SEARCH · Engineering Papers

Results for “VED”

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.

SourceResearch connectionFrom yearThrough year

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

At least 55 records · Page 3

All sources NASA NTRS DOE OSTI

DOE OSTI2017

Photoconductive detector of circularly polarized radiation based on a MIS structure with a CoPt layer

A photoconductive detector of circularly polarized radiation based on the metal–insulator–semiconductor structure of CoPt/(Al{sub 2}O{sub 3}/SiO{sub 2}/Al{sub 2}O{sub 3})/InGaAs/GaAs is created. The efficiency of detection of circularly polarized radiation is 0.75% at room temperature. The operation of the detector is based on the manifestation of the effect of magnetic circular dichroism in the CoPt layer, that is, the dependence of the CoPt transmission coefficient on the sign of the circular polarization of light and magnetization.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2017

Control of circular polarization of electroluminescence in spin light-emitting diodes based on InGaAs/GaAs/δ〈Mn〉 heterostructures

Circularly polarized luminescence of light-emitting InGaAs/GaAs structures with a delta-doped Mn layer in a GaAs barrier was studied. The structural parameters were varied by different ways, among them are homogeneous and delta-doping with acceptor impurity, and removal of donor doping from the technological process. As it was found, the magnitude and polarity of the degree of circular polarization of luminescence strongly depend on the technological mode chosen. Simultaneous modeling of wave functions of structures highlights a good agreement between the parameters of circularly polarized luminescence and spatial distribution of wave functions of heavy holes relative to the Mn delta-layer.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2018

Raising the Operating Temperature of (Ga,Mn)As/GaAs Spin Light Emitting Diodes by Applying Post-Growth Treatment

Abstract—Spin light emitting diodes (spin-LEDs) containing (Ga,Mn)As ferromagnetic layers are fabricated based on InGaAs/GaAs heterostructures and studied. We achieve increases in the operating temperatures of our spin-LEDs by subjecting the surface of the structures, prior to depositing an ohmic metal contact, to pulsed laser annealing. The fabricated devices produce circularly polarized electroluminescence when placed in an external magnetic field. The temperatures at which circularly polarized electroluminescence is still observed is raised from 30 K for unprocessed structure to 110 K for laser-annealed structures. The observed effect is linked to an increase in the Curie temperature of the (Ga,Mn)As layer as a result of laser impact.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2022

Cubit v.16.X

SAND2021-3051 O CUBIT is a full-featured software toolkit for robust generation of two- and three-dimensional finite element meshes (grids) and geometry preparation. Its main goal is to reduce the time required to generate meshes - particularly large hex meshes of complicated, interlocking assemblies. It is a solid-modeler-based preprocessor that meshes volumes and surfaces for finite element analysis. CUBIT also includes state-of-the-art smoothing algorithms and provides an extensive suite of tools for geometry decomposition and mesh generation. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Stimpson, Clint↗

DOE OSTI2017

Solid-State Thyratron Replacement. Final Report

Under this SBIR, DTI developed a solid-state switch as an alternative to legacy thyratron equipment. Our Phase II objective was to make a solid-state thyratron replacement that would provide equivalent or better performance, much higher reliability (at least a 20 year lifetime, compared to a thyratron’s two-year lifetime) and would sell for ~3x the cost of a thyratron, or less than $40k. We were successful in building a solid-state switch which could reliably function as a thyratron replacement. The unit was designed to directly replace the thyratrons currently being used at SLAC’s Linac Coherent Light Source (LCLS), and was built in a tank that was small enough to fit into the existing thyratron cabinet, providing a true form-fit-function replacement path. We tested the switch at the full operating specifications: 48 kV, 6.3 kA, and 1 µs risetime. We also demonstrated a peak-to-peak pulse jitter of 1.5 ns, which is five times shorter than is typical for thyratrons. This lower jitter would improve the performance of the LCLS beam. The predicted reliability is more than 80 years, which is 40 times greater than a thyratron.

47 OTHER INSTRUMENTATION↗

DOE OSTI2018

Polyalkylene Glycol (PAG) Based Lubricant for Light & Medium Duty Axles

The axle lubricant (SAE 75W-140) widely used in the market place over the last few years is primarily formulated from polyalpha olefin (PAO) basestock. This investigation focused on polyalkylene glycol (PAG) base stock which is significantly different chemically from PAO and because of their polar nature it is hypothesized that it can adsorb relatively easily on contacting surfaces resulting in significant friction reduction and improve fuel economy. Axle efficiency is generally quite high at high torque range but it could be quite low at low torque range representative of EPA drive cycles. Therefore, this project primarily focus on improving axle efficiency in the low torque range while maintaining all durability attributes. The objective of this project is to develop novel lubricant formulations that are expected to improve the fuel efficiency of light, medium, heavy-duty, and military vehicles by at least 2% over Society of Automotive Engineers (SAE) 75W-140 axle lubricants (improvement based on comparative results from engine dynamometer testing, chassis dynamometer testing or test track, e.g., SAE J1321) without adverse impacts on vehicle performance or durability. The proposed research project followed a structured approach beginning with lubricant formulations by varying base oil chemistry and additive components followed by physical property characterization. Lubricant performance evaluations started with simple laboratory friction, wear, oxidation, and corrosion tests which guided into identification of performance areas requiring improvements. This required an iterative lubricant reformulation. Once a few acceptable formulations were identified, they are evaluated in more complex component tests including ASTM tests (L-37 - load carrying capacity under low speed and high torque, L42 – load carrying properties under high speed and shock loading, L-60 - thermal and oxidative stability, and L-33-1 moisture corrosion resistance). Two issues were observed following L-42 tests; toxicological and foaming. This required reformulation and repeating all ASTM tests. During repeat tests it was observed that some of the additives precipitated out of solution. Since it was discovered quite late in the program, and the time it may need to find a solution to this issue, the team changed direction of the program to switch to use of an oil soluble PAG as a co-basestock (15-30%) with mineral oil used in a commercial formulation (SAE 75W-85). This formulation strategy helped passing all ASTM tests. This was followed by Ford proprietary axle efficiency, gear wear, and chassis roll dynamometer tests for fuel economy evaluations. Axle efficiency was improved up to 7% when a full PAG formulation was used but no such significant improvement was observed with PAG as co-basestock formulation. Chassis roll fuel economy improved by 1.9% with PAG co-basestock formulation (conforming to SAE 75W-85) compared to SAE 75W-140 baseline. However, it was observed that the improvement is mostly due to reduced viscosity of candidate formulation when compared to a full mineral oil based SAE 75W-85 formulation. PAG as co-basestock also exhibited unacceptable wear in proprietary gear wear test. This investigation highlighted challenges with PAG formulations including identification of additive components that stay in PAG solution for intended duration, and meeting component durability requirement.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Characterizing Integrase-Attachment Site Pairs: A Machine Learning Approach

Genomic islands (GIs) are mobile genetic elements that integrate into host genomes via self-encoded integrases at specific DNA sequences known as attachment (att) sites. The ability to predict the target att site from an integrase's protein sequence is a central challenge in genomics due to the sequence diversity of integrases and the subtlety of their DNA recognition motifs.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2026

Opportunities in AI/ML for the Rubin LSST Dark Energy Science Collaboration

The Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) will produce unprecedented volumes of heterogeneous astronomical data (images, catalogs, and alerts) that challenge traditional analysis pipelines. The LSST Dark Energy Science Collaboration (DESC) aims to derive robust constraints on dark energy and dark matter from these data, requiring methods that are statistically powerful, scalable, and operationally reliable. Artificial intelligence and machine learning (AI/ML) are already embedded across DESC science workflows, from photometric redshifts and transient classification to weak lensing inference and cosmological simulations. Yet their utility for precision cosmology hinges on trustworthy uncertainty quantification, robustness to covariate shift and model misspecification, and reproducible integration within scientific pipelines. This white paper surveys the current landscape of AI/ML across DESC's primary cosmological probes and cross-cutting analyses, revealing that the same core methodologies and fundamental challenges recur across disparate science cases. Since progress on these cross-cutting challenges would benefit multiple probes simultaneously, we identify key methodological research priorities, including Bayesian inference at scale, physics-informed methods, validation frameworks, and active learning for discovery. With an eye on emerging techniques, we also explore the potential of the latest foundation model methodologies and LLM-driven agentic AI systems to reshape DESC workflows, provided their deployment is coupled with rigorous evaluation and governance. Finally, we discuss critical software, computing, data infrastructure, and human capital requirements for the successful deployment of these new methodologies, and consider associated risks and opportunities for broader coordination with external actors.

Aubourg, Eric [APC, Paris] (ORCID:000000025592023X↗

NASA NTRS2016

Automated Segmentation and Classification of Coral using Fluid Lensing from Unmanned Airborne Platforms

In recent years, there has been a growing interest among biologists in monitoring the short and long term health of the world's coral reefs. The environmental impact of climate change poses a growing threat to these biologically diverse and fragile ecosystems, prompting scientists to use remote sensing platforms and computer vision algorithms to analyze shallow marine systems. In this study, we present a novel method for performing coral segmentation and classification from aerial data collected from small unmanned aerial vehicles (sUAV). Our method uses Fluid Lensing algorithms to remove and exploit strong optical distortions created along the air-fluid boundary to produce cm-scale resolution imagery of the ocean floor at depths up to 5 meters. A 3D model of the reef is reconstructed using structure from motion (SFM) algorithms, and the associated depth information is combined with multidimensional maximum a posteriori (MAP) estimation to separate organic from inorganic material and classify coral morphologies in the Fluid-Lensed transects. In this study, MAP estimation is performed using a set of manually classified 100 x 100 pixel training images to determine the most probable coral classification within an interrogated region of interest. Aerial footage of a coral reef was captured off the coast of American Samoa and used to test our proposed method. 90 x 20 meter transects of the Samoan coastline undergo automated classification and are manually segmented by a marine biologist for comparison, leading to success rates as high as 85%. This method has broad applications for coastal remote sensing, and will provide marine biologists access to large swaths of high resolution, segmented coral imagery.

algorithms segments classifications↗

NASA NTRS2017

Fluid Lensing and Applications to Remote Sensing of Aquatic Environments

The use of fluid lensing technology on UAVs is presented as a novel means for 3D imaging of aquatic ecosystems from above the water's surface at the centimeter scale. Preliminary results are presented from airborne fluid lensing campaigns conducted over the coral reefs of Ofu Island, American Samoa (2013) and the stromatolite reefs of Shark Bay, Western Australia (2014), covering a combined area of 15km2. These reef ecosystems were revealed with centimetre-scale 2D resolution, and an accompanying 3D bathymetry model was derived using fluid lensing, Structure from Motion and UAV position data. Data products were validated from in-situ survey methods including underwater calibration targets, depth measurements and millimetre-scale high-dynamic range gigapixel photogrammetry. Fluid lensing is an experimental technology that uses water transmitting wavelengths to passively image underwater objects at high-resolution by exploiting time-varying optical lensing events caused by surface waves. Fluid lensing data are captured from low-altitude, cost-effective electric UAVs to achieve multispectral imagery and bathymetry models at the centimetre scale over regional areas. As a passive system, fluid lensing is presently limited by signal-to-noise ratio and water column inherent optical properties to approximately 10 m depth over visible wavelengths in clear waters. The datasets derived from fluid lensing present the first centimetre-scale images of a reef acquired from above the ocean surface, without wave distortion. The 3D multispectral data distinguish coral, fish and invertebrates in American Samoa, and reveal previously undocumented, morphologically distinct, stromatolite structures in Shark Bay. These findings suggest fluid lensing and multirotor electric drones represent a promising advance in the remote sensing of aquatic environments at the centimetre scale, or 'reef scale' relevant to the conservation of reef ecosystems. Pending further development and validation of fluid lensing methods, these technologies present a solution for large-scale 3D surveys of shallow aquatic habitats with centimetre-scale spatial resolution and hourly temporal sampling.

Aquatic↗

NASA NTRS2017

NeMO-Net: The Neural Multi-Modal Observation and Training Network for Global Coral Reef Assessment

In the past decade, coral reefs worldwide have experienced unprecedented stresses due to climate change, ocean acidification, and anthropomorphic pressures, instigating massive bleaching and die-off of these fragile and diverse ecosystems. Furthermore, remote sensing of these shallow marine habitats is hindered by ocean wave distortion, refraction and optical attenuation, leading invariably to data products that are often of low resolution and signal-to-noise (SNR) ratio. However, recent advances in UAV and Fluid Lensing technology have allowed us to capture multispectral 3D imagery of these systems at sub-cm scales from above the water surface, giving us an unprecedented view of their growth and decay. Exploiting the fine-scaled features of these datasets, machine learning methods such as MAP, PCA, and SVM can not only accurately classify the living cover and morphology of these reef systems (below 8 percent error), but are also able to map the spectral space between airborne and satellite imagery, augmenting and improving the classification accuracy of previously low-resolution datasets. We are currently implementing NeMO-Net, the first open-source deep convolutional neural network (CNN) and interactive active learning and training software to accurately assess the present and past dynamics of coral reef ecosystems through determination of percent living cover and morphology. NeMO-Net will be built upon the QGIS platform to ingest UAV, airborne and satellite datasets from various sources and sensor capabilities, and through data-fusion determine the coral reef ecosystem makeup globally at unprecedented spatial and temporal scales. To achieve this, we will exploit virtual data augmentation, the use of semi-supervised learning, and active learning through a tablet platform allowing for users to manually train uncertain or difficult to classify datasets. The project will make use of Pythons extensive libraries for machine learning, as well as extending integration to GPU and High-End Computing Capability (HECC) on the Pleiades supercomputing cluster, located at NASA Ames. The project is being supported by NASAs Earth Science Technology Office (ESTO) Advanced Information Systems Technology (AIST-16) Program.

NeMO-Net↗

NASA NTRS2017

A 3D Active Learning Application for NeMO-Net, the NASA Neural Multi-Modal Observation and Training Network for Global Coral Reef Assessment

NeMO-Net, the NASA neural multi-modal observation and training network for global coral reef assessment, is an open-source deep convolutional neural network and interactive active learning training software aiming to accurately assess the present and past dynamics of coral reef ecosystems through determination of percent living cover and morphology as well as mapping of spatial distribution. We present an interactive video game prototype for tablet and mobile devices where users interactively label morphology classifications over mm-scale 3D coral reef imagery captured using fluid lensing to create a dataset that will be used to train NeMO-Nets convolutional neural network. The application currently allows for users to classify preselected regions of coral in the Pacific and will be expanded to include additional regions captured using our NASA FluidCam instrument, presently the highest-resolution remote sensing benthic imaging technology capable of removing ocean wave distortion, as well as lower-resolution airborne remote sensing data from the ongoing NASA CORAL campaign. Active learning applications present a novel methodology for efficiently training large-scale Neural Networks wherein variances in identification can be rapidly mitigated against control data. NeMO-Net periodically checks users input against pre-classified coral imagery to gauge their accuracy and utilize in-game mechanics to provide classification training. Users actively communicate with a server and are requested to classify areas of coral for which other users had conflicting classifications and contribute their input to a larger database for ranking. In partnering with Mission Blue and IUCN, NeMO-Net leverages an international consortium of subject matter experts to classify areas of confusion identified by NeMO-Net and generate additional labels crucial for identifying decision boundary locations in coral reef assessment.

NeMO=NET↗

NASA NTRS2017

NeMO-Net The Neural Multi-Modal Observation Training Network for Global Coral Reef Assessment

In the past decade, coral reefs worldwide have experienced unprecedented stresses due to climate change, ocean acidification, and anthropomorphic pressures, instigating massive bleaching and die-off of these fragile and diverse ecosystems. Furthermore, remote sensing of these shallow marine habitats is hindered by ocean wave distortion, refraction and optical attenuation, leading invariably to data products that are often of low resolution and signal-to-noise (SNR) ratio. However, recent advances in UAV and Fluid Lensing technology have allowed us to capture multispectral 3D imagery of these systems at sub-cm scales from above the water surface, giving us an unprecedented view of their growth and decay. Exploiting the fine-scaled features of these datasets, machine learning methods such as MAP, PCA, and SVM can not only accurately classify the living cover and morphology of these reef systems (below 8 error), but are also able to map the spectral space between airborne and satellite imagery, augmenting and improving the classification accuracy of previously low-resolution datasets.We are currently implementing NeMO-Net, the first open-source deep convolutional neural network (CNN) and interactive active learning and training software to accurately assess the present and past dynamics of coral reef ecosystems through determination of percent living cover and morphology. NeMO-Net will be built upon the QGIS platform to ingest UAV, airborne and satellite datasets from various sources and sensor capabilities, and through data-fusion determine the coral reef ecosystem makeup globally at unprecedented spatial and temporal scales. To achieve this, we will exploit virtual data augmentation, the use of semi-supervised learning, and active learning through a tablet platform allowing for users to manually train uncertain or difficult to classify datasets. The project will make use of Pythons extensive libraries for machine learning, as well as extending integration to GPU and High-End Computing Capability (HECC) on the Pleiades supercomputing cluster, located at NASA Ames. The project is being supported by NASAs Earth Science Technology Office (ESTO) Advanced Information Systems Technology (AIST-16) Program.

Remote Sensin↗

NASA NTRS2018

NASA Fluid Lensing & MiDAR: Next-Generation Remote Sensing Technologies for Aquatic Remote Sensing

We present two recent instrument technology developments at NASA, Fluid Lensing and MiDAR, and their application to remote sensing of Earth's aquatic systems. Fluid Lensing is the first remote sensing technology capable of imaging through ocean waves in 3D at sub-cm resolutions. MiDAR is a next-generation active hyperspectral remote sensing and optical communications instrument capable of active fluid lensing. Fluid Lensing has been used to provide 3D multispectral imagery of shallow marine systems from unmanned aerial vehicles (UAVs, or drones), including coral reefs in American Samoa and stromatolite reefs in Hamelin Pool, Western Australia. MiDAR is being deployed on aircraft and underwater remotely operated vehicles (ROVs) to enable a new method for remote sensing of living and nonliving structures in extreme environments. MiDAR images targets with high-intensity narrowband structured optical radiation to measure an objectâ€"TM"s non-linear spectral reflectance, image through fluid interfaces such as ocean waves with active fluid lensing, and simultaneously transmit high-bandwidth data. As an active instrument, MiDAR is capable of remotely sensing reflectance at the centimeter (cm) spatial scale with a signal-to-noise ratio (SNR) multiple orders of magnitude higher than passive airborne and spaceborne remote sensing systems with significantly reduced integration time. This allows for rapid video-frame-rate hyperspectral sensing into the far ultraviolet and VNIR wavelengths. Previously, MiDAR was developed into a TRL 2 laboratory instrument capable of imaging in thirty-two narrowband channels across the VNIR spectrum (400-950nm). Recently, MiDAR UV was raised to TRL4 and expanded to include five ultraviolet bands from 280-400nm, permitting UV remote sensing capabilities in UV A, B, and C bands and enabling mineral identification and stimulated fluorescence measurements of organic proteins and compounds, such as green fluorescent proteins in terrestrial and aquatic organics.

fluid lensing↗

NASA NTRS2018

NeMO-Net & Fluid Lensing: The Neural Multi-Modal Observation & Training Network for Global Coral Reef Assessment Using Fluid Lensing Augmentation of NASA EOS Data

We present preliminary results from NASA NeMO-Net, the first neural multi-modal observation and training network for global coral reef assessment. NeMO-Net is an open-source deep convolutional neural network (CNN) and interactive active learning training software in development which will assess the present and past dynamics of coral reef ecosystems. NeMO-Net exploits active learning and data fusion of mm-scale remotely sensed 3D images of coral reefs captured using fluid lensing with the NASA FluidCam instrument, presently the highest-resolution remote sensing benthic imaging technology capable of removing ocean wave distortion, as well as hyperspectral airborne remote sensing data from the ongoing NASA CORAL mission and lower-resolution satellite data to determine coral reef ecosystem makeup globally at unprecedented spatial and temporal scales. Aquatic ecosystems, particularly coral reefs, remain quantitatively misrepresented by low-resolution remote sensing as a result of refractive distortion from ocean waves, optical attenuation, and remoteness. Machine learning classification of coral reefs using FluidCam mm-scale 3D data show that present satellite and airborne remote sensing techniques poorly characterize coral reef percent living cover, morphology type, and species breakdown at the mm, cm, and meter scales. Indeed, current global assessments of coral reef cover and morphology classification based on km-scale satellite data alone can suffer from segmentation errors greater than 40%, capable of change detection only on yearly temporal scales and decameter spatial scales, significantly hindering our understanding of patterns and processes in marine biodiversity at a time when these ecosystems are experiencing unprecedented anthropogenic pressures, ocean acidification, and sea surface temperature rise. NeMO-Net leverages our augmented machine learning algorithm that demonstrates data fusion of regional FluidCam (mm, cm-scale) airborne remote sensing with global low-resolution (m, km-scale) airborne and spaceborne imagery to reduce classification errors up to 80% over regional scales. Such technologies can substantially enhance our ability to assess coral reef ecosystems dynamics.

satellite data↗

NASA NTRS2017

Weathering the Storm: Unmanned Aircraft Systems in the Maritime, Atmospheric and Polar Environments

Remotely piloted aircraft (RPA) have the potential to revolutionize local to regional data collection for geophysicists as platform and payload size decrease while aircraft capabilities increase. In particular, data from RPAs combine high-resolution imagery available from low flight elevations with comprehensive areal coverage, unattainable from ground investigations and difficult to acquire from manned aircraft due to budgetary and logistical costs. Low flight elevations are particularly important for detecting signals that decay exponentially with distance, such as electromagnetic fields. Onboard data processing coupled with high-bandwidth telemetry open up opportunities for real-time and near real-time data processing, producing more efficient flight plans through the use of payload-directed flight, machine learning and autonomous systems. Such applications not only strive to enhance data collection, but also enable novel sensing modalities and temporal resolution. NASAs Airborne Science Program has been refining the capabilities and applications of RPA in support of satellite calibration and data product validation for several decades. In this paper, we describe current platforms, payloads, and onboard data systems available to the research community. Case studies include Fluid Lensing for littoral zone 3D mapping, structure from motion for terrestrial 3D multispectral imaging, and airborne magnetometry on medium and small RPAs.

data collection↗

NASA NTRS2018

NASA Fluid Lensing & MiDAR - Next-Generation Remote Sensing Technologies for Aquatic Remote Sensing

Piti's Tepungan Bay and Tumon Bay, two of five marine preserves in Guam, have not been mapped to a level of detail sufficient to support proposed management strategies. This project addresses this gap by providing high resolution maps to promote sustainable, responsible use of the area while protecting natural resources. Dr. Chirayath, a research scientist at the NASA Ames Laboratory, developed a theoretical model and algorithm called 'Fluid Lensing'. Fluid lensing removes optical distortions caused by moving water, improving the clarity of the images taken of the corals below the surface. We will also be using MiDAR, a next-generation remote sensing instrument that provides real-time multispectral video using an array of LED emitters coupled with NASA's FluidCam Imaging System, which may assist Guam's coral reef response team in understanding the severity and magnitude of coral bleaching events. This project will produce a 3D orthorectified model of the shallow water coral reef ecosystems in Tumon Bay and Piti marine preserves. These 3D models may be printed, creating a tactile diorama and increasing understanding of coral reefs among various audiences, including key decision makers. More importantly, the final data products can enable accurate and quantitative health assessment capabilities for coral reef ecosystems.

Aquatic↗

NASA NTRS2018

NASA NeMO-Net

We present preliminary results from NASA NeMO-Net, the first neural multi-modal observation and training network for global coral reef assessment. NeMO-Net is an open-source deep convolutional neural network (CNN) and interactive active learning training software in development which will assess the present and past dynamics of coral reef ecosystems. NeMO-Net exploits active learning and data fusion of mm-scale remotely sensed 3D images of coral reefs captured using fluid lensing with the NASA FluidCam instrument, presently the highest-resolution remote sensing benthic imaging technology capable of removing ocean wave distortion, as well as hyperspectral airborne remote sensing data from the ongoing NASA CORAL mission and lower-resolution satellite data to determine coral reef ecosystem makeup globally at unprecedented spatial and temporal scales. Aquatic ecosystems, particularly coral reefs, remain quantitatively misrepresented by low- resolution remote sensing as a result of refractive distortion from ocean waves, optical attenuation, and remoteness. Machine learning classification of coral reefs using FluidCam mm-scale 3D data show that present satellite and airborne remote sensing techniques poorly characterize coral reef percent living cover, morphology type, and species breakdown at the mm, cm, and meter scales. Indeed, current global assessments of coral reef cover and morphology classification based on km-scale satellite data alone can suffer from segmentation errors greater than 40%, capable of change detection only on yearly temporal scales and decameter spatial scales, significantly hindering our understanding of patterns and processes in marine biodiversity at a time when these ecosystems are experiencing unprecedented anthropogenic pressures, ocean acidification, and sea surface temperature rise. NeMO-Net leverages our augmented machine learning algorithm that demonstrates data fusion of regional FluidCam (mm, cm-scale) airborne remote sensing with global low-resolution (m, km-scale) airborne and spaceborne imagery to reduce classification errors up to 80% over regional scales. Such technologies can substantially enhance our ability to assess coral reef ecosystems dynamics.

NASA↗