SEARCH · Engineering Papers

Results for “Sample size determination”

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 811 records · Page 45

All sources NASA NTRS DOE OSTI

NASA NTRS1997

New Potential Sources for Black Onaping Carbon

One intriguing and important issue of the Sudbury Structure concerns the source of the relatively large amount of C in the Onaping Formation Black member. This dilemma was recently addressed, and the conclusion was reached that an impactor could not have delivered all of the requisite C. Becker et al. have suggested that much of the C came from the impactor and reported the presence of interstellar He "caged" inside some fullerenes that may have survived the impact. So, conceivably, the C inventory in the Sudbury Structure comes from both target and impactor materials, although the known target rocks have little C. We discuss here the possibility of two terrestrial sources for at least some of the C: (1) impact evaporation/dissociation of C from carbonate target rocks and (2) the presence of heretofore-unrecognized C-rich (up to 26 wt%) siliceous "shale," fragments, which are found in the upper, reworked Black member. Experimental: Hypervelocity impact of a 0.635-diameter Al projectile into dolomite at 5.03 km/s (performed at the Ames Research Center vertical gun range) produced a thin, black layer (= 0.05 mm thick) that partially lined the crater and coated impactor remnants. Scanning electronic microscope (SEM) imagery shows this layer to be spongelike on a submicron scale and Auger spectroscopic analyses yield: 33% C, 22% Mg, 19% 0, and 9% Al (from the projectile). Elemental mapping shows that all of the available 0 is combined with Ca and Mg, Al is not oxidized, and C is in elemental form. Dissociation efficiency of C from CO2 is estimated to be <10% of crater volume. Raman spectroscopy indicates that the C is highly disorganized graphite. Another impact experiment [4] also produced highly disordered graphite from a limestone target (reducing collector), in addition to small amounts of diamond/lonsdaleite/chaoite (oxidizing collector). These experiments confirm the reduction of C from carbonates in impact vapor plumes. Observational: SEM observations and microprobe analyses of small, black shalelike inclusions in the upper Black Onaping indicate high C contents (7-26 wt% avg. = 16%). They contain mostly quartz and carbonaceous matter with small amounts of altered K-feldspar, clays, Fe oxide, and a sulfide. No evidence of shock is seen in quartz, and overall characteristics indicate a natural, lightly metamorphosed carbonaceous shale or mudstone that probably existed as a preimpact rock in the target region and distal fragments washed in during early crater filling. Fragments range in size from tens of microns to cm and increase in abundance in the upper Black toward the Onwatin contact, although their distribution is highly irregular. This increase corresponds to an increase in "organic" C with increasingly negative delta-13 C values and S, together with a decrease in fullerene abundance. In addition, we have found soot in acid-demineralized residues of the Onwatin but not in the Onaping samples. These data could be consistent with impact plume and atmospheric chemical processes, with possible diageneric ovedays. We are analyzing carbonaceous fractions of the Onaping and Onwatin to determine diagnostic C isotopic signatures Analyses by Whitehead et al. on bulk samples revealed no definitive source or processes, although delta-13 C values for "organic" C overlapped those for some meteorites. Discussion: If impact evaporation of Sudbury target carbonates did occur, then where are the carbonates? Distal carbonate (limestone/dolostone) exposures of the Espanola Formation (Huronian Supergroup) are generally thin-bedded, although remnants that partially encompass the Sudbury Crater are variable in thickness and may locally reach 250 m . If a carbonate thickness of 100-200 in existed at the target site, then copious amounts of C could have been reduced by impact processing of carbonates and also C-shale, depending on the efficiency of the processing and the amount of postimpact oxidation. Conclusion: The Sudbury crater offers a unique opportunity to study preserved characteristics of immediate carbonaceous fallback matter and particles of short-term residency in the impact plume as well as dust/aerosols from postimpact atmospheric processing.

Bunch, T. E.↗

NASA NTRS2003

A phase response curve to single bright light pulses in human subjects

The circadian pacemaker is differentially sensitive to the resetting effects of retinal light exposure, depending upon the circadian phase at which the light exposure occurs. Previously reported human phase response curves (PRCs) to single bright light exposures have employed small sample sizes, and were often based on relatively imprecise estimates of circadian phase and phase resetting. In the present study, 21 healthy, entrained subjects underwent pre- and post-stimulus constant routines (CRs) in dim light (approximately 2-7 lx) with maintained wakefulness in a semi-recumbent posture. The 6.7 h bright light exposure stimulus consisted of alternating 6 min fixed gaze (approximately 10 000 lx) and free gaze (approximately 5000-9000 lx) exposures. Light exposures were scheduled across the circadian cycle in different subjects so as to derive a PRC. Plasma melatonin was used to determine the phase of the onset, offset, and midpoint of the melatonin profiles during the CRs. Phase shifts were calculated as the difference in phase between the pre- and post-stimulus CRs. The resultant PRC of the midpoint of the melatonin rhythm revealed a characteristic type 1 PRC with a significant peak-to-trough amplitude of 5.02 h. Phase delays occurred when the light stimulus was centred prior to the critical phase at the core body temperature minimum, phase advances occurred when the light stimulus was centred after the critical phase, and no phase shift occurred at the critical phase. During the subjective day, no prolonged 'dead zone' of photic insensitivity was apparent. Phase shifts derived using the melatonin onsets showed larger magnitudes than those derived from the melatonin offsets. These data provide a comprehensive characterization of the human PRC under highly controlled laboratory conditions.

Randomized Controlled Trial↗

NASA NTRS2010

Estimating the Probability of Electrical Short Circuits from Tin Whiskers

To comply with lead-free legislation, many manufacturers have converted from tin-lead to pure tin finishes of electronic components. However, pure tin finishes have a greater propensity to grow tin whiskers than tin-lead finishes. Since tin whiskers present an electrical short circuit hazard in electronic components, simulations have been developed to quantify the risk of said short circuits occurring. Existing risk simulations make the assumption that when a free tin whisker has bridged two adjacent exposed electrical conductors, the result is an electrical short circuit. This conservative assumption is made because shorting is a random event that had an unknown probability associated with it. Note however that due to contact resistance electrical shorts may not occur at lower voltage levels. In our first article we developed an empirical probability model for tin whisker shorting. In this paper, we develop a more comprehensive empirical model using a refined experiment with a larger sample size, in which we studied the effect of varying voltage on the breakdown of the contact resistance which leads to a short circuit. From the resulting data we estimated the probability distribution of an electrical short, as a function of voltage. In addition, the unexpected polycrystalline structure seen in the focused ion beam (FIB) cross section in the first experiment was confirmed in this experiment using transmission electron microscopy (TEM). The FIB was also used to cross section two card guides to facilitate the measurement of the grain size of each card guide's tin plating to determine its finish .

Courey, Karim J.↗

NASA NTRS2009

Tin Whisker Electrical Short Circuit Characteristics

Existing risk simulations make the assumption that when a free tin whisker has bridged two adjacent exposed electrical conductors, the result is an electrical short circuit. This conservative assumption is made because shorting is a random event that has an unknown probability associated with it. Note however that due to contact resistance electrical shorts may not occur at lower voltage levels. In our first article we developed an empirical probability model for tin whisker shorting. In this paper, we develop a more comprehensive empirical model using a refined experiment with a larger sample size, in which we studied the effect of varying voltage on the breakdown of the contact resistance which leads to a short circuit. From the resulting data we estimated the probability distribution of an electrical short, as a function of voltage. In addition, the unexpected polycrystalline structure seen in the focused ion beam (FIB) cross section in the first experiment was confirmed in this experiment using transmission electron microscopy (TEM). The FIB was also used to cross section two card guides to facilitate the measurement of the grain size of each card guide's tin plating to determine its finish.

Courey, Karim J.↗

NASA NTRS2009

Developing an Empirical Model for Estimating the Probability of Electrical Short Circuits from Tin Whiskers

Courey, Karim J.↗

DOE OSTI2020

MESOSCALE THERMAL TRANSPORT MEASUREMENTS OF MULTI-PHASE AND POROUS NUCLEAR FUELS USING A SQUARE-WAVE PULSE THERMOREFLECTANCE TECHNIQUE

The safe and efficient operation of nuclear reactors require accurate knowledge of peak temperatures in the fuel assemblies. The temperature profiles are governed by the thermal transport properties of the fuel, namely the thermal conductivities (k) and thermal diffusivities (D). These values can be very difficult to measure as they are known to vary considerably from the measured bulk values of the fresh fuel, and quickly degrade with increasing burnup [1 -3] . Laser-based techniques have been effectively used for non-destructive and non-contact thermal transport measurements of a wide variety of materials, including nuclear materials [4-6] that would otherwise prove too hazardous or difficult to measure otherwise. In this study, a new thermoreflectance technique known as square-pulse transient thermoreflectance (SPTR) is described and used to determine the mesoscale thermal diffusivity of both uranium sesquisilicide (U3Si2) and uranium nitride (UN) phases in a composite fuel with micron level spatial resolution [7]. This technique employs a rapid train of square-wave pulses from an excitation laser to create a periodic heat flux on a gold coated sample surface. Surface adsorption results in transient film temperatures and hence rapid fluctuations in thermoreflectance that can be measured via a detection laser coupled with a digital oscilloscope. The lasers are coaxially focused on the sample surface, allowing for a sample measurement area of a single convolved laser spot size (~2 µm). A sensitivity analysis was conducted to identify key measurement parameters of this technique using reference materials with a range of thermal conductivities comparable to those of both ceramic, composite, and metal nuclear fuel types (1.4 – 27.2 W/m-K). The reference materials were measured using the new technique as well as a spatial-domain thermoreflectance technique (SDTR) previously reported for a comparison [8]. Additionally, measurements of several U3Si2 and UN phase regions of a polished UN/U3Si2 (70/30 vol.%) sample were taken, and the resulting calculated D values are reported, with both techniques showing excellent agreement between samples. This technique was used to scan a multiphase region at 5 micron increments to generate a local diffusivity map, demonstrating the utility of the techniques for measuring thermal transport properties in specimens with precipitates and secondary phases. Furthermore, the technique is currently being applied to measure thermal properties of Fast Flux Test Reactor (FFTR) irradiated metal fuel specimens whose porous microstructure make it very difficult to measure using other techniques. Comparison of the local scale measurements are compared with the pre-irradiated fuel samples to show the degradation of thermal transport in fuels due to pores from fission gas bubbles and other irradiation induced defects.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

DOE OSTI2021

Identifying crystallographic faces of the fluorites urania and thoria with rotational polarized Raman spectroscopy

Crystallite orientation identification is invaluable, but is often limited to small area identification or requires a large area sample. Nondestructive optical methods such as polarized Raman spectroscopy, in contrast, can be used to completely map a variety of sample sizes, but their potential is not yet fully realized. Here, we report a systematic study of polarized Raman scattering of high-quality, hydrothermally grown, single crystals of urania and thoria. The peak intensity variations for as-grown major crystal planes, post-growth polished crystal planes, and a post-growth polished non-crystallographic plane are directly linked to crystallographic orientation and crystal rotation, and agree with computed models. In particular, the parallel polarized peak intensity results are directly correlated with metal–oxygen–metal chains in the fluorite structure and can be used to determine both orientation and rotational alignment of a given crystal face if sufficiently small rotational steps are applied. These results are structure based, being applicable to the larger fluorite phase space, which is useful for optical, semiconductor, nuclear, and solid oxide fuel cell industries. Further, these results suggest that Raman spectroscopy can identify non-crystallographic orientations that are not discernable by traditional means.

36 MATERIALS SCIENCE↗

DOE OSTI2021

Kinetically Controlled Linker Binding in Rare Earth-2,5-Dihydroxyterepthalic Acid Metal–Organic Frameworks and Its Predicted Effects on Acid Gas Adsorption

In the pursuit of highly stable and selective metal–organic frameworks (MOFs) for the adsorption of caustic acid gas species, an entire series of rare earth MOFs have been explored. Each of the MOFs in this series (RE-DOBDC; RE = La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, and Lu; DOBDC = 2,5-dihydroxyterepthalic acid) was synthesized in the tetragonal space group I4/m. Crystallized MOF samples, specifically Eu-DOBDC, were seen to have a combination of monodentate and bidentate binding when synthesized under typical reaction conditions, resulting in a contortion of the structure. However, extended crystallization times determined that this binding is kinetically controlled and that the monodentate binding option was crystallographically eliminated by extended reaction times at higher temperatures. Furthermore, this series allows for the direct study of the effect of the metal center on the structure of the of the MOF; herein, the lanthanide metal ionic radii contraction across the periodic table results in a reduction of the MOF pore size and lattice parameters. Scanning electron microscopy–energy-dispersive spectroscopy was used to investigate the stages of crystal growth for these RE-DOBDC MOFs. All MOFs, except Er-DOBDC had a minimum of two stages of growth. These analogues were demonstrated by analysis of neutron diffraction (PND) to exhibit a cooperative rotational distortion of the secondary building unit, resulting in two crystallographically distinct linker sublattices. Computational modeling efforts were used to show distinct differences on acid gas (NO 2 and SO 2 ) binding energies for RE-DOBDC MOFs when comparing the monodentate/bidentate combined linker with the bidentate-only linker crystal structures.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2023

New signature method for identifying organic material in atmosphere and water samples

The levels of plastic contamination in the earth’s oceans have steadily increased over the past century, creating a threat to marine life and increasing climate change. Several methods have been developed over the past decade to try to determine the amount and accumulation rate of plastic impurities in the ecosystem. These methods have proven promising but suffer from long measurement times, bulky non-mobile systems, and sample degradation. We propose a smaller, faster system that employs an electron beam to examine seawater samples for polyethylene contamination levels. This process generates a full electron and photon energy spectrum analyses based on the subatomic particles generated from electron interactions. Our simulations have already been developed based on the known molecular composition of seawater and polyethylene, showing the possibility of discerning molecular plastics from microorganisms. With a system the size of a suitcase, this system would be easily deployable to remote locations around the world or installed as a monitoring system for manufacturing facilities. This concept could lead to a new methodology for the science of signatures and be applied to other biological hazard investigations, such as general virus detection.

43 PARTICLE ACCELERATORS↗

NASA NTRS1977

Testing of typical spacecraft materials in a simulated substorm environment

The test specimens were spacecraft paints, silvered Teflon, thermal blankets, and solar array segments. The samples, ranging in size from 300 to 1000 sq cm were exposed to monoenergetic electron energies from 2 to 20 keV at a current density of 1 NA/sq cm. The samples generally behaved as capacitors with strong voltage gradient at their edges. The charging characteristics of the silvered Teflon, Kapton, and solar cell covers were controlled by the secondary emission characteristics. Insulators that did not discharge were the spacecraft paints and the quartz fiber cloth thermal blanket sample. All other samples did experience discharges when the surface voltage reached -8 to -16kV. The discharges were photographed. The breakdown voltage for each sample was determined and the average energy lost in the discharge was computed.

Stevens, N. J.↗

NASA NTRS1993

The effect of an on-orbit near encounter on the number flux density of micron sized particles

Many materials and techniques have been developed by the authors to sample the flux of particles in Low Earth Orbit (LEO), and through regular insitu sampling of the flux in LEO, the materials and techniques have produced data which complement the data now being amassed by the Long Duration Exposure Facility (LDEF) research activities. Recent flight experiments on STS-32, STS-44, STS-46, and STS-52 have been conducted to develop an understanding of the spatial density as a function of size (mass) for particle sizes 1x10(exp -6) cm and larger. In addition to the enumeration of particle impacts, it was also the intent of these experiments that hypervelocity particles be captured and returned intact. Measurements were performed post-flight to determine the flux density, diameters, and subsequent effects on various optical, thermal control, and structural materials. During the course of the STS-44 mission, the Space Shuttle corrected its altitude by 26 km to evade a spent upper stage. The results of this near encounter suggests that a cloud of micron sized particles exist in the vicinity of the object. Data also suggest that the flux density is nearly two (2) orders of magnitude higher than background flux. A comparison of the number flux density along with microphotographs of the captured particles will be presented for the referenced shuttle flights.

Maag, Carl R.↗

NASA NTRS2014

Performance Evaluation of the Operational Air Quality Monitor for Water Testing Aboard the International Space Station

In the history of manned spaceflight, environmental monitoring has relied heavily on archival sampling. For short missions, this type of sample collection was sufficient; returned samples provided a snapshot of the presence of chemical and biological contaminants in the spacecraft air and water. However, with the construction of the International Space Station (ISS) and the subsequent extension of mission durations, soon to be up to one year, the need for enhanced, real-time environmental monitoring became more pressing. The past several years have seen the implementation of several real-time monitors aboard the ISS, complemented with reduced archival sampling. The station air is currently monitored for volatile organic compounds (VOCs) using gas chromatography-differential mobility spectrometry (Air Quality Monitor [AQM]). The water on ISS is analyzed to measure total organic carbon and biocide concentrations using the Total Organic Carbon Analyzer (TOCA) and the Colorimetric Water Quality Monitoring Kit (CWQMK), respectively. The current air and water monitors provide important data, but the number and size of the different instruments makes them impractical for future exploration missions. It is apparent that there is still a need for improvements in environmental monitoring capabilities. One such improvement could be realized by modifying a single instrument to analyze both air and water. As the AQM currently provides quantitative, compound-specific information for target compounds present in air samples, and many of the compounds are also targets for water quality monitoring, this instrument provides a logical starting point to evaluate the feasibility of this approach. In this presentation, we will discuss our recent studies aimed at determining an appropriate method for introducing VOCs from water samples into the gas phase and our current work, in which an electro-thermal vaporization unit has been interfaced with the AQM to analyze target analytes at the relevant concentrations at which they are routinely detected in archival water samples from the ISS.

Wallace, William T.↗

NASA NTRSDate not supplied

Particle Size Distribution Parameters from SAGE III/ISS Extinction Spectra

Stratospheric aerosols play key roles in the chemistry and radiation balance of the atmosphere and are a key input parameter for global chemistry and climate models. The degree to which aerosols impact chemistry and radiation balance depends primarily on the relative abundance of different sized particles within the sample volume, often referred to as the particle size distribution (PSD). If the PSD is accurately known then other key modeling parameters (e.g., surface area density and effective radius) can be derived. Historically, occultation observations from orbital instruments such as SAGE III/ISS have been used to infer these PSD parameters by inverting the extinction coefficient spectra. However, past efforts routinely failed to account for measurement uncertainty and lacked a rigorous estimate of the inferred PSD uncertainty. We carried out a series of simulations to evaluate the accuracy of these inferences and, for every valid SAGE III/ISS extinction spectrum, determined the range of PSD parameters that fell withing the bounds of the extinction error bars. Special application of this method was applied to estimate the impact of the 2022 Hunga Tonga eruption had on particle size distributions.

Travis N Knepp↗

DOE OSTI2025

Shielded magnetic small-angle neutron scattering for characterization of radioactive samples

The development of a Pb-shielded fixture for the execution of a small-angle neutron scattering (SANS)-based workflow for interrogation of highly irradiated nuclear materials has been explored. The Pb shielding was specially designed to reduce the detected radioactivity from the specimen during SANS experiments, and the overall configuration is termed shielded magnetic SANS (SM-SANS). Two FeCrAl-based alloys, C35M and 125YF, were examined with the SM-SANS technique using a free-form size distribution locally monodisperse model in both the as-received and irradiated states. Quantitative values derived from the free-form size distribution were compared with atom probe tomography experiments. Microstructural and compositional parameters determined using the two characterization techniques were complements of each other. The results demonstrate that the SM-SANS technique is an effective means of characterizing nanoscale clustering in irradiated material systems and provides new avenues for investigating radioactive material microstructures.

FeCrAl↗

DOE OSTI2024

Validation of the Scientific Program for the Dark Energy Spectroscopic Instrument

The Dark Energy Spectroscopic Instrument (DESI) was designed to conduct a survey covering 14,000 deg$^{2}$ over 5 yr to constrain the cosmic expansion history through precise measurements of baryon acoustic oscillations (BAO). The scientific program for DESI was evaluated during a 5 month survey validation (SV) campaign before beginning full operations. This program produced deep spectra of tens of thousands of objects from each of the stellar Milky Way Survey (MWS), Bright Galaxy Survey (BGS), luminous red galaxy (LRG), emission line galaxy (ELG), and quasar target classes. These SV spectra were used to optimize redshift distributions, characterize exposure times, determine calibration procedures, and assess observational overheads for the 5 yr program. In this paper, we present the final target selection algorithms, redshift distributions, and projected cosmology constraints resulting from those studies. We also present a One-Percent Survey conducted at the conclusion of SV covering 140 deg$^{2}$ using the final target selection algorithms with exposures of a depth typical of the main survey. The SV indicates that DESI will be able to complete the full 14,000 deg$^{2}$ program with spectroscopically confirmed targets from the MWS, BGS, LRG, ELG, and quasar programs with total sample sizes of 7.2, 13.8, 7.46, 15.7, and 2.87 million, respectively. These samples will allow exploration of the Milky Way halo, clustering on all scales, and BAO measurements with a statistical precision of 0.28% over the redshift interval z < 1.1, 0.39% over the redshift interval 1.1 < z < 1.9, and 0.46% over the redshift interval 1.9 < z < 3.5.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS1980

Mineralogic and petrologic studies of lunar samples and meteorites

Experimental and thermodynamic research on the pressure temperature limits of the stability of the mineral assemblages found in pristine, spinel bearing lunar highland lithologies demonstrated the likelihood that the minerals originated in the lower stratigraphic levels of the primordial crust. The phase equilibrium in silicate solid/liquid systems of planetary importance were thermochemically interpreted in order to model the early formation of the crusts and maneles of Earth and Moon sized planets. The petrography and chemical composition of coarse grained gabbro, the chemical analysis and age dating of clasts from Apollo 16 breccia, the analysis of glass particles from Apollo 16 soil samples, the study of Allende and Mokoia meteorites as a source of information about events in the solar nebula, and the hydrothermal alteration of amorphous materials were also investigated. The capabilities of a model for addressing the problem of the origin of the Earth's moon by the disruptive capture mechanism are examined as well as models of the thermal evolution of hypothetical meteorite bodies. Progress in determining the composition of stony meteorite specimens collected at the Allan Hills site during the Antarctic field exploration is reported.

Wood, J. A.↗

NASA NTRS2003

Identification of Presolar Spinel Grains from a Murray Residue by Multi-Detection Raster Imaging

Grain size separate CG from the Murray CM2 carbonaceous chondrite contains mostly spinel grains of average diameter 0.5 m. Zinner et al. found that approximately 1% of these spinel grains are of presolar origin as determined by their large O isotopic anomalies. These O isotopic measurements were made with the NanoSIMS on individual grains that were well separated from one another on a gold foil. The grains were selected for analysis from secondary electron and secondary O-16(-) images. The primary beam was then successively deflected onto these single grains for O isotopic analysis. While single grain analysis on dispersed samples is effective for finding relatively abundant anomalous grains, ion imaging in a raster mode on tightly packed grains might be more efficient for locating few anomalous grains among predominantly isotopically normal grains. In fact, this was the analysis mode used by Messenger et al. to discover presolar silicates in interplanetary dust particles. In an exploratory effort that is also geared toward establishing the optimum isotopic imaging technique in the search for presolar silicate grains in primitive meteorites, we measured O isotopic ratios in spinel grains from the Murray CG separate by raster imaging of areas with more or less tightly packed grains.

Nguyen, A.↗

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↗