Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Feature importance”

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 865 records · Page 48
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Detecting Important Drivers of Gridded Population Modeling With Machine Learning

High-resolution population datasets have been lever-aged across a broad swath of domains, such as climate change, public policy, humanitarian aid, and rescue operations, among others. Machine learning methods were adopted to generate high-resolution or gridded population estimates by using various geospatial input features such as buildings, roads, and nighttime lights. In this study, we evaluate the importance of population features using Random Forest models across three levels of analysis, utilizing permutation measures. Our research aims to address key questions to enhance our understanding of high-resolution population modeling, such as: Are certain features globally (10 countries collectively) more important than others? Do optimal features vary by country? Within each country, do feature importance differ across administrative units? What similarities exist in feature importance at the global, country, and administrative unit levels? To answer these questions, we leverage the Kneedle algorithm to automate the selection of optimum features. We find that there are patterns displayed by features across spatial boundaries, evidenced by the same feature being the most important indicator of population across 7 of the 10 countries modeled. Our findings indicate that while important features may vary across geographies, certain features consistently hold greater importance than others agnostic of geography.

Lebakula, Viswadeep [ORNL] (ORCID:0000000152935914↗
NASA NTRS1986

Atmospheric-water absorption features near 2.2 micrometers and their importance in high spectral resolution remote sensing

Selective absorption of electromagnetic radiation by atmospheric gases and water vapor is an accepted fact in terrestrial remote sensing. Until recently, only a general knowledge of atmospheric effects was required for analysis of remote sensing data; however, with the advent of high spectral resolution imaging devices, detailed knowledge of atmospheric absorption bands has become increasingly important for accurate analysis. Detailed study of high spectral resolution aircraft data at the U.S. Geological Survey has disclosed narrow absorption features centered at approximately 2.17 and 2.20 micrometers not caused by surface mineralogy. Published atmospheric transmission spectra and atmospheric spectra derived using the LOWTRAN-5 computer model indicate that these absorption features are probably water vapor. Spectral modeling indicates that the effects of atmospheric absorption in this region are most pronounced in spectrally flat materials with only weak absorption bands. Without correction and detailed knowledge of the atmospheric effects, accurate mapping of surface mineralogy (particularly at low mineral concentrations) is not possible.

Kruse, F. A.↗
NASA NTRS2006

Mechanisms for Diurnal Variability of Global Tropical Rainfall Observed from TRMM

The behavior and various controls of diurnal variability in tropical-subtropical rainfall are investigated using Tropical Rainfall Measuring Mission (TRMM) precipitation measurements retrieved from the three level-2 TRMM standard profile algorithms for the 1998 annual cycle. Results show that diurnal variability characteristics of precipitation are consistent for all three algorithms, providing assurance that TRMM retrievals are producing consistent estimates of rainfall variability. As anticipated, most ocean areas exhibit more rainfall at night, while over most land areas, rainfall peaks during daytime; however, important exceptions are noted. The dominant feature of the oceanic diurnal cycle is a rainfall maximum in late-evening-early-morning (LE-EM) hours, while over land the dominant maximum occurs in the mid- to late afternoon (MLA). In conjunction with these maxima are pronounced seasonal variations of the diurnal amplitudes. Amplitude analysis shows that the diurnal pattern and its seasonal evolution are closely related to the rainfall accumulation pattern and its seasonal evolution. In addition, the horizontal distribution of diurnal variability indicates that for oceanic rainfall, there is a secondary MLA maximum coexisting with the LE-EM maximum at latitudes dominated by large-scale convergence and deep convection. Analogously, there is a preponderancy for an LE EM maximum over land coexisting with the stronger MLA maximum, although it is not evident that this secondary continental feature is closely associated with the large-scale circulation. Neither of the secondary maxima exhibit phase behavior that can be considered semidiurnal in nature. Diurnal rainfall variability over the ocean associated with large-scale convection is clearly an integral component of the general circulation. Phase analysis reveals differences in regional and seasonal features of the diurnal cycle, indicating that underlying forcing mechanisms differ from place to place. This is underscored by the appearance of secondary ocean maxima in the presence of large-scale convection, along with other important features. Among these, there are clear-cut differences between the diurnal variability of seasonal rainfall over the mid-Pacific and Indian Ocean Basins. The mid-Pacific exhibits double maxima in spring and winter but only LE-EM maxima in summer and autumn, while the Indian Ocean exhibits double maxima in spring and summer and only an LE-EM maximum in autumn and winter. There are also evident daytime maxima within the major large-scale marine stratocumulus regions off the west coasts of continents. The study concludes with a discussion concerning how the observational evidence either supports or repudiates possible forcing mechanisms that have been suggested to explain diurnal rainfall variability.

Yang, Song↗
NASA NTRS2015

On the Sensitivity of the Diurnal Cycle in the Amazon to Convective Intensity

This presentation uses publicly available CERES and radiosonde data to investigate the sensitivity of thetropical convective diurnal cycle to atmosphere state. Averaging surface observations into regimes of convective intensitydefined by satellite shows great promise for physical understandingof convection.• Convective processes in the Amazon are highly variable seasonallyand locally.• Buoyancy/CIN more important JJA– Mesoscale/synoptic features easier to separate– Length/depth of buoyancy layer very important in DJF (EL).• Moisture more important DJF, esp. UTH– Humidity of lower atmosphere significantly impacts LTS, LCL and abilityfor parcels to reach LFC.• Lower level jet strength/direction important• Convective initiation correlated with LTS, LR, LTH, EL• Duration/Phase better correlated with humidity variables• Surface Flux amplitude well correlated with convection

Itterly, Kyle↗
NASA NTRS2000

Supporting Scientific Analysis within Collaborative Problem Solving Environments

Collaborative problem solving environments for scientists should contain the analysis tools the scientists require in addition to the remote collaboration tools used for general communication. Unfortunately, most scientific analysis tools have been designed for a "stand-alone mode" and cannot be easily modified to work well in a collaborative environment. This paper addresses the questions, "What features are desired in a scientific analysis tool contained within a collaborative environment?", "What are the tool design criteria needed to provide these features?", and "What support is required from the architecture to support these design criteria?." First, the features of scientific analysis tools that are important for effective analysis in collaborative environments are listed. Next, several design criteria for developing analysis tools that will provide these features are presented. Then requirements for the architecture to support these design criteria are listed. Sonic proposed architectures for collaborative problem solving environments are reviewed and their capabilities to support the specified design criteria are discussed. A deficiency in the most popular architecture for remote application sharing, the ITU T. 120 architecture, prevents it from supporting highly interactive, dynamic, high resolution graphics. To illustrate that the specified design criteria can provide a highly effective analysis tool within a collaborative problem solving environment, a scientific analysis tool that contains the specified design criteria has been integrated into a collaborative environment and tested for effectiveness. The tests were conducted in collaborations between remote sites in the US and between remote sites on different continents. The tests showed that the tool (a tool for the visual analysis of computer simulations of physics) was highly effective for both synchronous and asynchronous collaborative analyses. The important features provided by the tool (and made possible by the specified design criteria) are: 1. The tool provides highly interactive, dynamic, high resolution, 3D graphics. 2. All remote scientists can view the same dynamic, high resolution, 3D scenes of the analysis as the analysis is being conducted. 3. The responsiveness of the tool is nearly identical to the responsiveness of the tool in a stand-alone mode. 4. The scientists can transfer control of the analysis between themselves. 5. Any analysis session or segment of an analysis session, whether done individually or collaboratively, can be recorded and posted on the Web for other scientists or students to download and play in either a collaborative or individual mode. 6. The scientist or student who downloaded the session can, individually or collaboratively, modify or extend the session with his/her own "what if" analysis of the data and post his/her version of the analysis back onto the Web. 7. The peak network bandwidth used in the collaborative sessions is only 1K bit/second even though the scientists at all sites are viewing high resolution (1280 x 1024 pixels), dynamic, 3D scenes of the analysis. The links between the specified design criteria and these performance features are presented.

Watson, Velvin R.↗
NASA NTRS1984

Topological grid structure - A data structure for earth science modeling

The automated analysis of land surface features is increasingly important to earth scientists. User-friendly algorithms for studying these features can be integrated into geographic information systems through the use of topological grid structure, which maintains the simplicity and transportability of standard grid structure while providing the essential capability to treat groups of contiguous, identically-classified pixels (corresponding to lakes, forests, fields, etc.) as distinct spatial entities.

Goldberg, M.↗
DOE OSTI2022

Nonnegative canonical tensor decomposition with linear constraints: nnCANDELINC

Abstract There is an emerging interest for tensor factorization applications in big‐data analytics and machine learning. To speed up the factorization of extra‐large datasets, organized in multidimensional arrays (also known as tensors), easy to compute compression‐based tensor representations, such as, Tucker and tensor train formats, are used to approximate the initial large‐tensor. Further, tensor factorization is used to extract latent features that can facilitate discoveries of new mechanisms and signatures hidden in the data, where the explainability of the latent features is of principal importance. Nonnegative tensor factorization extracts latent features that are naturally sparse and parts of the data, which makes them easily interpretable. However, to take into account available domain knowledge and subject matter expertise, often additional constraints need to be imposed, which lead us to canonical decomposition with linear constraints (CANDELINC), a canonical polyadic decomposition with rank deficient factors. In CANDELINC, Tucker compression is used as a preprocessing step, which lead to a larger residual error but to more explainable latent features. Here, we propose a nonnegative CANDELINC (nnCANDELINC) accomplished via a specific nonnegative Tucker decomposition; we refer to as minimal or canonical nonnegative Tucker. We derive several results required to understand the specificity of nnCANDELINC, focusing on the difficulties of preserving the nonnegative rank of a tensor to its Tucker core and comparing the real valued to nonnegative case. Finally, we demonstrate nnCANDELINC performance on synthetic and real‐world examples.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Teleseismic Network Association with GENIE

In this report we investigate adapting the Graph Neural Interpretation Engine (GENIE), an associator developed for three-component dense monitoring networks, to regional to teleseismic association using a sparse network of array stations. We expand GENIE’s input features to include first-P detection time, azimuth, and slowness estimates. Additionally, we include a probability of detection (PDET) term which measures a station’s likelihood of detecting an event. To assess each feature’s relative importance, we train four models, each using an increasing set of node features and find that the PDET models perform the best. We define two measures of event complexity which demonstrate that all GENIE model versions perform better than the standard backprojection stack.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Enhancing Unknown Waveform Detection by Learning Intra and Inter-domain Dependencies with Advanced Attention Fusion Mechanisms

Detection of unknown waveforms in mission-critical communications is a crucial area of interest for the Department of Energy (DoE). Traditional methods and recent deep learning-based approaches often assume that the training set includes all possible classes, which is impractical for detecting new waveforms. This limitation gives rise to the problem of open-set recognition (OSR), which involves correctly identifying known classes while detecting and rejecting unknown or unseen classes. To address this limitation, we propose a novel dual-domain complex-valued neural architecture that jointly processes time-domain and frequency-domain signal representations using transformer mechanisms. A transformer model is a deep learning architecture that uses self-attention mechanisms to process and learn relationships in sequential data. Our model employs a cosine similarity loss to extract domain-specific features and incorporates a transformer architecture in the latent space to weigh the importance of different features from the time and frequency domains. The transformer layer includes stacked self-attention and cross-attention modules to learn intra-domain and inter-domain dependencies, creating a more holistic signal representation. An attention-based fusion module intelligently combines the time and frequency-domain features using multi-head attention, enabling the network to learn the optimal feature for each domain in each input signal. Quantitative results demonstrate the impact of these architectural choices on overall performance, showing significant improvement after incorporating self and cross-attention modules and using complex attention fusion over simple weighted fusion. Our ongoing work will focus on addressing the limitations of threshold-based OSR methods by developing a novel generative framework that integrates a conditional diffusion probabilistic model (DPM). DPM is a generative framework that learns to synthesize complex data by reversing a gradual noising process using a neural network trained to denoise step-by-step. Our goal is to leverage the inherent strengths of DPMs for identifying unknown signals more robustly. One primary advantage of using a DPM is its ability to provide a more reliable anomaly score based on the model's reconstruction error, rather than relying solely on classifier confidence. Additionally, the iterative denoising process of DPMs makes this approach naturally resilient to low Signal-to-Noise Ratio (SNR) conditions, where traditional methods often fail. By implementing this generative framework, we aim to enhance the model's capability to accurately detect unknown waveforms and maintain performance in challenging environments.

99 - GENERAL AND MISCELLANEOUS↗
NASA NTRS1985

Processing technology for high efficiency silicon solar cells

Recent advances in silicon solar cell processing have led to attainment of conversion efficiency approaching 20%. The basic cell design is investigated and features of greatest importance to achievement of 20% efficiency are indicated. Experiments to separately optimize high efficiency design features in test structures are discussed. The integration of these features in a high efficiency cell is examined. Ion implantation has been used to achieve optimal concentrations of emitter dopant and junction depth. The optimization reflects the trade-off between high sheet conductivity, necessary for high fill factor, and heavy doping effects, which must be minimized for high open circuit voltage. A second important aspect of the design experiments is the development of a passivation process to minimize front surface recombination velocity. The manner in which a thin SiO2 layer may be used for this purpose is indicated without increasing reflection losses, if the antireflection coating is properly designed. Details are presented of processing intended to reduce recombination at the contact/Si interface. Data on cell performance (including CZ and ribbon) and analysis of loss mechanisms are also presented.

Spitzer, M. B.↗
NASA NTRS1998

Mineralization of Bacteria in Terrestrial Basaltic Rocks: Comparison With Possible Biogenic Features in Martian Meteorite Allan Hills 84001

The identification of biogenic features altered by diagenesis or mineralization is important in determining whether specific features in terrestrial rocks and in meteorites may have a biogenic origin. Unfortunately, few studies have addressed the formation of biogenic features in igneous rocks, which may be important to these phenomena, including the controversy over possible biogenic features in basaltic martian meteorite ALH84001. To explore the presence of biogenic features in igneous rocks, we examined microcosms growing in basaltic small-scale experimental growth chambers or microcosms. Microbial communities were harvested from aquifers of the Columbia River Basalt (CRB) group and grown in a microcosm containing unweathered basalt chips and groundwater (technique described in. These microcosms simulated natural growth conditions in the deep subsurface of the CRB, which should be a good terrestrial analog for any putative martian subsurface ecosystem that may have once included ALH84001. Here we present new size measurements and photomicrographs comparing the putative martian fossils to biogenic material in the CRB microcosms. The range of size and shapes of the biogenic features on the CRB microcosm chips overlaps with and is similar to those on ALH84001 chips. Although this present work does not provide evidence for the biogenicity of ALH84001 features, we believe that, based on criteria of size, shape, and general morphology, a biogenic interpretation for the ALH84001 features remains plausible.

Thomas-Keprta, K. L.↗
NASA NTRS2021

Automated Data Accountability for Missions in Mars Rover Data

As the Mars Curiosity Rover transmits data to the JPL Ground Data System (GDS), it frequently observes data loss and corruption, requiring re-transmits from the rover and Ground Data System Analysts (GDSA) to monitor the downlink process. As new missions are launched, the GDSA team redistributes analysts to these new missions, causing shortages in previous missions. The GDSA team can significantly benefit from the automation and optimization of the downlink process of telemetry data. In fact, there is a need for a better understanding of why the data is corrupted, so that the GDSA team can best determine the root cause of the issues in the GDS. This paper presents machine learning and deep learning based approaches to automate and optimize the detection of data loss. We first created a pipeline to automatically accumulate data from the telemetry databases (MAROS, Telemetry Data Storage, and GDS Elastic Search Database) in the downlink process. With our newly created datasets, we perform feature selection to supplement the GDSA understanding of the downlink process and provide supplemental analysis on the importance of different features. We implement various machine learning and deep learning based models, including support vector machines, ensemble methods, and deep neural networks and evaluate their accuracies in identifying whether a downlink process is complete or incomplete. We utilize fast hyperparameter optimization methods that allow our models to quickly be re-trained, allowing them to quickly be tuned and optimized on daily incoming data in real time. This hyperparameter optimization also allows our methods to be quickly integrated into other JPL missions. Our results show that our best-performing machine learning and deep learning based models outperform the existing GDSA detection software by 6 accuracy points and can aid analysts by providing insights into the data accountability problem. Since these various machine learning and deep learning approaches vary significantly in interpretability, we provide a discussion on the tradeoffs between their performance and trustworthiness in helping detect issues in data transmission.

Divsalar, Dariush↗
DOE OSTI2024

An Investigation of LES Wall Modeling for Rayleigh–Bénard Convection via Interpretable and Physics-Aware Feedforward Neural Networks with DNS

Abstract The traditional approach of using the Monin–Obukhov similarity theory (MOST) to model near-surface processes in large-eddy simulations (LESs) can lead to significant errors in natural convection. In this study, we propose an alternative approach based on feedforward neural networks (FNNs) trained on output from direct numerical simulation (DNS). To evaluate the performance, we conduct both a priori and a posteriori tests. In the a priori (offline) tests, we compare the statistics of the surface shear stress and heat flux, computed from filtered DNS input variables, to the stress and flux obtained from the filtered DNS. Additionally, we investigate the importance of various input features using the Shapley additive explanations value and the conditional average of the filter grid cells. In the a posteriori (online) tests, we implement the trained models in the System for Atmospheric Modeling (SAM) LES and compare the LES-generated surface shear stress and heat flux with those in the DNS. Our findings reveal that vertical velocity, a traditionally overlooked flow quantity, is one of the most important input features for determining the wall fluxes. Increasing the number of input features improves the a priori test results but does not always improve the model performance in the a posteriori tests because of the differences in input variables between the LES and DNS. Last, we show that physics-aware FNN models trained with logarithmic and scaled parameters can well extrapolate to more intense convection scenarios than in the training dataset, whereas those trained with primitive flow quantities cannot. Significance Statement The traditional near-surface turbulence model, based on a shear-dominated boundary layer flow, does not represent near-surface turbulence in natural convection. Using a feedforward neural network (FNN), we can construct a more accurate model that better represents the near-surface turbulence in various flows and reveals previously overlooked controlling factors and process interactions. Our study shows that the FNN-generated models outperform the traditional model and highlight the importance of the near-surface vertical velocity. Furthermore, the physics-aware FNN models exhibit the potential to extrapolate to convective flows of various intensities beyond the range of the training dataset, suggesting their broader applicability for more accurate modeling of near-surface turbulence.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2004

The Development of a Plan for the Assessment, Improvement and Deployment of a Radar Acoustic Sounding System (RASS) for Wake Vortex Detection

This report describes the activities completed under a grant from the NASA Langley Research Center to develop a plan for the assessment, improvement, and deployment of a Radar Acoustic Sounding System (RASS) for the detection of wake vortices. A brief review is provided of existing alternative instruments for wake vortex detection. This is followed by a review of previous implementations and assessment of a RASS. As a result of this review, it is concluded that the basic features of a RASS have several advantages over other commonly used wake vortex detection and measurement systems. Most important of these features are the good fidelity of the measurements and the potential for all weather operation. To realize the full potential of this remote sensing instrument, a plan for the development of a RASS designed specifically for wake vortex detection and measurement has been prepared. To keep costs to a minimum, this program would start with the development an inexpensive laboratory-scale version of a RASS system. The new instrument would be developed in several stages, each allowing for a critical assessment of the instrument s potential and limitations. The instrument, in its initial stages of development, would be tested in a controlled laboratory environment. A jet vortex simulator, a prototype version of which has already been fabricated, would be interrogated by the RASS system. The details of the laboratory vortex would be measured using a Particle Image Velocimetry (PIV) system. In the early development stages, the scattered radar signal would be digitized and the signal post-processed to determine how extensively and accurately the RASS could measure properties of the wake vortex. If the initial tests prove to be successful, a real-time, digital signal processing system would be developed as a component of the RASS system. At each stage of the instrument development and testing, the implications of the scaling required for a full-scale instrument would be considered. It is concluded that a RASS system, developed for the specific application of wake vortex detection, could become part of a robust Aircraft Vortex Spacing System (AVOSS). This system, in turn, could contribute to Reduced Spacing Operations (RSO) in US airports and improvements in Terminal Area productivity (TAP).

Morris, Philip J.↗
DOE OSTI2025

On the minimum number of radiation field parameters to specify gas cooling and heating functions

Fast and accurate approximations of gas cooling and heating functions are needed for hydrodynamic galaxy simulations. We use machine learning to analyze atomic gas cooling and heating functions in the presence of a generalized incident local radiation field computed by Cloudy. We characterize the radiation field through binned radiation field intensities instead of the photoionization rates used in our previous work. We find a set of 6 energy bins whose intensities exhibit relatively low correlation. We use these bins as features to train machine learning models to predict Cloudy cooling and heating functions at fixed metallicity. We compare the relative SHapley Additive exPlanation (SHAP) value importance of the features. From the SHAP analysis, we identify a feature subset of 3 energy bins (0.5-1, 1-4, and 13-16Ry) with the largest importance and train additional models on this subset. We compare the mean squared errors and distribution of errors on both the entire training data table and a randomly selected 20% test set withheld from model training. The machine learning models trained with 3 and 6 bins, as well as 3 and 4 photoionization rates, have comparable accuracy everywhere, with errors ≳10 times smaller than for the interpolation table of Gnedin and Hollon (2012). We conclude that 3 energy bins (or 3 analogous photoionization rates: molecular hydrogen photodissociation, neutral hydrogen HI, and fully ionized carbon CVI) are sufficient to characterize the dependence of the gas cooling and heating functions on our assumed incident radiation field model.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

On the minimum number of radiation field parameters to specify gas cooling and heating functions

Fast and accurate approximations of gas cooling and heating functions are needed for hydrodynamic galaxy simulations. We use machine learning to analyze atomic gas cooling and heating functions computed by Cloudy in the presence of a generalized incident local radiation field. We characterize the radiation field through binned radiation field intensities instead of the photoionization rates used in our previous work. We find a set of 6 energy bins whose intensities exhibit relatively low correlation. We use these bins as features to train machine learning models to predict Cloudy cooling and heating functions at fixed metallicity. We compare the relative SHapley Additive exPlanation (SHAP) value importance of the features. From the SHAP analysis, we identify a feature subset of 3 energy bins ($0.5-1, 1-4$, and $13-16 \, \mathrm{Ry}$) with the largest importance and train additional models on this subset. We compare the mean squared errors and distribution of errors on both the entire training data table and a randomly selected 20% test set withheld from model training. The machine learning models trained with 3 and 6 bins, as well as 3 and 4 photoionization rates, have comparable accuracy everywhere, with errors $\gtrsim 10$ times smaller than for the interpolation table of Gnedin and Hollon (2012). We conclude that 3 energy bins (or 3 analogous photoionization rates: molecular hydrogen photodissociation, neutral hydrogen HI, and fully ionized carbon CVI) are sufficient to characterize the dependence of the gas cooling and heating functions on our assumed incident radiation field model.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1993

Decay of equatorial ring current ions and associated aeronomical consequences

The decay of the major ion species which constitute the ring current is studied by solving the time evolution of their distribution functions during the recovery phase of a moderate geomagnetic storm. In this work, only equatorially mirroring particles are considered. Particles are assumed to move subject to E x B and gradient drifts. They also experience loses along their drift paths. Two loss mechanisms are considered: charge exchange with neutral hydrogen atoms and Coulomb collisions with thermal plasma in the plasmasphere. Thermal plasma densities are calculated with a plasmaspheric model employing a time-dependent convection electric field model. The drift-loss model successfully reproduces a number of important and observable features in the distribution function. Charge exchange is found to be the major loss mechanism for the ring current ions; however the important effects of Coulomb collisions on both the ring current and thermal populations are also presented. The model predicts the formation of a low-energy (less than 500 eV) ion population as a result of energy degradation caused by Coulomb collision of the ring current ions with the plasmaspheric electrons; this population may be one source of the low-energy ions observed during active and quiet periods in the inner magnetosphere. The energy transferred to plasmaspheric electrons through Coulomb collisions with ring current ions is believed to be the energy source for the electron temperature enhancement and the associated 6300 A (stable auroral red (SAR) arc) emission in the subauroral region. The calculated energy deposition rate is sufficient to produce a subauroral electron temperature enhancement and SAR arc emissions that are consistent with observations of these quantities during moderate magnetic activity levels.

Fok, M.-C.↗
NASA NTRS2005

Surface Clutter Removal in Airborne Radar Sounding Data from the Dry Valleys, Antarctica

We have collected roughly 1,000 line-km of airborne radar sounding data over glaciers, rock/ice glaciers, permafrost, subsurface ice bodies, ice-covered saline lakes, and glacial deposits in Taylor and Beacon Valley. These data are being analyzed in order to develop techniques for discriminating between subsurface and off-nadir echoes and for detecting and characterizing subsurface interfaces. The identification of features on Mars exhibiting morphologies consistent with ice/rock mixtures, near-surface ice bodies and near-surface liquid water, and the importance of such features to the search for water on Mars, highlights the need for appropriate terrestrial analogs and analysis techniques in order to prepare for radar sounder missions to Mars. Climatic, hydrological, and geological conditions in the Dry Valleys of Antarctica are analogous in many ways to those on Mars. A crucial first step in the data analysis process is the discrimination of echo sources in the radar data. The goal is to identify all returns from the surface of off-nadir topography in order to positively identify subsurface echoes. This process will also be critical for radar data that will be collected in areas of Mars exhibiting significant topography, so that subsurface echoes are identified unambiguously. The positive detection and characterization of subsurface (including sub-ice) water is a primary goal of NASA's Mars exploration program. Our data over the Dry Valleys provides an opportunity to implement techniques we are developing to accomplish these goals.

Holt, J. W.↗