skip to main content

Title: Six Outbursts of Comet 46P/Wirtanen
Cometary activity is a manifestation of sublimation-driven processes at the surface of nuclei. However, cometary outbursts may arise from other processes that are not necessarily driven by volatiles. In order to fully understand nuclear surfaces and their evolution, we must identify the causes of cometary outbursts. In that context, we present a study of mini-outbursts of comet 46P/Wirtanen. Six events are found in our long-term lightcurve of the comet around its perihelion passage in 2018. The apparent strengths range from −0.2 to −1.6 mag in a 5" radius aperture, and correspond to dust masses between ∼104 to 106 kg, but with large uncertainties due to the unknown grain size distributions. However, the nominal mass estimates are the same order of magnitude as the mini-outbursts at comet 9P/Tempel 1 and 67P/Churyumov-Gerasimenko, events which were notably lacking at comet 103P/Hartley 2. We compare the frequency of outbursts at the four comets, and suggest that the surface of 46P has large-scale (∼10-100 m) roughness that is intermediate to that of 67P and 103P, if not similar to the latter. The strength of the outbursts appear to be correlated with time since the last event, but a physical interpretation with respect to solar insolation more » is lacking. We also examine Hubble Space Telescope images taken about 2 days following a near-perihelion outburst. No evidence for macroscopic ejecta was found in the image, with a limiting radius of about 2-m. « less
Authors:
; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; more » ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; « less
Award ID(s):
1440341 2034437
Publication Date:
NSF-PAR ID:
10280435
Journal Name:
ArXivorg
ISSN:
2331-8422
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract High-resolution infrared spectra of comet C/2014 Q2 Lovejoy were acquired with NIRSPEC at the W. M. Keck Observatory on two post-perihelion dates (UT 2015 February 2 and 3). H 2 O was measured simultaneously with CO, CH 3 OH, H 2 CO, CH 4 , C 2 H 6 , C 2 H 4 , C 2 H 2 , HCN, and NH 3 on both dates, and rotational temperatures, production rates, relative abundances, H 2 O ortho-to-para ratios, and spatial distributions in the coma were determined. The first detection of C 2 H 4 in a comet frommore »ground-based observations is reported. Abundances relative to H 2 O for all species were found to be in the typical range compared with values for other comets in the overall population to date. There is evidence of variability in rotational temperatures and production rates on timescales that are small compared with the rotational period of the comet. Spatial distributions of volatiles in the coma suggest complex outgassing behavior. CH 3 OH, HCN, C 2 H 6 , and CH 4 spatial distributions in the coma are consistent with direct release from associated ices in the nucleus and are peaked in a more sunward direction compared with co-measured dust. H 2 O spatial profiles are clearly distinct from these other four species, likely due to a sizable coma contribution from icy grain sublimation. Spatial distributions for C 2 H 2 , H 2 CO, and NH 3 suggest substantial contributions from extended coma sources, providing further evidence for distinct origins and associations for these species in comets. CO shows a different spatial distribution compared with other volatiles, consistent with jet activity from discrete nucleus ice sources.« less
  2. Context. Inferences about dark matter, dark energy, and the missing baryons all depend on the accuracy of our model of large-scale structure evolution. In particular, with cosmological simulations in our model of the Universe, we trace the growth of structure, and visualize the build-up of bigger structures from smaller ones and of gaseous filaments connecting galaxy clusters. Aims. Here we aim to reveal the complexity of the large-scale structure assembly process in great detail and on scales from tens of kiloparsecs up to more than 10 Mpc with new sensitive large-scale observations from the latest generation of instruments. We alsomore »aim to compare our findings with expectations from our cosmological model. Methods. We used dedicated SRG/eROSITA performance verification (PV) X-ray, ASKAP/EMU Early Science radio, and DECam optical observations of a ~15 deg 2 region around the nearby interacting galaxy cluster system A3391/95 to study the warm-hot gas in cluster outskirts and filaments, the surrounding large-scale structure and its formation process, the morphological complexity in the inner parts of the clusters, and the (re-)acceleration of plasma. We also used complementary Sunyaev-Zeldovich (SZ) effect data from the Planck survey and custom-made Galactic total (neutral plus molecular) hydrogen column density maps based on the HI4PI and IRAS surveys. We relate the observations to expectations from cosmological hydrodynamic simulations from the Magneticum suite. Results. We trace the irregular morphology of warm and hot gas of the main clusters from their centers out to well beyond their characteristic radii, r 200 . Between the two main cluster systems, we observe an emission bridge on large scale and with good spatial resolution. This bridge includes a known galaxy group but this can only partially explain the emission. Most gas in the bridge appears hot, but thanks to eROSITA’s unique soft response and large field of view, we discover some tantalizing hints for warm, truly primordial filamentary gas connecting the clusters. Several matter clumps physically surrounding the system are detected. For the “Northern Clump,” we provide evidence that it is falling towards A3391 from the X-ray hot gas morphology and radio lobe structure of its central AGN. Moreover, the shapes of these X-ray and radio structures appear to be formed by gas well beyond the virial radius, r 100 , of A3391, thereby providing an indirect way of probing the gas in this elusive environment. Many of the extended sources in the field detected by eROSITA are also known clusters or new clusters in the background, including a known SZ cluster at redshift z = 1. We find roughly an order of magnitude more cluster candidates than the SPT and ACT surveys together in the same area. We discover an emission filament north of the virial radius of A3391 connecting to the Northern Clump. Furthermore, the absorption-corrected eROSITA surface brightness map shows that this emission filament extends south of A3395 and beyond an extended X-ray-emitting object (the “Little Southern Clump”) towards another galaxy cluster, all at the same redshift. The total projected length of this continuous warm-hot emission filament is 15 Mpc, running almost 4 degrees across the entire eROSITA PV observation field. The Northern and Southern Filament are each detected at >4 σ . The Planck SZ map additionally appears to support the presence of both new filaments. Furthermore, the DECam galaxy density map shows galaxy overdensities in the same regions. Overall, the new datasets provide impressive confirmation of the theoretically expected structure formation processes on the individual system level, including the surrounding warm-hot intergalactic medium distribution; the similarities of features found in a similar system in the Magneticum simulation are striking. Our spatially resolved findings show that baryons indeed reside in large-scale warm-hot gas filaments with a clumpy structure.« less
  3. Obeid, Iyad Selesnick (Ed.)
    Electroencephalography (EEG) is a popular clinical monitoring tool used for diagnosing brain-related disorders such as epilepsy [1]. As monitoring EEGs in a critical-care setting is an expensive and tedious task, there is a great interest in developing real-time EEG monitoring tools to improve patient care quality and efficiency [2]. However, clinicians require automatic seizure detection tools that provide decisions with at least 75% sensitivity and less than 1 false alarm (FA) per 24 hours [3]. Some commercial tools recently claim to reach such performance levels, including the Olympic Brainz Monitor [4] and Persyst 14 [5]. In this abstract, we describemore »our efforts to transform a high-performance offline seizure detection system [3] into a low latency real-time or online seizure detection system. An overview of the system is shown in Figure 1. The main difference between an online versus offline system is that an online system should always be causal and has minimum latency which is often defined by domain experts. The offline system, shown in Figure 2, uses two phases of deep learning models with postprocessing [3]. The channel-based long short term memory (LSTM) model (Phase 1 or P1) processes linear frequency cepstral coefficients (LFCC) [6] features from each EEG channel separately. We use the hypotheses generated by the P1 model and create additional features that carry information about the detected events and their confidence. The P2 model uses these additional features and the LFCC features to learn the temporal and spatial aspects of the EEG signals using a hybrid convolutional neural network (CNN) and LSTM model. Finally, Phase 3 aggregates the results from both P1 and P2 before applying a final postprocessing step. The online system implements Phase 1 by taking advantage of the Linux piping mechanism, multithreading techniques, and multi-core processors. To convert Phase 1 into an online system, we divide the system into five major modules: signal preprocessor, feature extractor, event decoder, postprocessor, and visualizer. The system reads 0.1-second frames from each EEG channel and sends them to the feature extractor and the visualizer. The feature extractor generates LFCC features in real time from the streaming EEG signal. Next, the system computes seizure and background probabilities using a channel-based LSTM model and applies a postprocessor to aggregate the detected events across channels. The system then displays the EEG signal and the decisions simultaneously using a visualization module. The online system uses C++, Python, TensorFlow, and PyQtGraph in its implementation. The online system accepts streamed EEG data sampled at 250 Hz as input. The system begins processing the EEG signal by applying a TCP montage [8]. Depending on the type of the montage, the EEG signal can have either 22 or 20 channels. To enable the online operation, we send 0.1-second (25 samples) length frames from each channel of the streamed EEG signal to the feature extractor and the visualizer. Feature extraction is performed sequentially on each channel. The signal preprocessor writes the sample frames into two streams to facilitate these modules. In the first stream, the feature extractor receives the signals using stdin. In parallel, as a second stream, the visualizer shares a user-defined file with the signal preprocessor. This user-defined file holds raw signal information as a buffer for the visualizer. The signal preprocessor writes into the file while the visualizer reads from it. Reading and writing into the same file poses a challenge. The visualizer can start reading while the signal preprocessor is writing into it. To resolve this issue, we utilize a file locking mechanism in the signal preprocessor and visualizer. Each of the processes temporarily locks the file, performs its operation, releases the lock, and tries to obtain the lock after a waiting period. The file locking mechanism ensures that only one process can access the file by prohibiting other processes from reading or writing while one process is modifying the file [9]. The feature extractor uses circular buffers to save 0.3 seconds or 75 samples from each channel for extracting 0.2-second or 50-sample long center-aligned windows. The module generates 8 absolute LFCC features where the zeroth cepstral coefficient is replaced by a temporal domain energy term. For extracting the rest of the features, three pipelines are used. The differential energy feature is calculated in a 0.9-second absolute feature window with a frame size of 0.1 seconds. The difference between the maximum and minimum temporal energy terms is calculated in this range. Then, the first derivative or the delta features are calculated using another 0.9-second window. Finally, the second derivative or delta-delta features are calculated using a 0.3-second window [6]. The differential energy for the delta-delta features is not included. In total, we extract 26 features from the raw sample windows which add 1.1 seconds of delay to the system. We used the Temple University Hospital Seizure Database (TUSZ) v1.2.1 for developing the online system [10]. The statistics for this dataset are shown in Table 1. A channel-based LSTM model was trained using the features derived from the train set using the online feature extractor module. A window-based normalization technique was applied to those features. In the offline model, we scale features by normalizing using the maximum absolute value of a channel [11] before applying a sliding window approach. Since the online system has access to a limited amount of data, we normalize based on the observed window. The model uses the feature vectors with a frame size of 1 second and a window size of 7 seconds. We evaluated the model using the offline P1 postprocessor to determine the efficacy of the delayed features and the window-based normalization technique. As shown by the results of experiments 1 and 4 in Table 2, these changes give us a comparable performance to the offline model. The online event decoder module utilizes this trained model for computing probabilities for the seizure and background classes. These posteriors are then postprocessed to remove spurious detections. The online postprocessor receives and saves 8 seconds of class posteriors in a buffer for further processing. It applies multiple heuristic filters (e.g., probability threshold) to make an overall decision by combining events across the channels. These filters evaluate the average confidence, the duration of a seizure, and the channels where the seizures were observed. The postprocessor delivers the label and confidence to the visualizer. The visualizer starts to display the signal as soon as it gets access to the signal file, as shown in Figure 1 using the “Signal File” and “Visualizer” blocks. Once the visualizer receives the label and confidence for the latest epoch from the postprocessor, it overlays the decision and color codes that epoch. The visualizer uses red for seizure with the label SEIZ and green for the background class with the label BCKG. Once the streaming finishes, the system saves three files: a signal file in which the sample frames are saved in the order they were streamed, a time segmented event (TSE) file with the overall decisions and confidences, and a hypotheses (HYP) file that saves the label and confidence for each epoch. The user can plot the signal and decisions using the signal and HYP files with only the visualizer by enabling appropriate options. For comparing the performance of different stages of development, we used the test set of TUSZ v1.2.1 database. It contains 1015 EEG records of varying duration. The any-overlap performance [12] of the overall system shown in Figure 2 is 40.29% sensitivity with 5.77 FAs per 24 hours. For comparison, the previous state-of-the-art model developed on this database performed at 30.71% sensitivity with 6.77 FAs per 24 hours [3]. The individual performances of the deep learning phases are as follows: Phase 1’s (P1) performance is 39.46% sensitivity and 11.62 FAs per 24 hours, and Phase 2 detects seizures with 41.16% sensitivity and 11.69 FAs per 24 hours. We trained an LSTM model with the delayed features and the window-based normalization technique for developing the online system. Using the offline decoder and postprocessor, the model performed at 36.23% sensitivity with 9.52 FAs per 24 hours. The trained model was then evaluated with the online modules. The current performance of the overall online system is 45.80% sensitivity with 28.14 FAs per 24 hours. Table 2 summarizes the performances of these systems. The performance of the online system deviates from the offline P1 model because the online postprocessor fails to combine the events as the seizure probability fluctuates during an event. The modules in the online system add a total of 11.1 seconds of delay for processing each second of the data, as shown in Figure 3. In practice, we also count the time for loading the model and starting the visualizer block. When we consider these facts, the system consumes 15 seconds to display the first hypothesis. The system detects seizure onsets with an average latency of 15 seconds. Implementing an automatic seizure detection model in real time is not trivial. We used a variety of techniques such as the file locking mechanism, multithreading, circular buffers, real-time event decoding, and signal-decision plotting to realize the system. A video demonstrating the system is available at: https://www.isip.piconepress.com/projects/nsf_pfi_tt/resources/videos/realtime_eeg_analysis/v2.5.1/video_2.5.1.mp4. The final conference submission will include a more detailed analysis of the online performance of each module. ACKNOWLEDGMENTS Research reported in this publication was most recently supported by the National Science Foundation Partnership for Innovation award number IIP-1827565 and the Pennsylvania Commonwealth Universal Research Enhancement Program (PA CURE). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the official views of any of these organizations. REFERENCES [1] A. Craik, Y. He, and J. L. Contreras-Vidal, “Deep learning for electroencephalogram (EEG) classification tasks: a review,” J. Neural Eng., vol. 16, no. 3, p. 031001, 2019. https://doi.org/10.1088/1741-2552/ab0ab5. [2] A. C. Bridi, T. Q. Louro, and R. C. L. Da Silva, “Clinical Alarms in intensive care: implications of alarm fatigue for the safety of patients,” Rev. Lat. Am. Enfermagem, vol. 22, no. 6, p. 1034, 2014. https://doi.org/10.1590/0104-1169.3488.2513. [3] M. Golmohammadi, V. Shah, I. Obeid, and J. Picone, “Deep Learning Approaches for Automatic Seizure Detection from Scalp Electroencephalograms,” in Signal Processing in Medicine and Biology: Emerging Trends in Research and Applications, 1st ed., I. Obeid, I. Selesnick, and J. Picone, Eds. New York, New York, USA: Springer, 2020, pp. 233–274. https://doi.org/10.1007/978-3-030-36844-9_8. [4] “CFM Olympic Brainz Monitor.” [Online]. Available: https://newborncare.natus.com/products-services/newborn-care-products/newborn-brain-injury/cfm-olympic-brainz-monitor. [Accessed: 17-Jul-2020]. [5] M. L. Scheuer, S. B. Wilson, A. Antony, G. Ghearing, A. Urban, and A. I. Bagic, “Seizure Detection: Interreader Agreement and Detection Algorithm Assessments Using a Large Dataset,” J. Clin. Neurophysiol., 2020. https://doi.org/10.1097/WNP.0000000000000709. [6] A. Harati, M. Golmohammadi, S. Lopez, I. Obeid, and J. Picone, “Improved EEG Event Classification Using Differential Energy,” in Proceedings of the IEEE Signal Processing in Medicine and Biology Symposium, 2015, pp. 1–4. https://doi.org/10.1109/SPMB.2015.7405421. [7] V. Shah, C. Campbell, I. Obeid, and J. Picone, “Improved Spatio-Temporal Modeling in Automated Seizure Detection using Channel-Dependent Posteriors,” Neurocomputing, 2021. [8] W. Tatum, A. Husain, S. Benbadis, and P. Kaplan, Handbook of EEG Interpretation. New York City, New York, USA: Demos Medical Publishing, 2007. [9] D. P. Bovet and C. Marco, Understanding the Linux Kernel, 3rd ed. O’Reilly Media, Inc., 2005. https://www.oreilly.com/library/view/understanding-the-linux/0596005652/. [10] V. Shah et al., “The Temple University Hospital Seizure Detection Corpus,” Front. Neuroinform., vol. 12, pp. 1–6, 2018. https://doi.org/10.3389/fninf.2018.00083. [11] F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” J. Mach. Learn. Res., vol. 12, pp. 2825–2830, 2011. https://dl.acm.org/doi/10.5555/1953048.2078195. [12] J. Gotman, D. Flanagan, J. Zhang, and B. Rosenblatt, “Automatic seizure detection in the newborn: Methods and initial evaluation,” Electroencephalogr. Clin. Neurophysiol., vol. 103, no. 3, pp. 356–362, 1997. https://doi.org/10.1016/S0013-4694(97)00003-9.« less
  4. Abstract Centaurs are small bodies orbiting in the giant planet region that were scattered inward from their source populations beyond Neptune. Some members of the population display comet-like activity during their transition through the solar system, the source of which is not well understood. The range of heliocentric distances where the active Centaurs have been observed and their median lifetime in the region suggest that this activity is driven neither by water-ice sublimation nor entirely by supervolatiles. Here we present an observational and thermodynamical study of 13 Centaurs discovered in the Pan-STARRS1 detection database aimed at identifying and characterizing activemore »objects beyond the orbit of Jupiter. We find no evidence of activity associated with any of our targets at the time of their observations with the Gemini North telescope in 2017 and 2018, or in archival data from 2013 to 2019. Upper limits on the possible volatile and dust production rates from our targets are 1–2 orders of magnitude lower than production rates in some known comets and are in agreement with values measured for other inactive Centaurs. Our numerical integrations show that the orbits of six of our targets evolved interior to r ∼ 15 au over the past 100,000 yr, where several possible processes could trigger sublimation and outgassing, but their apparent inactivity indicates that either their dust production is below our detection limit or the objects are dormant. Only one Centaur in our sample—2014 PQ 70 —experienced a sudden decrease in semimajor axis and perihelion distance attributed to the onset of activity for some previously known inactive Centaurs, and therefore it is the most likely candidate for any future outburst. This object should be a target of high interest for any further observational monitoring.« less
  5. ABSTRACT The galaxy size–stellar mass and central surface density–stellar mass relationships are fundamental observational constraints on galaxy formation models. However, inferring the physical size of a galaxy from observed stellar emission is non-trivial due to various observational effects, such as the mass-to-light ratio variations that can be caused by non-uniform stellar ages, metallicities, and dust attenuation. Consequently, forward-modelling light-based sizes from simulations is desirable. In this work, we use the skirt  dust radiative transfer code to generate synthetic observations of massive galaxies ($M_{*}\sim 10^{11}\, \rm {M_{\odot }}$ at z = 2, hosted by haloes of mass $M_{\rm {halo}}\sim 10^{12.5}\, \rm {M_{\odotmore »}}$) from high-resolution cosmological zoom-in simulations that form part of the Feedback In Realistic Environments project. The simulations used in this paper include explicit stellar feedback but no active galactic nucleus (AGN) feedback. From each mock observation, we infer the effective radius (Re), as well as the stellar mass surface density within this radius and within $1\, \rm {kpc}$ (Σe and Σ1, respectively). We first investigate how well the intrinsic half-mass radius and stellar mass surface density can be inferred from observables. The majority of predicted sizes and surface densities are within a factor of 2 of the intrinsic values. We then compare our predictions to the observed size–mass relationship and the Σ1−M⋆ and Σe−M⋆ relationships. At z ≳ 2, the simulated massive galaxies are in general agreement with observational scaling relations. At z ≲ 2, they evolve to become too compact but still star forming, in the stellar mass and redshift regime where many of them should be quenched. Our results suggest that some additional source of feedback, such as AGN-driven outflows, is necessary in order to decrease the central densities of the simulated massive galaxies to bring them into agreement with observations at z ≲ 2.« less