Search NASA⌕ Search

SEARCH · Search NASA

Results for “difference map”

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 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
NASA NTRS1999

Point Spread Function and Transmittance Analyses for Conical and Hexapod Secondary Mirror Support Towers for the Next Generation Space Telescope (NGST)

Different secondary mirror support towers were modeled on the CODE V optical design/analysis program for the NGST Optical Telescope Assembly (OTA) B. The vertices of the NGST OTA B primary and secondary mirrors were separated by close to 9.0 m. One type of tower consisted of a hollow cone 6.0 m long, 2.00 m in diameter at the base, and 0.704 m in diameter at its top. The base of the cone was considered attached to the primary's reaction structure through a hole in the primary. Extending up parallel to the optical axis from the top of this cylinder were eight blades (pyramidal struts) 3.0 m long. A cross section of each these long blades was an isosceles triangle with a base of 0.010 m and a height of 0.100 m with the sharpest part of each triangle pointing inward. The eight struts occurred every 45 deg. The other type of tower was purely a hexapod arrangement and had no blades or cones. The hexapod consisted simply of six, very thin, circular struts, leaving in pairs at 12:00, 4:00, and 8:00 at the primary and traversing to the outer edge of the back of the secondary mount. At this mount, two struts arrived at each of 10:00, 2:00, and 6:00. The struts were attached to the primary mirror in a ring 3.5 m in diameter. They reached the back of the secondary mount, a circle 0.704 m in diameter. Transmittance analyses at two levels were performed on the secondary mirror support towers. Detailed transmittances were accomplished by the use of the CODE V optical design/analysis program and were compared to transmittance calculations that were almost back-of-the-envelope. Point spread function (PSF) calculations, including both diffraction and aberration effects, were performed on CODE V. As one goes out from the center of the blur (for a point source), the two types of support towers showed little difference between their PSF intensities until one reaches about the 3% level. Contours can be delineated on CODE V down to about 10 (exp -8) times the peak intensity, fine detailing the differences in the wings of the PSF. Between the first type of tower and the hexapod tower, below 3% of the peak intensity, the PSF contour maps well delineated significant differences in the outer portions of the PSFS. Such differences in the wings of the PSF significantly impact the resolution of brown dwarfs or extra solar planets from their star. The hexapod tower provided the best PSF.

Wilkerson, Gary W.↗
DOE OSTI2023

AlignOT: An Optimal Transport Based Algorithm for Fast 3D Alignment With Applications to Cryogenic Electron Microscopy Density Maps

Aligning electron density maps from Cryogenic electron microscopy (cryo-EM) is a first key step for studying multiple conformations of a biomolecule. As this step remains costly and challenging, with standard alignment tools being potentially stuck in local minima, we propose here a new procedure, called AlignOT, which relies on the use of computational optimal transport (OT) to align EM maps in 3D space. By embedding a fast estimation of OT maps within a stochastic gradient descent algorithm, our method searches for a rotation that minimizes the Wasserstein distance between two maps, represented as point clouds. Here, we quantify the impact of various parameters on the precision and accuracy of the alignment, and show that AlignOT can outperform the standard local alignment methods, with an increased range of rotation angles leading to proper alignment. We further benchmark AlignOT on various pairs of experimental maps, which account for different types of conformational heterogeneities and geometric properties. As our experiments show good performance, we anticipate that our method can be broadly applied to align 3D EM maps.

3D alignment↗
NASA NTRS1975

Nearshore coastal mapping

Two test sites of different water quality and bottom topography were used to test for maximum water depth penetration using the Skylab S-192 MSS for measurement of nearshore coastal bathymetry. Sites under investigation lie along the Lake Michigan coastline where littoral transport acts to erode sand bluffs and endangers developments along 1,200 miles of shore, and on the west coast of Puerto Rico where unreliable shoal location and depth information constitutes a safety hazard to navigation. The S-192 and S-190A and B provide data on underwater features because of water transparency in the blue/green portion of the spectrum. Depth of 20 meters were measured with the S-192 in the Puerto Rico test site. The S-190B photography with its improved spatial resolution clearly delineates the triple sand bar topography in the Lake Michigan test site. Several processing techniques were employed to test for maximum depth measurement with least error. The results are useful for helping to determine an optimum spectral bandwidth for future space sensors that will increase depth measurements for different water attenuation conditions where a bottom reflection is detectable.

Polcyn, F. C.↗
DOE OSTI2020

Area-Preserving Mapping of 3D Carotid Ultrasound Images Using Density-Equalizing Reference Map

Carotid atherosclerosis is a focal disease at the bifurcations of the carotid artery. To quantitatively monitor the local changes in the vessel-wall-plus-plaque thickness (VWT) and compare the VWT distributions for different patients or for the same patients at different ultrasound scanning sessions, a mapping technique is required to adjust for the geometric variability of different carotid artery models. In this work, we propose a novel method called density-equalizing reference map (DERM) for mapping 3D carotid surfaces to a standardized 2D carotid template, with an emphasis on preserving the local geometry of the carotid surface by minimizing the local area distortion. The initial map was generated by a previously described arc-length scaling (ALS) mapping method, which projects a 3D carotid surface onto a 2D non-convex L-shaped domain. A smooth and area-preserving flattened map was subsequently constructed by deforming the ALS map using the proposed algorithm that combines the density-equalizing map and the reference map techniques. This combination allows, for the first time, one-to-one mapping from a 3D surface to a standardized non-convex planar domain in an area-preserving manner. Evaluations using 20 carotid surface models show that the proposed method reduced the area distortion of the flattening maps by over 80% as compared to the ALS mapping method.

60 APPLIED LIFE SCIENCES↗
NASA NTRS1975

Applications of Skylab EREP photographs to mapping landforms and environmental geomorphology in the Great Plains and Midwest

The following evaluations of Skylab photographs were undertaken: (1) the 1290 Skylab S190A and S190B photographs of Illinois, Iowa, Kansas, Missouri, Nebraska, and South Dakota were evaluated in detail in terms of coverage, cloud cover, photographic quality, endlap, detectability of roads and stereorelief, and utility for geomorphologic mapping, and (2) the utility of the Skylab photos were tested for interpretive analytic mapping of geomorphologic features over large areas representative of different parts of this region. Photointerpretative maps of analytic geomorphology were obtained for various test areas representative of the varied landscapes in the region. These maps are useful for regional land-use planning, ground-water exploration, and other environmental geomorphologic-geologic applications. Compared with LANDSAT-1 MSS images, Skylab photos afford almost as extensive overviews of large areas but in considerably greater detail, and for many SL photos, moderate stereorelief. However, repetitive multiseasonal, cloud-free coverage by high-quality photos is very limited and many areas have no coverage at all.

Morrison, R. B.↗
DOE OSTI2021

The differences between mass- and light-derived structural parameters over time for MaNGA elliptical galaxies

ABSTRACT We apply stellar population synthesis analysis to obtain spatially resolved archaeological inferences for a large sample of ‘red and dead’ Elliptical galaxies (Classical Ellipticals, CLEs) from the MaNGA/SDSS-IV DR15 survey. From their 2D stellar light and mass maps, we explore the differences between the radial mass and light distributions in the rest-frame bands g, r, and i as functions of look-back time, tlb, or redshift, z. We characterize these differences through the ratios between the following mass- and light-derived global properties: sizes, concentrations, and effective surface densities. We find that the mass-to-light ratios of these properties change with tlb, more the more massive the galaxies are. The CLE galaxy archaeological progenitors are, on average, less compact, concentrated, and dense in light than in mass as z decreases. However, at later times, when also the evolution of the progenitors becomes passive at all radii, there is an upturn in these trends and the differences between mass and light in compactness/concentration decrease towards z ∼ 0. The trends in the ratios of mass-to-light sizes agree qualitatively with results from direct observations in galaxy surveys at different redshifts. We discuss the caveats and interpretations of our results, and speculate that the strong structural evolution found in some previous studies for early-type galaxies could be explained partially by photometric changes rather than by intrinsic structural changes.

Ibarra-Medel, H. (ORCID:0000000297906313)↗
DOE OSTI2024

ExaWorks software development kit: a robust and scalable collection of interoperable workflows technologies

Scientific discovery increasingly requires executing heterogeneous scientific workflows on high-performance computing (HPC) platforms. Heterogeneous workflows contain different types of tasks (e.g., simulation, analysis, and learning) that need to be mapped, scheduled, and launched on different computing. That requires a software stack that enables users to code their workflows and automate resource management and workflow execution. Currently, there are many workflow technologies with diverse levels of robustness and capabilities, and users face difficult choices of software that can effectively and efficiently support their use cases on HPC machines, especially when considering the latest exascale platforms. We contributed to addressing this issue by developing the ExaWorks Software Development Kit (SDK). The SDK is a curated collection of workflow technologies engineered following current best practices and specifically designed to work on HPC platforms. We present our experience with (1) curating those technologies, (2) integrating them to provide users with new capabilities, (3) developing a continuous integration platform to test the SDK on DOE HPC platforms, (4) designing a dashboard to publish the results of those tests, and (5) devising an innovative documentation platform to help users to use those technologies. Our experience details the requirements and the best practices needed to curate workflow technologies, and it also serves as a blueprint for the capabilities and services that DOE will have to offer to support a variety of scientific heterogeneous workflows on the newly available exascale HPC platforms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Particle Flow Distribution in a Fluidized-Particles Multitube Solar Receiver

A fluidized-particles two-tube solar receiver was tested at ambient temperature at PROMES Laboratory to investigate the influence of an inhomogeneous solar flux density on the particle mass flow rate between the tubes. The principle of this 3rd generation solar receiver is to fluidize the particles in a container, called dispenser, in which the tubes’ bottom are immersed. The fluidized particles are flowing upward the tubes by applying an overpressure in the dispenser. Air velocities are changed inside the tubes, thanks to air mass flow controllers, to represent temperature heterogeneity between the tubes. Air velocities from 0.05 up to 0.52 m/s were tested, both in homogeneous and heterogeneous conditions. In the heterogeneous ones, the differences in air velocity between the tubes aim to mimic a difference in temperature from 20 to 100 % with homogeneous air flow rates injected. The following conclusions were drawn. First, the particle mass flux in the tubes are the same with homogeneous air velocities, each one following a calibration map previously obtained. Second, different air velocities lead to different particle mass flux. Third, the rise of the total particle mass flux diminishes the pressure in the dispenser. Four, this diminution of pressure leads to a decrease of the particle mass flow rate of the tube with the lower air velocity. This influence can lead in some cases to the stop of the fluidized bed circulation in the affected tube.

Sahuquet, Guillaume (ORCID:0000000302752890)↗
NASA NTRS1977

Land use and land cover mapping: City of Palm Bay, Florida

Two different computer systems were compared for use in making land use and land cover maps. The Honeywell 635 with the LANDSAT signature development program (LSDP) produced a map depicting general patterns, but themes were difficult to classify as specific land use. Urban areas were unclassified. The General Electric Image 100 produced a map depicting eight land cover categories classifying 68 percent of the total area. Ground truth, LSDP, and Image 100 maps were all made to the same scale for comparison. LSDP agreed with the ground truth 60 percent and 64 percent within the two test areas compared and Image 100 was in agreement 70 percent and 80 percent.

Barile, D. D.↗
NASA NTRS1999

The Distribution of Titanium in Lunar Soils on the Basis of Sensor and In Situ Data Fusion

A variety of remote-sensing measurements have been used to map the distribution of elements on the Moon as a means of providing constraints on the processes from which its crust and major terranes originated. Discussed here is Ti, which is incorporated into refractory minerals such as ilmenite during the latter stages of differentiation, and is thus a most useful element for understanding mare basalt petrogenesis. One of the earliest Ti maps showed Ti variations in nearside maria on the basis of groundbased spectral reflectance measurements. A map of Ti derived from gamma-ray measurements on Apollo 15 and 16 was produced at about the same time, and was improved upon considerably by Davis and coworkers, who effectively removed sources of spurious variation from Fe and Al or REE (e.g., Th) interference, and calibrated Ti on the bases of landing-site soil averages. In recent years, spectral reflectance measurements from Clementine have been used by Lucey and coworkers to produce global Ti distribution maps as well. As we indicated previously, the Lucey and Davis maps agree to first order. Meanwhile, we are using the concept of sensor data fusion to combine measurements from the AGR (Apollo gamma-ray) and CSR (Clementine Spectral Reflectance) techniques with ground truth from lunar soils to utilize the differences between the two maps to understand the distribution of Ti within lunar soil components, as we have done with Fe. This technique should be verified and applied on Lunar Prospector gamma-ray measurements of Ti, as the calibrated data become available within the next couple of years. Lunar Ti is found principally in the mineral ilmenite, and is associated with certain components of lunar soil: crystalline Ilmenite mineral fragments and high Ti-bearing glass. All data indicate that Ti is associated with maria and mafic minerals. In AGR and CSR datasets, Ti is highest on the nearside and in the maria, particularly in southern Serenitatis/northern Tranquillitatis and northwestern Procellarum. Unlike CSR-derived values, AGR Ti values show modest increase (up to 0.7%) on the northern farside. Both techniques show a primarily unimodal distribution with a shoulder in the high Ti direction and a primary mode at approximately 0.2% (0.15% for CSR and 0.25% for AGR data), but the AGR Ti data have more structure in the shoulder, including an apparent minor mode at 1.2% Ti, representing non-highland areas. Recalibrating the two datasets on the basis of matching the peaks and range on the histograms would not account for this additional structure in the AGR data. Low Ti areas, which occur predominantly on the farside, are not represented well in the lunar sample collection, although several meteorites thought to be of lunar highland origin show Ti abundances averaging about 0.2%. (Error bars on AGR Ti values are about 0.5%.) Gamma-ray measurements reflect intrinsic Ti surface composition, regardless of the physical or chemical state of Ti. Lucey and coworkers have attempted to produce an equivalent bulk Ti map by normalizing the spectral feature at 415 nm to remove the effects of physical variations from soil to soil. The normalization is based on laboratory measurements of available samples, primarily nearside maria, and is undoubtedly optimized for the proportions in which Ti- components are found in these soils. In particular, mare soils tend to have a much higher proportion of opaque mineral grains, which appears to be well correlated with Ti abundance. The relationship between Ti abundance and proportion of Ti-bearing glass (weighted for Ti abundance) is not as direct. Also shows the relationship between landing site soil average, AGR, and CSR Ti abundances (with CSR filtered to match resolution and field of view of AGR data). AGR Ti values show the best agreement with Ti soil averages, CSR Ti values being somewhat higher in high-Ti soils, and lower in low-Ti soils. These results would tend to support the argument that the technique for deriving Ti from the CSR measurements is optimized for one of the primary Ti-bearing components, most likely opaque mineral grains, and is thus over-compensating and reporting higher Ti values in mare areas, and underreporting Ti in highland areas with much lower proportions of opaque mineral grains. More information is contained in original.

Clark, P. E.↗
DOE OSTI2020

Accounting for Training Data Error in Machine Learning Applied to Earth Observations

Remote sensing, or Earth Observation (EO), is increasingly used to understand Earth system dynamics and create continuous and categorical maps of biophysical properties and land cover, especially based on recent advances in machine learning (ML). ML models typically require large, spatially explicit training datasets to make accurate predictions. Training data (TD) are typically generated by digitizing polygons on high spatial-resolution imagery, by collecting in situ data, or by using pre-existing datasets. TD are often assumed to accurately represent the truth, but in practice almost always have error, stemming from (1) sample design, and (2) sample collection errors. The latter is particularly relevant for image-interpreted TD, an increasingly commonly used method due to its practicality and the increasing training sample size requirements of modern ML algorithms. TD errors can cause substantial errors in the maps created using ML algorithms, which may impact map use and interpretation. Despite these potential errors and their real-world consequences for map-based decisions, TD error is often not accounted for or reported in EO research. Here we review the current practices for collecting and handling TD. We identify the sources of TD error, and illustrate their impacts using several case studies representing different EO applications (infrastructure mapping, global surface flux estimates, and agricultural monitoring), and provide guidelines for minimizing and accounting for TD errors. To harmonize terminology, we distinguish TD from three other classes of data that should be used to create and assess ML models: training reference data, used to assess the quality of TD during data generation; validation data, used to iteratively improve models; and map reference data, used only for final accuracy assessment. We focus primarily on TD, but our advice is generally applicable to all four classes, and we ground our review in established best practices for map accuracy assessment literature. EO researchers should start by determining the tolerable levels of map error and appropriate error metrics. Next, TD error should be minimized during sample design by choosing a representative spatio-temporal collection strategy, by using spatially and temporally relevant imagery and ancillary data sources during TD creation, and by selecting a set of legend definitions supported by the data. Furthermore, TD error can be minimized during the collection of individual samples by using consensus-based collection strategies, by directly comparing interpreted training observations against expert-generated training reference data to derive TD error metrics, and by providing image interpreters with thorough application-specific training. We strongly advise that TD error is incorporated in model outputs, either directly in bias and variance estimates or, at a minimum, by documenting the sources and implications of error. TD should be fully documented and made available via an open TD repository, allowing others to replicate and assess its use. To guide researchers in this process, we propose three tiers of TD error accounting standards. Finally, we advise researchers to clearly communicate the magnitude and impacts of TD error on map outputs, with specific consideration given to the likely map audience.

58 GEOSCIENCES↗
NASA NTRS2005

X-Ray Fluorescence Solvent Detection at the Substrate-Adhesive Interface

With environmental regulations limiting the use of volatile organic compounds, low-vapor pressure solvents have replaced traditional degreasing solvents for bond substrate preparation. When used to clean and prepare porous bond substrates such as phenolic composites, low vapor pressure solvents can penetrate deep into substrate pore networks and remain there for extended periods. Trapped solvents can interact with applied adhesives either prior to or during cure, potentially compromising bond properties. Currently, methods for characterizing solvent time-depth profiles in bond substrates are limited to bulk gravimetric or sectioning techniques. While sectioning techniques such as microtome allow construction of solvent depth profiles, their depth resolution and reliability are limited by substrate type. Sectioning techniques are particularly limited near the adhesive-substrate interface where depth resolution is further limited by adhesive-substrate hardness and, in the case of a partially cured adhesive, mechanical properties differences. Additionally, sectioning techniques cannot provide information about lateral solvent diffusion. Cross-section component mapping is an alternative method for measuring solvent migration in porous substrates that eliminates the issues associated with sectioning techniques. With cross-section mapping, the solvent-wiped substrate is sectioned perpendicular rather than parallel to the wiped surface, and the sectioned surface is analyzed for the solvent or solvent components of interest using a two-dimensional mapping or imaging technique. Solvent mapping can be performed using either direct or indirect methods. With a direct method, one or more solvent components are mapped using red or Raman spectroscopy together with a moveable sample stage and/or focal plane array detector. With an indirect method, an elemental "tag" not present in the substrate is added to the solvent before the substrate is wiped. Following cross sectioning, the tag element can then be mapped by its characteristic x-ray emission using either x-ray fluorescence, or electron-beam energy-and wavelength-dispersive x-ray spectrometry. The direct mapping techniques avoid issues of different diffusion or migration rates of solvents and elemental tags, while the indirect techniques avoid spectral resolution issues in cases where solvents and substrates have adjacent or overlapping peaks. In this study, cross-section component indirect mapping is being evaluated as a method for measuring migration of d-limonene based solvents in glass-cloth phenolic composite (GCP) prior to and during subsequent bonding and epoxy adhesive cure.

Wurth, Laura↗
NASA NTRS1986

Multipolarization radar images for geologic mapping and vegetation discrimination

NASA has developed an airborne SAR that simultaneously yields image data in four linear polarizations in L-band with 10-m resolution over a swath of about 10 km. Signal data are recorded both optically and digitally and annotated in each of the channels to facilitate completely automated digital correlation. Comparison of the relative intensities of the different polarizations furnishes discriminatory mapping information. Local intensity variations in like-polarization images result from topographic effects, while strong cross polarization responses denote the effects of vegetation cover and, in some cases, possible scattering from the subsurface. In each of the areas studied, multiple polarization data led to the discrimination and mapping of unique surface unit features.

Evans, D. L.↗
NASA NTRS2001

A Two-Wheel Observing Mode for the MAP Spacecraft

The Microwave Anisotropy Probe (MAP) is a follow-on to the Differential Microwave Radiometer (DMR) instrument on the Cosmic Background Explorer (COBE). Due to the MAP project's limited mass, power, and budget, a traditional reliability concept including fully redundant components was not feasible. The MAP design employs selective hardware redundancy, along with backup software modes and algorithms, to improve the odds of mission success. This paper describes the effort to develop a backup control mode, known as Observing II, that will allow the MAP science mission to continue in the event of a failure of one of its three reaction wheel assemblies. This backup science mode requires a change from MAP's nominal zero-momentum control system to a momentum-bias system. In this system, existing thruster-based control modes are used to establish a momentum bias about the sun line sufficient to spin the spacecraft up to the desired scan rate. Natural spacecraft dynamics exhibits spin and nutation similar to the nominal MAP science mode with different relative rotation rates, so the two reaction wheels are used to establish and maintain the desired nutation angle from the sun line. Detailed descriptions of the ObservingII control algorithm and simulation results will be presented, along with the operational considerations of performing the rest of MAP's necessary functions with only two wheels.

Starin, Scott R.↗
NASA NTRS1988

A generic fine-grained parallel C

With the present availability of parallel processors of vastly different architectures, there is a need for a common language interface to multiple types of machines. The parallel C compiler, currently under development, is intended to be such a language. This language is based on the belief that an algorithm designed around fine-grained parallelism can be mapped relatively easily to different parallel architectures, since a large percentage of the parallelism has been identified. The compiler generates a FORTH-like machine-independent intermediate code. A machine-dependent translator will reside on each machine to generate the appropriate executable code, taking advantage of the particular architectures. The goal of this project is to allow a user to run the same program on such machines as the Massively Parallel Processor, the CRAY, the Connection Machine, and the CYBER 205 as well as serial machines such as VAXes, Macintoshes and Sun workstations.

Hamet, L.↗
NASA NTRS1975

Approaches to digital snow mapping with LANDSAT-1 data

Applying the same LANDSAT-1 data to three substantially different image processing systems, a snow mapping task was performed. LARSYS Ver.3, STANSORT-2, and General Electric Image-100 did all the jobs of detecting the snowline in forested mountainous terrain, and to determine the snowcovered area. While the control and accuracy achieved with LARSYS is remarkable, time and effort to perform the processing favor the systems STANSORT and Image-100. The experiences and results demonstrate the need for a fast interactive system for operational snowmapping with multispectral satellite data.

Itten, K. I.↗
DOE OSTI2024

Performance evaluation of the USGS velocity model for the San Francisco Bay Area

In this study, we evaluated the performance of the United States Geological Survey velocity model developed for the San Francisco Bay Area (SFBA), version 21.1. The evaluation was performed through high-resolution three-dimensional physics-based ground motion simulations of seven small-magnitude earthquakes (ranging from magnitude 3.8 to 4.4) that occurred on the eastern side of the San Francisco Bay. The simulations were performed in the frequency range from 0 to 5 Hz with a minimum shear-wave velocity of 250 m/s, which allowed the capture of wave propagation effects of the near-surface soft materials that characterize local basins. Based on the direct comparison of Fourier amplitude spectra between recorded and simulated ground motions for more than 250 stations, we found that the velocity model generally performs well in the frequency range of 0.2–5 Hz. The median value of the Fourier amplitude residuals was found to be near zero for all seven earthquakes. The slight over-prediction of 0.2 log-natural units at frequencies above 3 Hz in our simulations was attributed to the potentially inaccurate representation of the source radiation pattern by a double-couple point source model, and simple representation of shallow small-scale underground structural complexity in the velocity model. Maps of spectral amplitude differences between the simulated and recorded data were used to identify areas responsible for systematic ground motion over-predictions or under-predictions. For example, while some sub-domains over soft sediments show over-prediction patterns, the block east of the Hayward fault is prone to exhibit patterns of under-prediction. These maps can be used to guide future refinements of the SFBA velocity model. Since our simulation methodology allows for the decoupling of the source and wave propagation effects, the ground motion data generated by our simulations can also be used to quantify the epistemic uncertainty due to the velocity model, in empirically based ground motion estimates for the SFBA.

58 GEOSCIENCES↗
DOE OSTI2022

Characterizing vertical upper ocean temperature structures in the European Arctic through unsupervised machine learning

In-situ observations of subsurface ocean temperatures are, in many regions, inconsistently distributed in time and space. These spatio-temporal inconsistencies in the observational network lead to difficulties in utilizing those observations effectively for ocean model evaluation or understanding larger-scale ocean characteristics. Model accuracy of subsurface ocean characteristics is especially important within regions that contain complex ocean structures. One such region is the European Arctic which not only contains several types of water masses with unique characteristics, but also wintertime sea ice coverage and complex bathymetry. This study presents an unsupervised neural networking technique that can be used in combination with traditional ocean model evaluation techniques to provide additional information on the accuracy of modeled vertical ocean temperature profiles. Self-organizing maps is an unsupervised machine learning technique that we apply to approximately twenty thousand Argo and CTD temperature profiles from 2012 to 2020 in the European Arctic to categorize the observed vertical ocean temperature structures in the top 150 m. The observed ocean profile categories, or neurons, defined by the self-organizing map show strong spatial and temporal dependencies. We then use the neuron weights, or the learned temperature profile structure of each neuron, to validate the spatial and temporal variability of modeled vertical temperature structures. This analysis gives us new insights about the model’s capabilities to reproduce specific vertical structures of the top-most ocean layer within different regions and seasons. Mapping modeled ocean temperature profiles onto the neuron-space of the observationally-defined self organized map highlights the potential of this method to advance our understanding of model deficiencies in that region.

54 ENVIRONMENTAL SCIENCES↗