Search NASA⌕ Search

SEARCH · Search NASA

Results for “Counting statistics”

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 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
NASA NTRS1984

Evidence for material between Saturn's A and F rings from the Voyager 2 photopolarimeter experiment

The region in the Saturn system between the F ring and the outer edge of the A ring is an area that appears, in images from the imaging experiment, to be virtually devoid of material except for three small satellites. Near the orbit of 1980S28, Atlas - the innermost satellite - the Voyager Photopolarimeter Stellar Occultation data show a discontinuity in count rate which marks a boundary between the tenuous material near the outer edge of the A ring and the orbit of Atlas. The data pertaining to this region have been examined with the aid of statistics and models generated from other similar ring structures. It is concluded that the discontinuity is real, implying the existence of tenuous material of normal optical depth of 0.01 to 0.006 in this region.

Graps, A. L.↗
NASA NTRS1992

NASA Pocket Statistics

NASA administrative and organizational information is presented along with summaries of space flight activity including the NASA Major Launch Record, and NASA procurement, financial and manpower data. The Major Launch Record includes all launches of Scout class and larger vehicles. Vehicle and spacecraft development flights are also included. Shuttle missions are counted as one launch and one payload, where free flying payloads are not involved. Satellites deployed from the cargo bay of the Shuttle and placed in a separate orbit or trajectory are counted as an additional payload.

Source record↗
NASA NTRS2003

Analysis of ETMS Data Quality for Traffic Flow Management Decisions

The data needed for air traffic flow management decision support tools is provided by the Enhanced Traffic Management System (ETMS). This includes both the tools that are in current use and the ones being developed for future deployment. Since the quality of decision support provided by all these tools will be influenced by the quality of the input ETMS data, an assessment of ETMS data quality is needed. Motivated by this desire, ETMS data quality is examined in this paper in terms of the unavailability of flight plans, deviation from the filed flight plans, departure delays, altitude errors and track data drops. Although many of these data quality issues are not new, little is known about their extent. A goal of this paper is to document the magnitude of data quality issues supported by numerical analysis of ETMS data. Guided by this goal, ETMS data for a 24-hour period were processed to determine the number of aircraft with missing flight plan messages at any given instant of time. Results are presented for aircraft above 18,000 feet altitude and also at all altitudes. Since deviation from filed flight plan is also a major cause of trajectory-modeling errors, statistics of deviations are presented. Errors in proposed departure times and ETMS-generated vertical profiles are also shown. A method for conditioning the vertical profiles for improving demand prediction accuracy is described. Graphs of actual sector counts obtained using these vertical profiles are compared with those obtained using the Host data for sectors in the Fort Worth Center to demonstrate the benefit of preprocessing. Finally, results are presented to quantify the extent of data drops. A method for propagating track positions during ETMS data drops is also described.

Chatterji, Gano B.↗
NASA NTRS2021

Characterization Study of TestBed Infrastructure Performance in a Distributed Simulation Environment: Baseline Analysis

Characterization of the performance of Air Traffic Management Exploration (ATM-X) TestBed integration environment has been investigated and documented for one system configuration for progressively increasing traffic. Several statistical parameters were used to assess the performance of the TestBed distributed system such as mean, standard deviation, skewness, and kurtosis of latency, and update rate for aircraft state messages that are transmitted through the simulated system under investigation. It is necessary to assess the performance characteristics of distributed systems in terms of the indicated statistical parameters mentioned above. It is critical to verify the system performance with respect to a researcher’s required system performance. Computer host specifications are documented in terms of Central Processing Unit (CPU) clock speed and core count. Transmission Control Protocol/ Internet Protocol (TCP/IP) message protocol was used for data transmission. The system network topology also contributes to the latency and update rate variations from the one imposed by the data source. The motivation for selecting the TestBed infrastructure as the focus of this study can be attributed to the number of services and capabilities it provides that help simplify the process of preparing and conducting a simulation. These capabilities include an easy to use GUI for simulation configuration, access to TestBed library by the end-user of other simulation software components, a modular adapter paradigm that allows simple connectivity of external software to TestBed, connectivity with other simulation laboratories, and a Software Development Kit (SDK) for quicker development. Two types of traffic generators, Air Traffic Generator (ATG) and Multi Aircraft Control System (MACS) were used to generate messages that were injected into the TestBed distributed environment. Eight different air traffic scenarios with progressively increasing loads were generated for each air traffic simulator. The corresponding air traffic loads between the two simulators had an identical number of aircraft per scenario, but different flight plans. It was observed that the performance of MACS degraded for air traffic scenarios containing more than 200 aircraft (37.5 KB/s nominal throughput). However, ATG performed adequately under all tested air traffic loads up to 1200 aircraft (225. KB/s nominal throughput). The tests show that MACS exhibits better latency performance with smaller aircraft loads when compared to ATG. The tests also show that the TestBed infrastructure successfully transmits 1200 aircraft without significant degradation of its performance. From the latency trends for both MACS and ATG, it is clear that as aircraft load increases, the latency in the system increases as well as its standard deviation. Likewise, the trends for the update data rate for both MACS and ATG show that as the aircraft load increases, so does the standard deviation and mean of the update rates which can be attributed to the performance of MACS and ATG applications. The analysis of the results of this study have proven that the overall system performance is dependent on the individual performance of each system component that is connected to TestBed, which subsequently propagates into the system. All TestBed characterization tests were conducted in SimLabs at NASA Ames Research Center in November 2019. This study addresses the need for a baseline TestBed characterization, and the results will serve as a reference for more complex simulation systems.

Air Traffic Management simulations↗
NASA NTRSDate not supplied

Transcriptomics-based Machine Learning Analysis Predicts Space-Exposed Murine Livers

Limited sample sizes, high data dimensionality, and sensitivity to technical and biological variability of next generation sequencing (NGS), typically limits machine learning (ML) approaches in spaceflight studies that include radiation effects. However, pooling smaller studies while addressing intra- and inter-study variabilities allows for ML predictive modeling. Here, integration methods were applied to whole transcriptome shotgun sequencing (RNA-seq) data from six mouse liver GeneLab datasets (GLDS) (n ranging from 6 to 39 samples) from with a total of 81 spaceflight and ground-control samples to determine top features (i.e. genes) relevant to spaceflight including the effect of radiation exposure. RNASeq counts were normalized for each study, then merged and scaled across all datasets. Data dimensionality was reduced using a minimum redundancy maximum relevance (MRMR) methodology. Redundancy and relevance were computed using the Pearson correlation and F-statistic, respectively. The top 100 MRMR features were used to predict spaceflight vs. ground-control samples using Random Forest (RF), Support Vector Machine (SVM), and Linear Discriminant Analysis (LDA) classifiers with 5-fold cross validation (CV). Principal component analysis (PCA) on the complete feature set versus the MRMR features shows separation between spaceflight samples and ground controls (Figure 1A). The ML-based gene sets were compared against differential gene expression results obtained with DESeq2 from individual GLDS. Using all features or randomly sampled subsets at matching set sizes with MRMR, a maximum classifier accuracy of 69% was shown on the test set over 5 folds. For all classifiers, CV training using at least the top 30 MRMR genes show minimum 89% accuracy and 0.95 AUC value on the test set over 5 folds (Figure 1B). Baseline set analysis on differentially expressed genes (DEGs) identified using padj ≤ 0.05 show 295 DEGs that overlap at least two studies and 13 DEGs that overlap three studies (Figure 1C). Set analysis between the top 100 MRMR features and the DEGs showed 47 genes that overlap at least one study and 24 genes that overlap two studies. Over-representation analysis showed overlapping biological processes related to fatty acid and lipid metabolism which may indicate these processes in the response to spaceflight stressors. MRMR feature selection for the selected ML methods improve performance relative to a classifier built on all features or randomly sampled subsets. Permutation feature importance within the decorrelated MRMR features showed concordance in feature ranking between ML methods. A challenge of applying ML methods across heterogeneous NGS data is accounting for signal:noise. Here, signal validation across studies was shown by intersecting sets between top MRMR genes and DEGs from DESeq2 analysis. Non-intersecting sets introduce opportunity to explore genes relevant to differentiating space flight exposed groups and implementing ML methods across existing NGS datasets may overcome sample size limitations.

Machine Learning↗
NASA NTRSDate not supplied

Transcriptomics-based Machine Learning (ML) Analysis Predicts Space-Exposed Murine Livers

Limited sample sizes, high data dimensionality, and sensitivity to technical and biological variability of next generation sequencing (NGS), typically limits machine learning (ML) approaches in spaceflight studies that include radiation effects. However, pooling smaller studies while addressing intra- and inter-study variabilities allows for ML predictive modeling. Here, integration methods were applied to whole transcriptome shotgun sequencing (RNA-seq) data from six mouse liver GeneLab datasets (GLDS) (n ranging from 6 to 39 samples) from with a total of 81 spaceflight and ground-control samples to determine top features (i.e. genes) relevant to spaceflight including the effect of radiation exposure. RNASeq counts were normalized for each study, then merged and scaled across all datasets. Data dimensionality was reduced using a minimum redundancy maximum relevance (MRMR) methodology. Redundancy and relevance were computed using the Pearson correlation and F-statistic, respectively. The top 100 MRMR features were used to predict spaceflight vs. ground-control samples using Random Forest (RF), Support Vector Machine (SVM), and Linear Discriminant Analysis (LDA) classifiers with 5-fold cross validation (CV). Principal component analysis (PCA) on the complete feature set versus the MRMR features shows separation between spaceflight samples and ground controls (Figure 1A). The ML-based gene sets were compared against differential gene expression results obtained with DESeq2 from individual GLDS. Using all features or randomly sampled subsets at matching set sizes with MRMR, a maximum classifier accuracy of 69% on the test set over 5 folds. For all classifiers, CV training using at least the top 30 MRMR genes show minimum 89% accuracy and 0.95 AUC value on the test set over 5 folds (Figure 1B). Baseline set analysis on differentially expressed genes (DEGs) identified using padj ≤ 0.05 show 295 DEGs that overlap at least two studies and 13 DEGs that overlap three studies (Figure 1C). Set analysis between the top 100 MRMR features and the DEGs showed 47 genes that overlap at least one study and 24 genes that overlap two studies. Over-representation analysis showed overlapping biological processes related to fatty acid and lipid metabolism which may indicate these processes in the response to spaceflight stressors. MRMR feature selection for the selected ML methods improve performance relative to a classifier built on all features or randomly sampled subsets. Permutation feature importance within the decorrelated MRMR features showed concordance in feature ranking between ML methods. A challenge of applying ML methods across heterogeneous NGS data is accounting for signal:noise. Here, signal validation across studies was shown by intersecting sets between top MRMR genes and DEGs from DESeq2 analysis. Non-intersecting sets introduce opportunity to explore genes relevant to differentiating space flight exposed groups and implementing ML methods across existing NGS datasets may overcome sample size limitations.

Machine Learning↗
NASA NTRSDate not supplied

Transcriptomics-based Machine Learning Analysis Predicts Space-Exposed Murine Livers

Limited sample sizes, high data dimensionality, and sensitivity to technical and biological variability of next generation sequencing (NGS), typically limits machine learning (ML) approaches in spaceflight studies that include radiation effects. However, pooling smaller studies while addressing intra- and inter-study variabilities allows for ML predictive modeling. Here, integration methods were applied to whole transcriptome shotgun sequencing (RNA-seq) data from six mouse liver GeneLab datasets (GLDS) (n ranging from 6 to 39 samples) from with a total of 81 spaceflight and ground-control samples to determine top features (i.e. genes) relevant to spaceflight including the effect of radiation exposure. RNASeq counts were normalized for each study, then merged and scaled across all datasets. Data dimensionality was reduced using a minimum redundancy maximum relevance (MRMR) methodology. Redundancy and relevance were computed using the Pearson correlation and F-statistic, respectively. The top 100 MRMR features were used to predict spaceflight vs. ground-control samples using Random Forest (RF), Support Vector Machine (SVM), and Linear Discriminant Analysis (LDA) classifiers with 5-fold cross validation (CV). Principal component analysis (PCA) on the complete feature set versus the MRMR features shows separation between spaceflight samples and ground controls (Figure 1A). The ML-based gene sets were compared against differential gene expression results obtained with DESeq2 from individual GLDS. Using all features or randomly sampled subsets at matching set sizes with MRMR, a maximum classifier accuracy of 69% was shown on the test set over 5 folds. For all classifiers, CV training using at least the top 30 MRMR genes show minimum 89% accuracy and 0.95 AUC value on the test set over 5 folds (Figure 1B). Baseline set analysis on differentially expressed genes (DEGs) identified using padj ≤ 0.05 show 295 DEGs that overlap at least two studies and 13 DEGs that overlap three studies (Figure 1C). Set analysis between the top 100 MRMR features and the DEGs showed 47 genes that overlap at least one study and 24 genes that overlap two studies. Over-representation analysis showed overlapping biological processes related to fatty acid and lipid metabolism which may indicate these processes in the response to spaceflight stressors. MRMR feature selection for the selected ML methods improve performance relative to a classifier built on all features or randomly sampled subsets. Permutation feature importance within the decorrelated MRMR features showed concordance in feature ranking between ML methods. A challenge of applying ML methods across heterogeneous NGS data is accounting for signal:noise. Here, signal validation across studies was shown by intersecting sets between top MRMR genes and DEGs from DESeq2 analysis. Non-intersecting sets introduce opportunity to explore genes relevant to differentiating space flight exposed groups and implementing ML methods across existing NGS datasets may overcome sample size limitations.

Machine Learning↗
NASA NTRS2000

Changes in ribbon synapses and rough endoplasmic reticulum of rat utricular macular hair cells in weightlessness

This study combined ultrastructural and statistical methods to learn the effects of weightlessness on rat utricular maculae. A principle aim was to determine whether weightlessness chiefly affects ribbon synapses of type II cells, since the cells communicate predominantly with branches of primary vestibular afferent endings. Maculae were microdissected from flight and ground control rat inner ears collected on day 13 of a 14-day spaceflight (F13), landing day (R0) and day 14 postflight (R14) and were prepared for ultrastructural study. Ribbon synapses were counted in hair cells examined in a Zeiss 902 transmission electron microscope. Significance of synaptic mean differences was determined for all hair cells contained within 100 section series, and for a subset of complete hair cells, using SuperANOVA software. The synaptic mean for all type II hair cells of F13 flight rats increased by 100%, and that for complete cells by 200%. Type I cells were less affected, with synaptic mean differences statistically insignificant in complete cells. Synapse deletion began within 8 h upon return to Earth. Additionally, hair cell laminated rough endoplasmic reticulum of flight rats was reversibly disorganized on R0. Results support the thesis that synapses in type II hair cells are uniquely affected by altered gravity. Type II hair cells may be chiefly sensors of gravitational and type I cells of translational linear accelerations.

short duration↗
NASA NTRS2009

Group Capability Model

The Group Capability Model (GCM) is a software tool that allows an organization, from first line management to senior executive, to monitor and track the health (capability) of various groups in performing their contractual obligations. GCM calculates a Group Capability Index (GCI) by comparing actual head counts, certifications, and/or skills within a group. The model can also be used to simulate the effects of employee usage, training, and attrition on the GCI. A universal tool and common method was required due to the high risk of losing skills necessary to complete the Space Shuttle Program and meet the needs of the Constellation Program. During this transition from one space vehicle to another, the uncertainty among the critical skilled workforce is high and attrition has the potential to be unmanageable. GCM allows managers to establish requirements for their group in the form of head counts, certification requirements, or skills requirements. GCM then calculates a Group Capability Index (GCI), where a score of 1 indicates that the group is at the appropriate level; anything less than 1 indicates a potential for improvement. This shows the health of a group, both currently and over time. GCM accepts as input head count, certification needs, critical needs, competency needs, and competency critical needs. In addition, team members are categorized by years of experience, percentage of contribution, ex-members and their skills, availability, function, and in-work requirements. Outputs are several reports, including actual vs. required head count, actual vs. required certificates, CGI change over time (by month), and more. The program stores historical data for summary and historical reporting, which is done via an Excel spreadsheet that is color-coded to show health statistics at a glance. GCM has provided the Shuttle Ground Processing team with a quantifiable, repeatable approach to assessing and managing the skills in their organization. They now have a common frame of reference across NASA/contractor lines to communicate and mitigate any critical skills concerns.

Olejarski, Michael↗
NASA NTRS2015

The Impact of Radiation Damage on Photon Counting with an EMCCD for the WFIRST-AFTA Coronagraph

WFIRST-AFTA is a 2.4m class NASA observatory designed to address a wide range of science objectives using two complementary scientific payloads. The Wide Field Instrument (WFI) offers Hubble quality imaging over a 0.28 square degree field of view, and will gather NIR statistical data on exoplanets through gravitational microlensing. The second instrument is a high contrast coronagraph that will carry out the direct imaging and spectroscopic analysis of exoplanets, providing a means to probe the structure and composition of planetary systems. The coronagraph instrument is expected to operate in low photon flux for long integration times, meaning all noise sources must be kept to a minimum. In order to satisfy the low noise requirements, the Electron Multiplication (EM)-CCD has been baselined for both the imaging and spectrograph cameras. The EMCCD was selected in comparison with other candidates because of its low effective electronic read noise at sub-electron values with appropriate multiplication gain setting. The presence of other noise sources, however, such as thermal dark signal and Clock Induced Charge (CIC), need to be characterised and mitigated. In addition, operation within a space environment will subject the device to radiation damage that will degrade the Charge Transfer Efficiency (CTE) of the device throughout the mission lifetime. Here we present our latest results from pre- and post-irradiation testing of the e2v CCD201-20 BI EMCCD sensor, baselined for the WFIRST-AFTA coronagraph instrument. A description of the detector technology is presented, alongside considerations for operation within a space environment. The results from a room temperature irradiation are discussed in context with the nominal operating requirements of AFTA-C and future work which entails a cryogenic irradiation of the CCD201-20 is presented.

Peddada, Pavani↗
NASA NTRS2013

Single-Photon-Sensitive HgCdTe Avalanche Photodiode Detector

The purpose of this program was to develop single-photon-sensitive short-wavelength infrared (SWIR) and mid-wavelength infrared (MWIR) avalanche photodiode (APD) receivers based on linear-mode HgCdTe APDs, for application by NASA in light detection and ranging (lidar) sensors. Linear-mode photon-counting APDs are desired for lidar because they have a shorter pixel dead time than Geiger APDs, and can detect sequential pulse returns from multiple objects that are closely spaced in range. Linear-mode APDs can also measure photon number, which Geiger APDs cannot, adding an extra dimension to lidar scene data for multi-photon returns. High-gain APDs with low multiplication noise are required for efficient linear-mode detection of single photons because of APD gain statistics -- a low-excess-noise APD will generate detectible current pulses from single photon input at a much higher rate of occurrence than will a noisy APD operated at the same average gain. MWIR and LWIR electron-avalanche HgCdTe APDs have been shown to operate in linear mode at high average avalanche gain (M > 1000) without excess multiplication noise (F = 1), and are therefore very good candidates for linear-mode photon counting. However, detectors fashioned from these narrow-bandgap alloys require aggressive cooling to control thermal dark current. Wider-bandgap SWIR HgCdTe APDs were investigated in this program as a strategy to reduce detector cooling requirements.

Huntington, Andrew↗
NASA NTRS1994

Galaxy pairs in deep HST images: Evidence for evolution in the galaxy merger rate

We use four deep serendipitous fields observed with the Hubble Space Telescope (HST) Wide-Field Camera to constrain the rate of galaxy merging between the current epoch and z approximately equals 0.7. Since most mergers occur between members of bound pairs, the merger rate is given to a good approximation by (half) the rate of disappearance of galaxies in pairs. An objective criterion for pair membership shows that 34% +/- 9% of our HST galaxies with I = 18-22 belong to pairs, compared to 7% locally. This means that about 13% of the galaxy population has disappeared due to merging in the cosmic epoch corresponding to this magnitude interval (or 0.1 approximately less than z approximately less than 0.7). Our pair fraction is a lower limit: correction for pair members falling below our detection threshold might raise the fraction to approximately 50%. Since we address only two-galaxy merging, these values do not include physical systems of higher multiplicity. Incorporating I-band field-galaxy redshift distributions, the pair fraction grows with redshift as alpha(1 + z)(exp 3.5 +/- 0.5) and the merger rate as (1 + z)(exp 2.5 +/- 0.5). This may have significant implications for the interpretation of galaxy counts (disappearance of faint blue galaxies), the cosmological evolution of faint radio sources and quasars (which evolve approximately as (1 + z)(exp 3), the similarity in the power law is necessary but not sufficient evidence for a causal relation), statistics of QSO companions, the galaxy content in distant clusters, and the merging history of a 'typical' galaxy.

Burkey, Jordan M.↗
NASA NTRS2024

A Review of Twenty-Six Years of U.S. Weather Radar Detection of Meteorite Falls

Weather radars have proven to be a valuable tool for detection, rapid recovery, and analysis of meteorite falls and fall statistics. The U.S. operates a nationwide weather radar network (NEXRAD) which will be discussed here, but it is important to acknowledge that weather radars operate worldwide and efforts are underway to utilize this rich resource. By the author’s count, since the NEXRAD system came online in the late 1990s it has detected 34 recovered meteorite falls and another 33 unrecovered probable falls. Weather radars are powerful tools because they directly detect meteorites falling after the end of their luminous flight, during the "dark flight" period. This allows rapid recovery through detection and modeling of fall locations down to as little as tens of meters. They are also the only tool that analyzes meteorites that survive their parent fireball, before many or all of them are lost to terrain. This allows direct measurement of meteorite mass, particle size distribution, and with additional work should allow calculation of pre-atmospheric mass and a measure of whole-body strength.

M D Fries↗
NASA NTRS1972

Statistical load data processing

A recorder system has been installed on two operational fighter aircrafts. Signal values from a c.g.-acceleration transducer and a strain-gage installation at the wing root were sampled and recorded in digital format on the recorder system. To analyse such load-time histories for fatigue evaluation purposes, a number of counting methods are available in which level crossings, peaks, or ranges are counted. Ten different existing counting principles are defined. The load-time histories are analysed to evaluate these counting methods. For some of the described counting methods, the counting results might be affected by arbitrarily chosen parameters such as the magnitude of load ranges that will be neglected and other secondary counting restrictions. Such influences might invalidate the final counting results entirely. The evaluation shows that for the type of load-time histories associated with most counting methods, a sensible value of the parameters involved can be found.

Vandijk, G. M.↗
NASA NTRS1995

Timing the Geminga pulsar with EGRET data

The pulsation of Geminga has been detected to date only at high energies (E greater than 0.1 keV). Since x-ray exposures are short and Geminga is at best only marginally detected in gamma-rays at E less than 30 MeV, the primary means of timing Geminga is with high-energy gamma-rays. The EGRET observations of Geminga now span 4 years. These data are analyzed to determine the 1995 ephemeris for Geminga which is provided here. We continue to count every revolution of Geminga during the GRO mission with a rotational phase resolution which improves with additional exposure. Proper motion is now apparent in gamma-ray timing, consistent with the optical measurement of Bignami et al. With improved statistics, two additional peaks are tentatively detected in the 'minor bridge' region. More exposure is required to confirm them. If found to be real, they are difficult to understand with polar cap models, but are expected for the outer gap model, and provide sorely needed constraints.

Mattox, J. R.↗
NASA NTRS1995

Effects of Space Missions on the Human Immune System: A Meta-Analysis

Future spaceflight will require travelers to spend ever-increasing periods of time in microgravity. Optimal functioning of the immune system is of paramount importance for the health and performance of these travelers. A meta-analysis statistical procedure was used to analyze immune system data from crew members in United States and Soviet space missions from 8.5 to 140 days duration between 1968 and 1985. Ten immunological parameters (immunoglobulins A, G, M, D, white blood cell (WBC) count, number of lymphocytes, percent total lymphocytes, percent B lymphocytes, percent T lymphocytes, and lymphocyte reactivity to mitogen) were investigated using multifactorial, repeated measure analysis of variance. With the preflight level set at 100, WBC count increased to 154 +/- 14% (mean +/- SE; p less than or equal to 0.05) immediately after flight; there was a decrease in lymphocyte count (83 +/- 4%; p less than or equal to 0.05) and percent of total lymphocytes (69 +/- 1%; p less than or equal to 0.05) immediately after flight, with reduction in RNA synthesis to phytohemagglutinin (PHA) to 51 +/- 21% (p less than or equal to 0.05) and DNA synthesis to PHA to 61 +/- 8% (p less than or equal to 0.05) at the first postflight measurement. Thus, some cellular immunological functions are decreased significantly following spaceflight. More data are needed on astronauts' age, aerobic power output, and parameters of their exercise training program to determine if these immune system responses are due solely to microgravity exposure or perhaps to some other aspect of spaceflight.

Greenleaf, J. E.↗
NASA NTRS2001

Timing the Geminga Pulsar with EGRET Data

The pulsation of Geminga has been detected to date only at high energies (E greater than 0.1 keV). Since X-ray exposures are short and Geminga is at best only marginally detected in gamma-rays at E less than 30 MeV, the primary means of timing Geminga is with high-energy gamma-rays. The EGRET observations of Geminga now span 4 years. These data are analyzed to determine the 1995 ephemeris for Geminga which is provided here. We continue to count every revolution of Geminga during the GRO mission with a rotational phase resolution which improves with additional exposure. Proper motion is now apparent in the gamma-ray timing, consistent with the optical measurement of Bignami et al. With improved statistics, two addition peaks are tentatively detected in the "minor bridge" region. More exposure is required to confirm them. If found to be real, they are difficult to understand with polar cap models, but are expected for the outer gap model, and provide sorely needed constraints.

Mattox, J. R.↗
NASA NTRS1998

U.S. Civil Rotorcraft Accidents, 1963 through 1997

The U.S. National Transportation Safety Board (NTSB) has recorded 8,436 rotorcraft accidents during the period mid - 1963 through the end of 1997. Review and analysis of the NTSB summary narrative for each accident has been completed. In addition, FAA (Federal Aviation Administration) counts of the growing registered rotorcraft fleet over this period has obtained. Taken together, a large and informative data base is now available, which indicates that the accident rate (on a per airframe basis) has changed very little since the mid 1970s. The data base, even in the summary form provided by this paper, offers suggestions for safer designs and improved flight operations. For analysis purposes, each accident has been placed in one of 21 top level categories as defined by the NTSB. Analysis of this grouping shows that 70 percent of rotorcraft accidents are associated with four categories. The accident count in these top four categories are: (1) 2,408 Loss of engine power (2) 1,322 In flight collision with object (3) 1,114 Loss of control (4) 1,083 Airframe/component/system failure or malfunction. Single engine rotorcraft dominate these accident statistics because of their sheer numbers over the study period. One-third of the loss of engine power accidents with these aircraft is fuel/air mixture related and fuel exhaustion is a common event. This appears to be the case whether a piston or turbine engine is installed. This paper provides similar study results in the other major mishap categories. It shows that both minor and major design and flight operations changes can -- and should -- be made to reduce rotorcraft accidents in the future. The paper outlines these changes and suggests how they may be made.

Harris, Franklin D.↗