Search NASA⌕ Search

SEARCH · Search NASA

Results for “data streaming”

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 811 records · Page 45
All sourcesNASA NTRSDOE OSTI
NASA NTRS2003

Possible Stick-Slip Mechanism for Whillans Ice Stream

Tidally-induced stick-slip motion in the mouth of Whillans Ice Stream provides a unique natural experiment in ice-stream response behavior and from which we might learn a great deal about subglacial till properties and sub-ice-stream conditions. At the IGS Symposium on Fast Glacier Flow (Yakutat, 2002), we reported our observations of stick- slip motion and demonstrated its synchronicity with tidal forcing. Recently, we have completed additional processing of our GPS data in differential mode. It reveals more details of the stick-slip events and illustrates that within 30 seconds, the temporal interval of our data, the ice stream accelerates to a speed corresponding to a completely lubricated bed. While details of individual events vary, there seems to be strong evidence of an elastic rebound on the time scale of one hour following most events. This suggests the event involves the release of stored elastic strain energy in the ice. The similar displacements of events suggest further that till or subglacial hydrologic properties limit the amount of elastic strain released in any single event. We follow a line of reasoning that dilatant strengthening limits the slip displacement and present model of the stick-slip process. To match the observed delay between the peak ocean tide and stick-slip events, our model includes a propagating pressure wave in the subglacial hydrologic system between the grounding line, where the rising tide first increases the subglacial water pressure and regions upstream where stored elastic strain increases the basal shear stress. This high-tide event is released when the increased water pressure reaches the region of increased shear stress. Dilatant strengthening stops the event by increasing pore volume and lowering the water pressure. Following this event, falling tide increases the normal forces, compresses the till and increases pore pressure again, leading to the second falling-tide event we observe every tidal cycle.

Bindschadler, Robert↗
NASA NTRS2018

Space-to-Ground Interactions While Conducting Scientific Fieldwork Under Mars Mission Constraints

The Biologic Analog Science Associated with Lava Terrains (BASALT) project is a 4-year program dedicated to iteratively designing, implementing, and evaluating concepts of operations (ConOps) and supporting capabilities to enable and enhance scientific exploration for future human Mars missions. BASALT incorporates three field deployments during which real (non-simulated) biological and geochemical field science is conducted at two high-fidelity Mars analog locations under simulated Mars mission conditions, including communication de-lays and data transmission limitations. BASALT’s primary science objective is to investigate how the redox conditions of altered basaltic environments affect the development of microbial communities in these Mars-relevant settings. Field sites include the active East Rift Zone on the Big Island of Hawaii, reminiscent of early Mars when basaltic volcanism and interaction with water were widespread, and the dormant eastern Snake River Plain in Idaho, similar to present-day Mars where basaltic volcanism is rare and most evidence for volcano-driven hydrothermal activity is relict. BASALT’s primary science operations objective is to investigate exploration ConOps and capabilities that facilitate scientific return during human-robotic exploration under Mars mission constraints. Each field deployment consists of ten extravehicular activities (EVAs) on the volcanic flows in which two extravehicular and two intravehicular (IV) crew-members conduct the science while communicating across time delay and under bandwidth constraints with an Earth-based Mission Support Center (MSC) comprised of expert scientists and operators. Communication latencies of 5 and 15-minute one-way light time and low (0.512 Mb/s uplink, 1.54 Mb/s down-link) and high (5.0 Mb/s uplink, 10.0 Mb/s downlink) bandwidth conditions are being evaluated. EVA crewmembers communicate with the MSC via voice and text messaging and provide scientific instrument data, still imagery, video streams, and GPS tracking information. The MSC reviews this data across delay and provides recommendations for presampling and sampling tasks. The scientists used dynamic leaderboards (priority rank-ing lists), to track and rank candidate samples relative to one another and against the science objectives for the current EVA and the overall mission. Updates to the dynamic leaderboards are relayed regularly to the IV crewmembers to provide scientific feedback from Earth and to help minimize crew idle time (time spent waiting for Earth input during which no productive tasks are performed). EVA timelines are strategically designed to enable continuous (delayed) feedback from an Earth-based science team while simultaneously minimizing crew idle time. Such timelines are operationally advantageous, reducing transport costs by eliminating the need for crews to return to the same locations on multiple EVAs while still providing opportunities for recommendations from science experts on Earth, and scientifically advantageous by minimizing the potential for cross-contamination across sites. This paper will highlight the space-to-ground interaction results from the three BASALT field deployments, including planned versus actual EVA time-line data, ground assimilation times (the amount of time available to the MSC to provide input to the crew), and idle time. Furthermore, we describe how these results vary under the different communication latency and bandwidth conditions. Together, these data will provide a basis for guiding and prioritizing capability development for future human exploration missions.

Beaton, Kara H.↗
NASA NTRS1986

Gulf stream velocity structure through combined inversion of hydrographic and acoustic Doppler data

Near-surface velocities from an acoustic Doppler instrument are used in conjunction with CTD/O2 data to produce estimates of the absolute flow field off Cape Hatteras. The data set consists of two transects across the Gulf Stream made by the R/V Endeavor cruise EN88 in August 1982. An inverse procedure is applied which makes use of both the acoustic Doppler data and property conservation constraints. Velocity sections at approximately 73 deg. W and 71 deg. W are presented with formal errors of 1-2 cm/s. The net Gulf Stream transports are estimated to be 116 + or - 2 Sv across the south leg and 161 + or - 4 Sv across the north. A Deep Western Boundary Current transport of 4 + or - 1 Sv is also estimated. While these values do not necessarily represent the mean, they are accurate estimates of the synoptic flow field in the region.

Pierce, S. D.↗
NASA NTRS1985

On boundary layer transition in high-subsonic and transonic flow under the influence of acoustic disturbances and free-stream turbulence

The existing data base which quantifies the roles played by the Mach number, turbulence level, and acoustic noise in the flow quality of wind tunnels is reviewed. Attention is given to the effects on end (ET) and beginning (BT) transitions in flows over a 10 deg cone in transonic channels. The Re of both ET and BT flows were significantly affected by the Mach number, acoustic noise level and the noise frequency content. A unit increment in the Mach number increased the Re beyond 3,000,000, while noise below the 1 percent level in the rms pressure coefficient did not alter the transition phenomenon. More data are required on the effects of turbulence, which was a significant factor, and in the settling chamber, test section, plenum and diffuser to trace the locations and magnitudes of the acoustic sources.

Murthy, S. V.↗
NASA NTRS2009

Complex Event Recognition Architecture

Complex Event Recognition Architecture (CERA) is the name of a computational architecture, and software that implements the architecture, for recognizing complex event patterns that may be spread across multiple streams of input data. One of the main components of CERA is an intuitive event pattern language that simplifies what would otherwise be the complex, difficult tasks of creating logical descriptions of combinations of temporal events and defining rules for combining information from different sources over time. In this language, recognition patterns are defined in simple, declarative statements that combine point events from given input streams with those from other streams, using conjunction, disjunction, and negation. Patterns can be built on one another recursively to describe very rich, temporally extended combinations of events. Thereafter, a run-time matching algorithm in CERA efficiently matches these patterns against input data and signals when patterns are recognized. CERA can be used to monitor complex systems and to signal operators or initiate corrective actions when anomalous conditions are recognized. CERA can be run as a stand-alone monitoring system, or it can be integrated into a larger system to automatically trigger responses to changing environments or problematic situations.

Fitzgerald, William A.↗
DOE OSTI2025

Applying Transfer Learning for Street-Scale Nuisance Flood Forecasting in Coastal-Urban Cities

An important challenge with Machine Learning (ML) is its transferability; that is, whether a ML model trained on one set of data can be applied to a second set of data without requiring a full re-training of the model. Transfer Learning (TL) addresses this challenge by transferring knowledge learned in the source domain (the data it was trained on) to the target domain (a second set of data that is statistically different but related, which the model was not trained on). This study investigates the use of TL for street-scale nuisance flood forecasting by exploring whether a ML model trained on data collected for one set of streets can effectively forecast flooding for another set of streets in the same city using TL. The envisioned use case is a city deploying a new flood depth monitoring sensor on a street and using TL to apply a ML model, trained on sensor data from an existing flood depth sensor network, to this new street. Eventually, the new flood depth sensor will have a sufficient dataset for training its own ML model, but TL can be used to fill the gap in time while this new dataset is being generated. This method is explored using a Long Short-Term Memory (LSTM) model trained on data for the flood-prone streets of Norfolk City, Virginia. The data used for training includes environmental time series (rainfall, tide), topographic features (Digital Elevation Model (DEM), Topographic Wetness Index (TWI), Depth To Water (DTW)), and street-scale flood depth time series obtained from a high-fidelity physics-based model, acting as a synthetic street-scale stream depth sensor dataset since actual stream depth sensor data is generally unavailable for most cities. A set of 180 flood-prone streets was used to train a base model, while another set of 180 flood-prone streets was used to re-train that model using different TL strategies. The results show that full-weight re-training proved most effective and minimal re-training of only the output layer was insufficient. The advantage of TL was most pronounced when target data was limited, meaning data collected at the new water depth sensor location included generally less than 18 flood events. As target data increased beyond 18 flood events, the benefit of TL diminished relative to training a ML model directly on the local flood events. These findings can assist cities as they implement street-scale flood sensing systems to create accurate forecasts for new sensing locations that do not yet have sufficient data records to train a local ML model.

Roy, Binata [Univ. of Virginia, Charlottesville, V↗
NASA NTRS2002

Experimental Investigation of Transition to Turbulence as Affected by Passing Wakes

Experimental results from a study of the effects of passing wakes upon laminar-to-turbulent transition in a low-pressure turbine passage are presented. The test section geometry is designed to simulate the effects of unsteady wakes resulting from rotor-stator interaction upon laminar-to-turbulent transition in turbine blade boundary layers and separated flow regions over suction surfaces. Single-wire, thermal anemometry techniques were used to measure time-resolved and phase-averaged, wall-normal profiles of velocity, turbulence intensity, and intermittency at multiple streamwise locations over the turbine airfoil suction surface. These data are compared to steady state, wake-free data collected in the same geometry to identify the effects of wakes upon laminar-to-turbulent transition. Results are presented for flows with a Reynolds number based on suction surface length and exit velocity of 50,000 and an approach flow turbulence intensity of 2.5 percent. From these data, the effects of passing wakes and associated increased turbulence levels and varying pressure gradients on transition and separation in the near-wall flow are presented. The results show that the wakes affect transition both by virtue of their difference in turbulence level from that of the free-stream but also by virtue of their velocity deficit relative to the freestream velocity, and the concomitant change in angle of attack and temporal pressure gradients. The results of this study seem to support the theory that bypass transition is a response of the near-wall viscous layer to pressure fluctuations imposed upon it from the free-stream flow. The data also show a significant lag between when the wake is present over the surface and when transition begins. The accompanying CD-ROM includes tabulated data, animations, higher resolution plots, and an electronic copy of this report.

Kaszeta, Richard W.↗
NASA NTRS2002

Experimental Investigation of Transition to Turbulence as Affected by Passing Wakes

Experimental results from a study of the effects of passing wakes upon laminar-to-turbulent transition in a low-pressure turbine passage are presented. The test section geometry is designed to simulate the effects of unsteady wakes resulting from rotor-stator interaction upon laminar-to-turbulent transition in turbine blade boundary layers and separated flow regions over suction surfaces. Single-wire, thermal anemometry techniques were used to measure time- resolved and phase-averaged, wall-normal profiles of velocity, turbulence intensity, and intermittency at multiple streamwise locations over the turbine airfoil suction surface. These data are compared to steady state, wake-free data collected in the same geometry to identify the effects of wakes upon laminar-to-turbulent transition. Results are presented for flows with a Reynolds number based on suction surface length and exit velocity of 50,000 and an approach flow turbulence intensity of 2.5 percent. From these data, the effects of passing wakes and associated increased turbulence levels and varying pressure gradients on transition and separation in the near-wall flow are presented. The results show that the wakes affect transition both by virtue of their difference in turbulence level from that of the free-stream but also by virtue of their velocity deficit relative to the freestream velocity, and the concomitant change in angle of attack and temporal pressure gradients. The results of this study seem to support the theory that bypass transition is a response of the near-wall viscous layer to pressure fluctuations imposed upon it from the free-stream flow. The data also show a significant lag between when the wake is present over the surface and when transition begins. The accompanying CD-ROM includes tabulated data, animations, higher resolution plots, and an electronic copy of this report.

Kaszeta, Richard W.↗
NASA NTRS1994

X-ray bright points and high-speed wind streams: A preliminary analysis from Yohkoh and Ulysses data

The following aspect of the solar wind mass flux, and of its variation, is examined: whether coronal plumes might be responsible for the long-term variability of the mass flux in high-speed streams emanating from coronal holes. The assumption that plumes are rooted in coronal bright points (BP's) is made. The behavior of X-ray BP's, imaged by the Yohkoh soft X-ray telescope (SXT), during a seven month period when Ulysses experiments observed a series of recurrent high-speed streams, is analyzed. If plumes/BP's are sources of the wind mass flux, changes in the coronal hole BP density to mimic changes of the mass flux in high-speed streams are expected. SOHO will have the capability of measuring the solar wind speed/density at small heliocentric distances while simultaneously observing coronal BP's and coronal plumes.

Poletto, Giannina↗
DOE OSTI2024

Exploring DAOS as a Burst Buffer for a 100 Gbps DAQ Real-Time Streaming System

We present an experimental evaluation of a burst buffer for a real-time DAQ streaming system designed to transmit instrument data to remote data centers. The system is based on EJ-FAT, a load balancing system capable of Nx 100Gbps streams, distributing data from event sources to processing nodes. We explore applying the DAOS system as a burst buffer to serve a number of purposes: improve resiliency, elasticity and add new functions into the processing pipeline. In the evaluation a sender transmits events over a 100Gbps network to a receiver integrated with DAOS to store the reassembled events using DAOS APIs. We evaluate the system for possible bottlenecks and provide end-to-end evaluation with a burst buffer using DAOS storage abstractions. We show that a receiver node can support 38.1 Gbps. This proves the viability of our approach and allows us to extend this work to investigate scale-out properties and new streaming optimizations.

Mei, Xinxin↗
NASA NTRSDate not supplied

Mapping National Forest Aboveground Biomass in Mexico By Integrating GEDI and Landsat Times Series Data

Mexico is one of the countries with great potential for the UN's Reducing Emissions from Deforestation and Forest Degradation (REDD+) program, a key nature-based solution for the forest sector. To monitor carbon stock changes, there is a growing demand for unbiased Monitoring Reporting Verification (MRV) systems to facilitate effective forest management and climate change mitigation strategies. Remote sensing-based national aboveground biomass density (AGBD) estimation over Mexico is scarce and often limited to one-time static mapping, leading to spatiotemporal inconsistency in inputs. As an effort under NASA's Carbon Monitoring System (CMS) program, we have developed a remote sensing-based approach to create consistent historical AGBD maps of Mexico using multi-stream remote sensing data, including spaceborne lidar GEDI and long-term Landsat time series, as well as topographic information. We employ the continuous change detection and classification (CCDC) algorithm for temporal modeling of Landsat surface reflectance, followed by the inference of forest AGBD using a random forest machine learning algorithm with the temporal information of land surface dynamics extracted by the CCDC as input. GEDI provides unprecedented forest structure and AGBD sampling datasets for model training and validation practices. In this presentation, we share the progress made in developing a spatially explicit mapping of historical AGBD changes associated with land surface changes and post-disturbance landscapes.

Taejin Park↗
DOE OSTI2025

The Artificial Scientist: in-Transit Machine Learning of Plasma Simulations

Large-scale simulations or scientific experiments produce petabytes of data per run. This poses massive challenges for I/O and storage when scientific analysis workflows are run manually offline. Unsupervised deep learning-based techniques to extract patterns and non-linear relations from these large amounts of data provide a way to build scientific understanding from raw data, reducing the need for manual pre-selection of analysis steps, but require exascale compute and memory to process the full dataset available. In this paper, we demonstrate a heterogeneous streaming workflow in which plasma simulation data is streamed directly to a Machine Learning (ML) application training a model on the simulation data in-transit, completely circumventing the capacity-constrained filesystem bottleneck. This workflow employs openPMD to provide a high level interface to describe scientific data and also uses ADIOS2, to transfer volumes of data that exceed the capabilities of the filesystem. We employ experience replay to avoid catastrophic forgetting in learning from this non-steady state process in a continual manner and adapt it to improve model convergence while learning in-transit. As a proof-of-concept, we approach the ill-posed inverse problem of predicting particle dynamics from radiation in a particle-incell (PIConGPU) simulation of the Kelvin-Helmholtz instability (KHI). We detail hardware-software co-design challenges as we scale PIConGPU to full Frontier, the Top-1 system as of June 2024 Top500 list.

Kelling, Jeffrey [Helmholtz-Zentrum Dresden Rossen↗
NASA NTRS1959

Effects of Nose Radius and Extreme Cooling on Boundary-Layer Transition for Three Smooth 15°-Cone-Cylinders in Free Flight at Mach Numbers to 8.50

Three highly polished 15°- included- angle cone- cylinders with hemispherical tips of several diameters ( 2, 3, and 4 in.) have been flown in order to obtain boundary- layer transition data at very low wall to local stream temperature ratios, and heat- transfer data. All surfaces had a 2-microinch average roughness height. Laminar flow existed over the entire hemispherical nose of the 2- and 3-inch-tip- diameter models throughout the complete flight history. Extreme cooling to wall to local stream temperature ratios at the sonic point as low as 0.20 did not cause transition on the nose for diameters as large as 3 inches. However, extreme cooling did cause early transition on the 4-inch model where it appears probable that transition occurred forward of the 45° station at a wall to local stream temperature ratio of about 0.26. Variations in tip diameter influenced transition downstream of the nose under conditions of extreme cooling. The 2-inch- tip model was laminar at all cone- cylinder stations at temperature ratios as low as 0.32 whereas the 3- and 4-inch-tip models were turbulent at the same local flow conditions but at higher wall to local temperature ratios. Transition on the cone and cylinder of the 3- and 4-inch- tip bodies appeared to be sensitive to local Mach number, and occurred at higher local temperature ratios when values of local Mach number were higher. Increasing the nose diameter from 2 to 3 inches significantly changed the local flow conditions for which laminar flow existed on the cone- cylinder afterbody. However, a further increase in tip size t o a 4-inch diameter had no discernable effect on the local flow conditions at transition. The transition results of the 3- and 4-inch-nose-diameter smooth bodies are similar to those observed on a 7/8-inch-nose-diameter body with roughened surfaces. Turbulent boundary layers resulted in both cases at very low wall to local stream temperature ratios. Both laminar and turbulent heat-transfer data were in good agreement with theoretical Stanton numbers when heat-transfer reduction due to tip blunting was considered.

CYLINDRICAL AFTERBODY↗
NASA NTRS2004

Science Goal Monitor: Science Goal Driven Automation for NASA Missions

Infusion of automation technologies into NASA s future missions will be essential because of the need to: (1) effectively handle an exponentially increasing volume of scientific data, (2) successfully meet dynamic, opportunistic scientific goals and objectives, and (3) substantially reduce mission operations staff and costs. While much effort has gone into automating routine spacecraft operations to reduce human workload and hence costs, applying intelligent automation to the science side, i.e., science data acquisition, data analysis and reactions to that data analysis in a timely and still scientifically valid manner, has been relatively under-emphasized. In order to introduce science driven automation in missions, we must be able to: capture and interpret the science goals of observing programs, represent those goals in machine interpretable language; and allow spacecrafts onboard systems to autonomously react to the scientist's goals. In short, we must teach our platforms to dynamically understand, recognize, and react to the scientists goals. The Science Goal Monitor (SGM) project at NASA Goddard Space Flight Center is a prototype software tool being developed to determine the best strategies for implementing science goal driven automation in missions. The tools being developed in SGM improve the ability to monitor and react to the changing status of scientific events. The SGM system enables scientists to specify what to look for and how to react in descriptive rather than technical terms. The system monitors streams of science data to identify occurrences of key events previously specified by the scientist. When an event occurs, the system autonomously coordinates the execution of the scientist s desired reactions. Through SGM, we will improve om understanding about the capabilities needed onboard for success, develop metrics to understand the potential increase in science returns, and develop an operational prototype so that the perceived risks associated with increased use of automation can be reduced.

Koratkar, Anuradha↗
DOE OSTI2026

Data, model inputs, and analysis scripts associated with a manuscript on stream intermittency controls across spatial scales in Pacific Northwest watersheds

NOTE: The manuscript associated with this data package is currently in review. The data may be revised based on reviewer feedback. Upon manuscript acceptance, this data package will be updated with the final dataset and additional metadata. This data package is associated with the manuscript "Hydroclimatic Memory and Watershed Template Shape Stream Intermittency: Multi-scale Attribution Using Process-based Simulation and Explainable ML" by Niroula et al. (2026), submitted to Water Resources Research (WRR). The study investigates the dominant controls on stream intermittency across local, reach, and watershed scales using a coupled process-based simulation and explainable machine-learning framework. Long-term daily simulations from the Advanced Terrestrial Simulator (ATS) were used to generate wetness states and ponded-depth responses over river-corridor cells. These ATS outputs were then aggregated across scales and used to train XGBoost (eXtreme Gradient Boosting) models. SHAP (SHapley Additive exPlanations) was applied to quantify the relative importance of hydroclimatic forcings, watershed template attributes, and antecedent-memory effects in shaping intermittency behavior. The analysis is carried out for three contrasting Pacific Northwest watersheds: Oak Creek (OCW), American River Watershed (ARW), and H.J. Andrews (HJA). Across these testbeds, the package contains ATS-ready watershed inputs, ATS run configuration and selected output files, model-evaluation data products, intermittency-analysis datasets, machine-learning target-feature tables, SHAP outputs, and notebooks used to organize, analyze, and visualize results. At a high level, the package documents a workflow in which ATS provides the physically based simulation backbone and explainable machine learning is used as a post-processing attribution tool. The contents are intended to support interpretation of the manuscript figures and results, provide context for how intermittency metrics were generated at multiple scales, and preserve the key artifacts needed to understand and reuse the analysis workflow. The package contains a high-level directory summary file (`summary.txt`) and four main content folders (1) `evaluation_plots` contains evaluation figures and supporting evaluation datasets; (2) `intermittency_plots` contains intermittency-focused analysis notebook and prepared datasets; (3) `ml-training-and-shap_values_plots` contains ML training inputs, SHAP outputs, and figure-generation notebooks; and (4) `watershed_mesh_and_ats_input` contains ATS model setup materials, forcing inputs, geometry, and selected run files. More specifically, the `evaluation_plots` folder contains the notebook used for ATS evaluation plotting and site-specific evaluation datasets. These include evapotranspiration and water-balance products for three watersheds, as well as an Oak Creek field-measurement discharge file. The `intermittency_plots` folder contains the notebook used for intermittency analysis and the prepared datasets used to analyze intermittent and non-intermittent wetness behavior across the study watersheds. The `ml-training-and-shap_values_plots` folder contains notebooks and outputs for the machine-learning and explainability workflow. This includes the main XGBoost and SHAP notebook(s), a beeswarm plotting notebook, target-feature tables for machine-learning training, SHAP summary tables, and per-sample SHAP value archives. The `watershed_mesh_and_ats_input` folder contains ATS-related watershed inputs and supporting materials. This includes mesh and shape products, ATS-readable LAI and meteorological forcing inputs, selected ATS spinup and transient-run files, and a watershed workflow example notebook. Subdirectories are organized by watershed where applicable.All files are .cpg (codepage files), .csv (comma-separated values), .dbf (database files), .exo (Exodus mesh format), .h5 (HDF5 format), .ipynb (Jupyter notebooks), .pkl (Python pickle), .prj (projection files), .sh (shell scripts), .shp (shapefile geometry), .shx (shapefile index), .txt (text files), or .xml (markup data).

Advanced Terrestrial Simulator↗
NASA NTRS1978

Large-scale Gulf Stream frontal study using GEOS 3 radar altimeter data

From data obtained by the GEOS 3 radar altimeter, sea surface heights are found by both editing and filtering the raw sea surface height measurements and then referencing these processed data to a 5 foot by 5 foot geoid. Any trend between the processed data and the geoid is removed by subtracting out a linear fit to the residuals in the open ocean. Data from individual passes are further processed by applying a minimum variance technique at the subsatellite crossing points to produce surface topography maps for the 6 months and an overall mean map which reveal important details about the Gulf Stream system. The differences between the monthly mean and the overall mean are calculated for each of the 6 months to show the temporal and spatial changes of the Gulf Stream front and spawned eddies. The standard deviation map is even more informative and shows preferred locations of Gulf Stream meanders.

Huang, N. E.↗