV. Importantly, we detected both outbreaks early on, before they were over, implying that had this method been applied in real-time as data became available, one would have been able to intervene and possibly prevent the extent of these outbreaks. This approach is scalable to processing hundreds of thousands of sequences, making it useful for current and future real-time epidemiological investigations, including public health monitoring using large databases and especially for rapid outbreak identification.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

The inheritance of anthracnose (Colletotrichum sublineola) resistance in sorghum differential lines QL3 and IS18760

Anthracnose caused by the fungal pathogen C. sublineola is an economically important constraint on worldwide sorghum production. The most effective strategy to safeguard yield is through the introgression of resistance alleles. This requires elucidation of the genetic basis of the different resistance sources that have been identified. In this study, 223 recombinant inbred lines (RILs) derived from crossing anthracnose-differentials QL3 (96 RILs) and IS18760 (127 RILs) with the common susceptible parent PI609251 were evaluated at four field locations in the United States (Florida, Georgia, Texas, and Puerto Rico) for their anthracnose resistance response. Both RIL populations were highly susceptible to anthracnose in Florida and Georgia, while in Puerto Rico and Texas they were segregating for anthracnose resistance response. A genome scan using a composite linkage map of 982 single nucleotide polymorphisms (SNPs) detected two genomic regions of 4.31 and 0.85 Mb on chromosomes 4 and 8, respectively, that explained 10–27% of the phenotypic variation in Texas and Puerto Rico. In parallel, a subset of 43 RILs that contained 67% of the recombination events were evaluated against anthracnose pathotypes from Arkansas (2), Puerto Rico (2) and Texas (4) in the greenhouse. A genome scan showed that the 7.57 Mb region at the distal end of the short arm of chromosome 5 is associated with the resistance response against the pathotype AMP-048 from Arkansas. Comparative analysis identified the genomic region on chromosome 4 overlaps with an anthracnose resistance locus identified in another anthracnose-differential line, SC414-12E, indicating this genomic region is of interest for introgression in susceptible sorghum germplasm. Candidate gene analysis for the resistance locus on chromosome 5 identified an R -gene cluster that has high similarity to another R -gene cluster associated with anthracnose resistance on chromosome 9.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS2017

The Ultracompact Nature of the Black Hole Candidate X-Ray Binary 47 Tuc X9

47 Tuc X9 is a low-mass X-ray binary (LMXB) in the globular cluster 47 Tucanae, and was previously thought to be a cataclysmic variable. However, Miller-Jones et al. recently identified a radio counterpart to X9 (inferring a radio X-ray luminosity ratio consistent with black hole LMXBs), and suggested that the donor star might be a white dwarf. We report simultaneous observations of X9 performed by Chandra, NuSTAR and Australia Telescope Compact Array. We find a clear 28.18+/- 0.02-min periodic modulation in the Chandra data, which we identify as the orbital period, confirming this system as an ultracompact X-ray binary. Our X-ray spectral fitting provides evidence for photoionized gas having a high oxygen abundance in this system, which indicates a CO white dwarf donor. We also identify reflection features in the hard X-ray spectrum, making X9 the faintest LMXB to show X-ray reflection. We detect an approx. 6.8-d modulation in the X-ray brightness by a factor of 10, in archival Chandra, Swift and ROSAT data. The simultaneous radio X-ray flux ratio is consistent with either a black hole primary or a neutron star primary, if the neutron star is a transitional millisecond pulsar. Considering the measured orbital period (with other evidence of a white dwarf donor), and the lack of transitional millisecond pulsar features in the X-ray light curve, we suggest that this could be the first ultracompact black hole X-ray binary identified in our Galaxy.

accretion↗
DOE OSTI2020

Unified architecture for data-driven metadata tagging of building automation systems

This article presents a Unified Architecture (UA) for automated point tagging of Building Automation System (BAS) data, based on a combination of data-driven approaches. Advanced energy analytics applications—including fault detection and diagnostics and supervisory control—have emerged as a significant opportunity for improving the performance of our built environment. Effective application of these analytics depends on harnessing structured data from the various building control and monitoring systems, but typical BAS implementations do not employ any standardized metadata schema. While standards such as Project Haystack and Brick Schema have been developed to address this issue, the process of structuring the data, i.e., tagging the points to apply a standard metadata schema, has, to date, been a manual process. This process is typically costly, labor-intensive, and error-prone. In this work we address this gap by proposing a UA that automates the process of point tagging by leveraging the data accessible through connection to the BAS, including time-series data and the raw point names. The UA intertwines supervised classification and unsupervised clustering techniques from machine learning and leverages both their deterministic and probabilistic outputs to inform the point tagging process. Furthermore, we extend the UA to embed additional input and output data-processing modules that are designed to address the challenges associated with the real-time deployment of this automation solution. We test the UA on two datasets for real-life buildings: (i) commercial retail buildings and (ii) office buildings from the National Renewable Energy Laboratory (NREL) campus. We report the proposed methodology correctly applied 85–90% and 70–75% of the tags in each of these test scenarios, respectively for two significantly different building types used for testing UA's fully-functional prototype. The proposed UA, therefore, offers promising approach for automatically tagging BAS data as it reaches close to 90% accuracy. Further building upon this framework to algorithmically identify the equipment type and their relationships is an apt future research direction to pursue.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRS2011

Parametric Testing of Launch Vehicle FDDR Models

For the safe operation of a complex system like a (manned) launch vehicle, real-time information about the state of the system and potential faults is extremely important. The on-board FDDR (Failure Detection, Diagnostics, and Response) system is a software system to detect and identify failures, provide real-time diagnostics, and to initiate fault recovery and mitigation. The ERIS (Evaluation of Rocket Integrated Subsystems) failure simulation is a unified Matlab/Simulink model of the Ares I Launch Vehicle with modular, hierarchical subsystems and components. With this model, the nominal flight performance characteristics can be studied. Additionally, failures can be injected to see their effects on vehicle state and on vehicle behavior. A comprehensive test and analysis of such a complicated model is virtually impossible. In this paper, we will describe, how parametric testing (PT) can be used to support testing and analysis of the ERIS failure simulation. PT uses a combination of Monte Carlo techniques with n-factor combinatorial exploration to generate a small, yet comprehensive set of parameters for the test runs. For the analysis of the high-dimensional simulation data, we are using multivariate clustering to automatically find structure in this high-dimensional data space. Our tools can generate detailed HTML reports that facilitate the analysis.

Schumann, Johann↗
DOE OSTI2021

Evidence of a Dwarf Galaxy Stream Populating the Inner Milky Way Halo

Stellar streams produced from dwarf galaxies provide direct evidence of the hierarchical formation of the Milky Way. Here, we present the first comprehensive study of the LMS-1 stellar stream, that we detect by searching for wide streams in the Gaia EDR3 data set using the STREAMFINDER algorithm. This stream was recently discovered by Yuan et al. We detect LMS-1 as a 60° long stream to the north of the galactic bulge, at a distance of ~20 kpc from the Sun, together with additional components that suggest that the overall stream is completely wrapped around the inner Galaxy. Using spectroscopic measurements from the Large Sky Area Multi-Object Fiber Spectroscopic Telescope, the Sloan Digital Sky Survey, and the Apache Point Observatory Galactic Evolution Experiment, we infer that the stream is very metal-poor (<[Fe/H]> = –2.1) with a significant metallicity dispersion (σ [Fe/H] = 0.4), and it possesses a large radial velocity dispersion (σ v = 20 ± 4 km s–1 ). These estimates together imply that LMS-1 is a dwarf galaxy stream. The orbit of LMS-1 is close to polar, with an inclination of 75° to the galactic plane. Both the orbit and metallicity of LMS-1 are remarkably similar to the globular clusters NGC 5053, NGC 5024, and the stellar stream Indus. As a result, these findings make LMS-1 an important contributor to the stellar population of the inner Milky Way halo.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1997

A 2000 Solar Mass Rotating Molecular Disk Around NGC 6334A

We present millimeter and centimeter wave spectroscopic observations of the H II region NGC 6334A. We have mapped the source in several transitions of CO, CS, and NH3. The molecular emission shows a distinct flattened structure in the east-west direction. This structure is probably a thick molecular disk or torus (2.2 x 0.9 pc) responsible for the bipolarity of the near-infrared (NIR) and radio continuum emission which extends in two "lobes" to the north and south of the shell-like H II region. The molecular disk is rotating from west to east (omega approximately equals 2.4 km/s.pc) about an axis approximately parallel to the radio and NIR emission lobes. By assuming virial equilibrium, we find that the molecular disk contains approximately 2000 solar mass. Single-component gas excitation model calculations show that the molecular gas in the disk is warmer and denser (T(sub k) approximately equals 60 K, n approximately equals 3000/cc) than the gas to the north and south (T(sub k) approximately equals 50 K, n approximately equals 400/cc). High resolution (approximately 5 sec) NH3 (3, 3) images of NGC 6334A reveal several small (approximately 0.1 pc) clumps, one of which lies southwest of the radio continuum shell, and is spatially coincident with a near-infrared source, IRS 20. A second NH3 clump is coincident with an H2O maser and the center of a molecular outflow. The dense gas tracers, CS J = 5 approaches 4 and 7 approaches 6, peak near IRS 20 and the H2O maser, not at NGC 6334A. IRS 20 has a substantial far-infrared (FIR) luminosity L(sub FIR) approximately 10(exp 5) solar luminosity, which indicates the presence of an O 7.5 star but has no detected radio continuum (F(sub 6 cm) < 0.02 Jy). The combination of dense gas, a large FIR luminosity and a lack of radio continuum can best be explained if IRS 20 is a protostar. A third clump of NH3 emission lies to the west of IRS 20 but is not associated with any other molecular or continuum features. The star formation activity in the region has moved west of NGC 6334A to IRS 20 and the H2O maser position. We suggest that NGC 6334A, IRS 20, and the H2O maser spot are part of a "protocluster" of stars which is condensing from the massive molecular disk. The similarity between the structure around NGC 6334A and other large (r approximately 1 pc), massive (M approximately 10(exp 3) solar mass), rotating disks (K3-50A and G10.6-0.4) suggests that this may be a common mechanism by which open clusters form.

Kraemer, Kathleen E.↗
DOE OSTI2022

Experimental study of the Si 30 ( He 3 , d ) P 31 reaction and thermonuclear reaction rate of Si 30 ( p , γ ) P 31

Background: Abundance anomalies in some globular clusters, such as the enhancement of potassium and the depletion of magnesium, can be explained in terms of an earlier generation of stars polluting the presently observed ones. It was shown that the potential range of temperatures and densities of the polluting sites depends on the strength of a few critical reaction rates. The 30 Si(p,γ) 31 P reaction has been identified as one of these important reactions. Purpose: The key ingredient for evaluating the thermonuclear 30 Si(p,γ) 31 P reaction rate is the strength of the resonances which, at low energy, are proportional to their proton width. Therefore, the goal of this work is to determine the proton widths of unbound 31 P states. Method: States in 31 P were studied at the Maier-Leibnitz-Laboratorium using the one-proton 30 Si( 3 He,d) 31 P transfer reaction. Deuterons were detected with the high resolution one quadrupole three dipoles (Q3D) magnetic spectrometer. Angular distribution and spectroscopic factors were extracted for 27 states, and proton widths and resonance strengths were calculated for the unbound states. Results: Several 31 P unbound states have been observed for the first time in a one-proton transfer reaction. Above 20 MK, the 30 Si(p,γ) 31 P reaction rate is now entirely estimated from the observed properties of 31 P states. The reaction rate uncertainty from all resonances other than the E$^{c.m.}_{r}$ = 149 keV resonance has been reduced down to less than a factor of 2 above that temperature. The unknown spin and parity of the E$^{c.m.}_{r}$ = 149 keV resonance dominates the uncertainty in the rate in the relevant temperature range. Conclusion: Here, the remaining source of uncertainty on the 30 Si(p,γ) 31 P reaction rate comes from the unknown spin and parity of the E$^{c.m.}_{r}$ = 149 keV resonance which can change the reaction rate by a factor of 10 in the temperature range of interest.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2022

Mitigate: An Adaptive Network Data Anonymization Tool Using Condensation-Based Differential Privacy

Modern network devices collect a large amount of data that can be analyzed to identify bottlenecks, anomalies, cyber-attacks, etc. Therefore, there is often a need to analyze such collections of network data quite often by an external expert or by the research community. However, these collections of data contain sensitive, proprietary information. In order for the network data to be shared, it must first be anonymized. The overall objective of this project is to develop an innovative privacy management tool to anonymize network data and achieve sufficient privacy, acceptable data utility, and efficient data analysis at the same time. No existing anonymization methods can achieve all of these at the same time. The core of this technology is a differential private clustering algorithm that provides strong privacy protection, preserves data properties important for subsequent analysis, and allows the party receiving the anonymized data to conduct analysis directly on anonymized data without the need of decryption or any extra processing. The research carried out was to design, implement and verify a solution to this problem by completing the following tasks: 1) developing the core technology; 2) developing a context based method that automatically recommends fields that must be anonymized; 3) conducted experiments showing superior results using our approach compared to existing tools, and 4) developed an intuitive but basic user interface. The research that was conducted generated novel algorithmic techniques that utilize state-of-the-art methods such as condensation, differential privacy preservation, clustering, automated tuning based on contextual awareness, and recommendation techniques to specify columns to users for anonymization leading to optimal privacy that allows research analysis on the dataset. Experiments were conducted to evaluate the efficacy of these novel algorithmic techniques by performing analysis on original non-anonymized datasets, then conducting analysis on the same yet anonymized datasets and comparing the results of the analyses. Overall, the anonymized analysis results were within 1% of the original results, verifying that the generated technology not only guarantees a high level of privacy but also enables research analysis as if it were conducted on the original dataset. Potential applications of this technology include anonymization of any type of structured network datasets that contain sensitive identifiers, such as IP addresses, that can be used in multiple applications. For example, to create an AI or machine learning model for cyber security, e.g., to detect attacks, or for performance analysis, e.g., identify bottlenecks or predict performance. In addition, a market analysis that was conducted for potential applications of this technology identified a broader range of applications of our anonymization technology beyond the network sector that includes healthcare, banking, insurance, securities, finance (FISB), data brokering, cloud services, ad sales, and government.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1997

NGC 5291: Implications for the Formation of Dwarf Galaxies

The possible formation and evolution of dwarf irregular galaxies from material derived from perturbed evolved galaxies is addressed via an H I study of a likely example, the peculiar system NGC 5291. This system, located in the western outskirts of the cluster Abell 3574, contains the lenticular galaxy NGC 5291 which is in close proximity to a disturbed companion and is flanked by an extensive complex of numerous knots extending roughly 4 min north and 4 min south of the galaxy. In an initial optical and radio study, Longmore et al. (1979, MNRAS, 188, 285) showed that these knots have the spectra of vigorous star-forming regions, and suggested that some may in fact be young dwarf irregular galaxies. High resolution 21-cm line observations taken with the VLA are presented here and reveal that the H I distribution associated with this system encompasses not only the entire N-S complex of optical knots, but also forms an incomplete ring or tail that extends approximately 3 min to the west. The H I associated with NGC 5291 itself shows a high velocity range; the Seashell is not detected. The formation mechanism for this unusual system is unclear and two models - a large, low-luminosity ram-swept disk, and a ram-swept interaction-are discussed. The H I in the system contains numerous concentrations, mostly along the N-S arc of the star-forming complexes, which generally coincide with one or more optical knots; the larger H I features contain several x 10(exp 9) solar mass of gas. Each of the knots is compared to a set of criteria designed to determine if these objects are bound against their own internal kinetic energy and are tidally stable relative to the host galaxy. An analysis of the properties of the H I concentrations surrounding the optical star-forming complexes indicates that at least the largest of these is a bound system; it also possesses a stellar component. It is suggested that this object is a genuinely young dwarf irregular galaxy that has evolved from the material associated with the system and that this entire complex contains several proto- or young dwarf irregular galaxies in various stages of development. We are therefore witnessing the early evolution of a number of genuinely young galaxies. Given the evident importance of the NGC 5291 system as a 'nursery' for young galaxies, careful modeling is required if we are to understand this remarkable galaxy.

Malphrus, Benjamin K.↗
DOE OSTI2022

Single atoms and small clusters of atoms may accompany Au and Pd dendrimer-encapsulated nanoparticles

Here we report the presence of small clusters of atoms (<1 nm) (SCs) and single atoms (SAs) in solutions containing 1–2 nm dendrimer-encapsulated nanoparticles (DENs). Au and Pd DENs were imaged using aberration-corrected scanning transmission electron microscopy (ac-STEM), and energy dispersive spectroscopy (EDS) was used to identify and quantify the SAs/SCs. Two main findings have emerged from this work. First, the presence or absence of SAs/SCs depends on both the terminal functional group of the dendrimer (–NH 2 or –OH) and the elemental composition of the DENs (Au or Pd). Second, dialysis can be used to remove the majority of SAs/SCs in cases where a high density of SAs/SCs are present. The foregoing conclusions provide insights into the mechanisms for Au and Pd DEN synthesis and stability. Ultimately, these results demonstrate the need for careful characterization of systems containing nanoparticles to ensure that SAs/SCs, which may be below the detection limit of most analytical methods, are taken into consideration (especially for catalysis experiments).

74 ATOMIC AND MOLECULAR PHYSICS↗
DOE OSTI2024

FORESTR: Finding, Organizing, Representing, Explaining, Summarizing, and Thinning Random forests

Random forests have become popular models used for data driven predictions. As a result, random forests are currently used or being considered for high-consequence mission applications in national security, such as the prediction of yield from optical signals and malware detection. While random forests may provide accurate predictions, the complexity of the algorithm causes a lack of interpretability. Random forests are an ensemble of regression or decision trees. Individual regression and decision trees are interpretable, but ensembles are inherently difficult to interpret due to the compilation of many models. We aim to increase the interpretability of random forests by finding patterns in the ensemble of trees that can be used to “thin” (or remove) trees. As a starting point, in this report, we develop a new distance metric for quantifying the similarity between trees based on their topologies (i.e., shapes). We base the metric on a novel distance metric for graphs that is a proper mathematical distance, is invariant to transformations, has registration between graphs, and computes topological evolutions between graphs. We use the tree distance metric to compute tree statistics such as a “mean tree” and to identify clusters of trees. We apply the developed methodology to a toy dataset and a mission relevant product inspection dataset to demonstrate how the metric can provide insight into random forests. Furthermore, we discuss the limitations of the approach and ideas for future research into how the metric could be used as a thinning tool to develop less complex models.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1993

Ly alpha and IR galaxy companions of high redshift damped Ly alpha QSO absorbers

We have used a Near-Infrared Camera and Multi-Object Spectrometer (NICMOS3) HgCdTe 256x256 array detector with the Infrared (IR) camera on the 2.3m telescope at Steward Observatory to image several Quasi-Stellar Object (QSO) fields. The limiting magnitude is K'(2.1 microns) = 21.0 - 21.5 mag per square arcsec for a 3 sigma detection in 3 hours of in-field chopping observations. Each QSO line-of-sight samples several known absorbers with Mg2(lambda)2796-2803 A and/or C4(lambda)1548-1551 A absorption doublets. The equivalent width distributions of the low and high ionization absorption lines of the absorber sample are identical to those of the parent population of all absorbers. This selection process, used already for a spectroscopic survey of Mg2 absorption lines in C4-selected absorption systems at high z, gives a methodical approach to observing, reduces the observer biases, and makes a more efficient use of telescope time. This selection guarantees that imaging of the sample of QSO fields will provide complete sampling of the whole population of high z QSO absorbers. Follow-up optical and IR spectroscopy of these objects is scheduled for redshift measurement and confirmation of the absorbing galaxies and the cluster members.

Caulet, Adeline↗
NASA NTRS2022

Chile Wildfires: Utilizing NASA and NOAA Earth Observations to Determine Lightning-ignited Wildfire Risks in Central Chile

In recent years, Central Chile has experienced wildfires of increasing frequency and intensity which threaten natural resources and communities. The Corporación Nacional Forestal (CONAF) responds to wildfires caused by a variety of ignitions, including lightning, but it is difficult to determine the prevalence of lightning-ignited wildfires based solely on ground observations. In collaboration with CONAF and the Embassy of Chile, Agricultural Office, the team used Earth observations to map coincidence of lightning strikes and wildfire ignitions. The Active Fire Product of Suomi NPP Visible Infrared Imaging Radiometer Suite (VIIRS) identified wildfires as thermal anomalies, which the team compared to the lightning events detected by NOAA’s GOES-16 Geostationary Lightning Mapper (GLM). Next, the team mapped lightning strike frequency and lightning related wildfires across the study area. Finally, the team calculated and mapped a relative estimate of lightning-ignited wildfire vulnerability across the year, fire season (December – March), and off-season (April – November) by summing the following factors: lightning frequency, the Normalized Difference Moisture Index (NDMI) and land surface temperature (LST). These risks were then weighted by fuel availability. Preliminary analysis of the lightning fire relationship showed a spatiotemporal coincidence, primarily in the South-central region of study, near Temuco, and isolated areas on the Andean front. The team identified areas at risk of lightning-induced wildfires, predominantly in the northern third of the study area and along the Andean front. Adjusting the relative weight of risk factors and improving the lightning and fire coincidence map by clustering VIIRS thermal anomalies into fire events could reduce discrepancies and improve risk assessments for future work.

Christopher Matechik↗