Stream dissolved oxygen (DO) dynamics are an outcome of metabolic activity and subsequently regulate ecosystem functions such as in‐stream solute and sediment reactions. The synchronization of DO signals in and across stream networks is both a cause and effect of the mode and timing of these functions, but there is limited empirical evidence for network patterns of DO synchrony. We used high frequency DO measurements at 42 sites spanning five catchments and stream orders to evaluate DO signal synchrony in response to variation in light (a driver of photosynthesis) and discharge (a control on DO signal spatial extent). We hypothesized that stream network DO synchrony arises when regional controls dominate: when light inputs are synchronous and when longitudinal hydrologic connectivity is high. By complement, we predicted that DO signal synchrony decreases as light becomes more asynchronous and stream flows decline or become discontinuous. Our results supported this hypothesis: greater DO signal synchrony arose with increasing light synchrony and flow connectivity. A model including these two controls explained 70% of variation in DO synchrony. We conclude that DO synchrony patterns within‐ and across‐networks support the current paradigm of discharge and light control on stream metabolic activity. Finally, we propose that DO synchrony patterns are likely a useful prerequisite for scaling subdaily metabolism estimates to network and regional scales.
- Award ID(s):
- 1723773
- NSF-PAR ID:
- 10182977
- Date Published:
- Journal Name:
- Proceedings of the 57th Design Automation Conference (DAC 2020), San Francisco, CA
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
Abstract -
null (Ed.)We present Calyx, a new intermediate language (IL) for compiling high-level programs into hardware designs. Calyx combines a hardware-like structural language with a software-like control flow representation with loops and conditionals. This split representation enables a new class of hardware-focused optimizations that require both structural and control flow information which are crucial for high-level programming models for hardware design. The Calyx compiler lowers control flow constructs using finite-state machines and generates synthesizable hardware descriptions. We have implemented Calyx in an optimizing compiler that translates high-level programs to hardware. We demonstrate Calyx using two DSL-to-RTL compilers, a systolic array generator and one for a recent imperative accelerator language, and compare them to equivalent designs generated using high-level synthesis (HLS). The systolic arrays are 4.6× faster and 1.11× larger on average than HLS implementations, and the HLS-like imperative language compiler is within a few factors of a highly optimized commercial HLS toolchain. We also describe three optimizations implemented in the Calyx compiler.more » « less
-
more » « less
Abstract We present a novel method for measuring the fluctuating basal normal and shear stresses of debris flows by using along‐channel seismic recordings. Our method couples a simple parameterization of a debris flow as a seismic source with direct measurements of seismic path effects using empirical Green's functions generated with a force hammer. We test this method using two large‐scale (8 and 10 m3) experimental flows at the U.S. Geological Survey debris‐flow flume that were recorded by dozens of three‐component seismic sensors. The seismically derived basal stress fluctuations compare well in amplitude and timing to independent force plate measurements within the valid frequency range (15–50 Hz). We show that although the high‐frequency seismic signals provide band‐limited forcing information, there are systematic relations between the fluctuating stresses and independently measured flow properties, especially mean basal shear stress and flow thickness. However, none of the relationships are simple, and since the flow properties also correlate with one another, we cannot isolate a single factor that relates in a simple way to the fluctuating forces. Nevertheless, our observations, most notably the gradually declining ratio of fluctuating to mean basal stresses during flow passage and the distinctive behavior of the coarse, unsaturated flow front, imply that flow style may be a primary control on the conversion of translational to vibrational kinetic energy. This conversion ultimately controls the radiation of high‐frequency seismic waves. Thus, flow style may provide the key to revealing the nature of the relationship between fluctuating forces and other flow properties.
-
more » « less
SUMMARY Yasur volcano, Vanuatu is a continuously active open-vent basaltic-andesite stratocone with persistent and long-lived eruptive activity. We present results from a seismo-acoustic field experiment at Yasur, providing locally dense broad-band seismic and infrasonic network coverage from 2016 July 27 to August 3. We corroborate our seismo-acoustic observations with coincident video data from cameras deployed at the crater and on an unoccupied aircraft system (UAS). The waveforms contain a profusion of signals reflecting Yasur’s rapidly occurring and persistent explosive activity. The typical infrasonic signature of Yasur explosions is a classic short-duration and often asymmetric explosion waveform characterized by a sharp compressive onset and wideband frequency content. The dominant seismic signals are numerous repetitive very-long-period (VLP) signals with periods of ∼2–10 s. The VLP seismic events are ‘high-rate’, reoccurring near-continuously throughout the data set with short interevent times (∼20–60 s). We observe variability in the synchronization of seismic VLP and acoustic sources. Explosion events clearly delineated by infrasonic waveforms are underlain by seismic VLPs. However, strong seismic VLPs also occur with only a weak infrasonic expression. Multiplet analysis of the seismic VLPs reveals a systematic progression in the seismo-acoustic source decoupling. The same dominant seismic VLP multiplet occurs with and without surficial explosions and infrasound, and these transitions occur over a timescale of a few days during our field campaign. We subsequently employ template matching, stacking, and full-waveform inversion to image the source mechanism of the dominant VLP multiplet. Inversion of the dominant VLP multiplet stack points to a composite source consisting of either a dual-crack (plus forces) or pipe-crack (plus forces) mechanism. The derived mechanisms correspond to a point-source directly beneath the summit vents with centroid depths in the range ∼900–1000 m below topography. All mechanisms suggest a northeast trending crack dipping relatively shallowly to the northwest and indicate a VLP source centroid and mechanism controlled by a stable structural geologic feature beneath Yasur. We interpret the results in the framework of gas slug ascent through the conduit responsible for Yasur explosions. The VLP mechanism and timing with infrasound (when present) are explained by a shallow-buffered top-down model in which slug ascent is relatively aseismic until reaching the base of a shallow section. Slug disruption in this shallow zone triggers a pressure disturbance that propagates downward and couples at the conduit base (VLP centroid). If the shallow section is open, an explosion propagates to the surface, producing infrasound. In the case of (the same multiplet) VLPs occurring without surficial explosions and weak or no infrasound, the decoupling of the dominant VLPs at ∼900–1000 m depth from surficial explosions and infrasound strongly indicates buffering of the terminal slug ascent. This buffering could be achieved by a variety of conditions at or directly beneath the vents, such as a high-viscosity layer of crystal-rich magma, a debris cap from backfill, a foam layer, or a combination of these. The dominant VLP at Yasur captured by our experiment has a source depth and mechanism separated from surface processes and is stable over time.
-
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 describe 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.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3373087.3375332
