ALBERT

Description: The objectives of this program are as follows: modelling of guided waves in fiber-reinforced plates in terms of different modes; and analysis of scattering by transverse cracks using modal representation. A hybrid numerical method combining the finite element representation of a region around the crack with the modal representation in the exterior region will be used in this program. Modes will be obtained using the through-the-thickness discretization of the displacement field.

Description: In June 1978, the Seasat satellite was launched carrying, among other instruments, the Seasat-A scatterometer system (SASS), which produced ambiguous wind speed and direction data at the ocean surface. A fifteen day subset of dealiased wind vector data with the inherent ambiguities removed was produced for the period of September 6-20, 1978. On September 8, SASS began to observe a development of frontal cyclogenesis in the South Pacific off the east coast of New Zealand, in an area of few surface observations. A large mature cyclone contained weak warm and cold fronts and an occlusion with a strong horizontal wind shear. Satellite imagery shows that a strong upper-level jet streak was moving rapidly over the area of the surface frontal occlusion and as the jet passed over this area a new vortex formed. This cyclogenesis event was studied using 50-km resolution scatterometer surface wind data. High-resolution fields of wind vectors, divergence and vorticity are computed and plotted from the scatterometer data to study the structure and development of the newly formed cyclonic vortex, not otherwise possible using conventional observations.

Description: The NASA scatterometer (NSCAT) is a spaceborne scatterometer scheduled to be deployed in the mid-1990s. An analysis of the wind retrieval error distribution for wind estimates based on backscatter measurements made by the NSCAT instrument is presented. The results are based on an end-to-end simulation of the scatterometer instrument and data processing. In general, the distribution of the wind speed error, when normalized, is independent of the true wind speed and direction. The wind speed error can be characterized by a normal distribution. The wind direction error is independent of the true wind speed, but depends on the true wind direction. Details for wind vectors with true wind speeds from 3 m/s to 33 m/s and true wind directions from 0 to 360 deg are presented.

Description: This paper considers complex transcendental eigenvalue problems where one is interested in pairs of eigenvalues that are restricted to take real values only. Such eigenvalue problems arise in dynamic stability analysis of nonconservative physical systems, i.e., flutter analysis of aeroelastic systems. Some available solution methods are discussed and a new method is presented. Two computational approaches are described for analytical evaluation of the sensitivities of these eigenvalues when they are dependent on other parameters. The algorithms presented are illustrated through examples.

Description: Modulation of a rain wave pattern by longer waves has been studied. An analytical model taking into account capillarity effects and obliquity of short waves has been developed. Modulation rates in wave number and amplitude have been computed. Experiments were carried out in a wave tank. First results agree with theoretical models, but higher values of modulation rates are measured. These results could be taken into account for understanding the radar response from the sea surface during rain.

Description: Rainfall modification of directional scatterometer response from the sea surface was simulated in wind-wave tank experiments. Data show that for the range of conditions in laboratory experiments, rain enhances radar cross section for all azimuthal angles relative to wind direction. This result broadens previous measurements, which showed that scatterometer response increases with increasing rainfall for radars pointing upwind. But more to the point, the data also show that the directional dynamic-range of scatterometry diminishes rapidly as rainfall rate increases. Thus, while it may be possible to determine wind speed and direction during rain, it will require adequate system sensitivity.

Description: An account of the construction of surface pressure fields from Seasat-A satellite scatterometer (SASS) winds as carried out by different methods, and the comparison of these pressure fields with those derived from in situ ship observations is presented. On the assumption that the pressure adjusts itself instantaneously to the motion field, it may be computed by various methods. One of these makes use of planetary boundary theory, and of the possible techniques in this category a two-layer iterative scheme admitting of the parametrization of diabatic and baroclinic effects and of secondary flow was chosen. A second method involves the assumption of zero two-dimensional divergence, leading to a Laplace's equation (the balance equation) in pressure, with the wind field serving as a forcing function. This method does not accommodate adiabatic or baroclinic effects, and requires a knowledge of the pressure at all boundary points. Two comparison fields are used for validation: the conventional operational analyses of the US National Meteorological Center (NMC), and the special analyses of the Gulf of Alaska Experiment (GOASEX), which were done by hand. The results of the computations were as follows: (1) The pressure fields, as computed from the SASS winds alone, closely approximated the NMC fields in regions where reasonable in situ coverage was available (typically, one or two mb differences over most of the chart, three to four mb in extreme cases); (2) In some cases the SASS-derived pressure fields displayed high-resolution phenomena not detected by the NMC fields, but evident in the GOASEX data; and, (3) As expected, the pressure fields derived from the balance equation were much smoother and less well resolved than the SASS-derived or NMC fields. The divergence as measured from the SASS winds is smaller than, but of the same order of magnitude as, the vorticity.

Description: The accuracy of temperature, pressure, potential temperature, and horizontal wind measurements is discussed in connection with the use of Meteorological Measurement System data in the AAOE. The vertical distribution of temperature measurements and latitudinal variations of the zonal wind for 12 flights over Antarctica during the 1987 AAOE campaign are summarized. Model atmospheres from 0 to 32 km at 70 deg and 55 deg S for the August-September period are constructed. Above the 420 K isentropic surface, the polar vortex remains strong throughout August and September of 1987.

Description: The formulation and evolution of polar stratospheric ice clouds are simulated using a one-dimensional model of cloud microphysics. It is found that the optical thickness and particle size of ice clouds depend on the cooling rate of the air in which the cloud formed. It is necessary that there be an energy barrier to ice nucleation upon the preexisting aerosols in order to account for the cooling rate dependence of the cloud properties.

Description: The history of minimum temperatures at 50 and 70 mb is examined from NMC, UK Met O and ECMWF analyses. MSU channel 24 data are similarly inspected. South Pole sonde data are used to calculate saturation humidity mixing ratio as a function of altitude and time throughout 1987. Saturation with respect to ice could be maintained for water mixing ratios of 3.5 ppmv for a period of about 80 days from mid-June to mid-September. Dehydration to mixing ratios of 1 ppmv or less was possible sporadically. Data from the ER-2 flights between 53 S and 72 S are used in conjunction with particle size measurements and air parcel trajectories to demonstrate the dehydration occurring over Antarctica. Water mixing ratios at the latitude of Punta Arens (53 S), in conjunction with tracer measurements and trajectory analysis, show that at potential temperatures from about 325 to 400 K, the dryness (less than 3 ppmv) had its origin over Antarctica rather than in the tropics. Water mixing ratios within the Antarctic vortex varied from 1.5 to 3.8 ppmv, with a strong isentropic gradient being evident in the region of high potential vorticity gradients.

Description: Measurements of total reactive nitrogen, NOy, total water vapor, and aerosols were made as part of the Airborne Antarctic Ozone Experiment. The measurements were made using instruments located onboard the NASA ER-2 aircrafts which conducted twelve flights over the Antarctic continent reaching altitudes of 18 km at 72 S latitude. Each instrument utilized an ambient air sample and provided a measurement up to 1 Hz or every 200 m of flight path. The data presented focus on the flights of Aug. 17th and 18th during which Polar Stratospheric Clouds (PSCs) were encountered containing concentrations of 0.5 to 1.0 micron diameter aerosols greater than 1 cm/cu. The temperature pressure during these events ranged as low as 184 K near 75 mb pressure, with water values near 3.5 ppm by volume (ppmv). With the exception of two short periods, the PSC activity was observed at temperatures above the frost point of water over ice. The data gathered during these flights are analyzed and presented.

Description: Delamination growth caused by local buckling of a delaminated group of plies was investigated. Delamination growth was assumed to be governed by the strain energy release rates, G(1), G(2) and G(3). The strain energy release rates were calculated using a geometrically nonlinear three-dimensional finite element analysis. The program is described and several checks of the analysis are discussed. Based on a limited parametric study, the following conclusions were reached: (1) the problem is definitely mixed mode (in some cases G(1) is larger than G(2), for other cases the opposite is true); (2) in general, there is a large gradient in the strain energy release rates along the delamination front; (3) the locations of maximum G(1) and G(2) depend on the delamination shape and the applied strain; (4) the mode 3 component was negligible for all cases considered; and (5) the analysis predicted that parts of the delamination would overlap. The results presented did not impose contact constraints to prevent overlapping. Further work is needed to determine the effects of allowing the overlapping.

Description: A multichannel statistical approach is used to retrieve rainfall rates from the brightness temperature T(B) observed by passive microwave radiometers flown on a high-altitude NASA aircraft. T(B) statistics are based upon data generated by a cloud radiative model. This model simulates variabilities in the underlying geophysical parameters of interest, and computes their associated T(B) in each of the available channels. By further imposing the requirement that the observed T(B) agree with the T(B) values corresponding to the retrieved parameters through the cloud radiative transfer model, the results can be made to agree quite well with coincident radar-derived rainfall rates. Some information regarding the cloud vertical structure is also obtained by such an added requirement. The applicability of this technique to satellite retrievals is also investigated. Data which might be observed by satellite-borne radiometers, including the effects of nonuniformly filled footprints, are simulated by the cloud radiative model for this purpose.

Description: This paper presents the results of a field program using a ground-based Raman lidar system to observe changes in moisture profiles as a cold and a warm front passed over the NASA/Goddard Space Flight Center in Greenbelt, Maryland. The lidar operating only during darkness is capable of providing continuous high vertical resolution profiles of water vapor mixing ratio and aerosol scattering ratio from near the surface to about 7 km altitude. The lidar data acquired on three consecutive nights from shortly after sunset to shortly before sunrise, along with upper air data from specially launched rawinsondes, have provided a unique visualization of the detailed structure of the two fronts.

Description: The propagation of elastic waves of the type induced by impact are studied for a special class of cylinders. In this class are thin finite cylinders, for which Mindlin's equations are appropriate, on which point masses are attached. An approximation is derived valid for frequencies below cutoff of the lowest shear mode. An eigenfunction expansion is used to compute the transient response of the coupled system.

Description: The method of boundary elements is adapted to the dynamics of elastic axisymmetric solids. Three scalar potentials are used, each satisfying the Helmholtz equation and each represented by its own source distribution. The sources are distributed over a surface enclosing the solid boundary. Stress or displacement conditions on the boundary are realized by third-order derivatives of the potentials. Discrete ring elements on the source enclosure combine their influences at discrete ring elements on the solid boundary.

Description: An improved version of the GISS Model II cumulus parameterization designed for long-term climate integrations is used to study the effects of entrainment and multiple cloud types on the January climate simulation. Instead of prescribing convective mass as a fixed fraction of the cloud base grid-box mass, it is calculated based on the closure assumption that the cumulus convection restores the atmosphere to a neutral moist convective state at cloud base. This change alone significantly improves the distribution of precipitation, convective mass exchanges, and frequencies in the January climate. The vertical structure of the tropical atmosphere exhibits quasi-equilibrium behavior when this closure is used, even though there is no explicit constraint applied above cloud base.

Description: While the standard mode acceleration formulation in structural dynamics has often been interpreted to suggest that the reason for improved convergence obtainable is that the dynamic correction factor is divided by the modal frequencies-squared, an alternative formulation is presented which clearly indicates that the only difference between mode acceleration and mode displacement data recovery is the addition of a static correction term. Attention is given to the advantages in numerical implementation associated with this alternative, as well as to an illustrative example.

Description: The joint frequency distribution technique was used to analyze buoyancy fluxes in the marine atmospheric boundary layer (MABL) for the cloud street regime noted during the Genesis of Atlantic Lows Experiment. It is found that for the lower half of the MABL, the buoyancy flux is mainly generated by the rising thermals and the sinking compensating ambient air, and is mainly consumed by the entrainment and detrainment of thermals, penetrative convection, and the entrainment from the MABL top. If the buoyancy flux is primarily driven by the temperature flux, these buoyancy-flux generating processes should be the same for the lower boundary layers over land and ocean. The results of the scale analysis of the buoyancy flux agree well with those obtained for mesoscale cellular convection during the Air-Mass Transformation Experiment.

Description: A catalog of 2678 galaxies within an area of about 40 sq deg centered on the Fornax Cluster has been compiled based on 26 deep large-scale plates obtained with the 2.5-m Las Campanas Observatory reflector. The catalog includes 340 likely cluster members and 2338 likely background galaxies. Radial velocities are given for 89 of the galaxies. The spatial distributions of various types of galaxies have been modeled as the sum of a King (1962) model cluster component superimposed on a uniform background. Using maximum-likelihood fits to these distributions, a core radius of 0.7 deg is found for a King model fit to the cluster, suggesting that there are few cluster members contained in the sample of background galaxies.

Description: Results are presented on numerical experiments that were carried out to investigate the mechanisms of the observed variabilities in wind and convection associated with supercloud clusters (SCCs), westerly wind bursts, and 30-60 day oscillations in the western Pacific region. It is shown that the generation of a 30-60 day eastward propagating precipitation pattern in the Lau and Peng (1987) model, which can be identified as SCC, is accompanied by convective clusters coming in opposite direction to that of the SCC itself. The results suggest that the westward propagating cloud clusters are produced at the initial stage of the 30-60 day disturbance due to mutual adjustment of the large-scale flow and heating.

Description: Six radio telescopes were operated as the first Southern Hemisphere VLBI array in April and May 1982. Observations were made at 2.3 and 8.4 GHz. This array provided VLBI modeling and hybrid imaging of celestial radio sources in the Southern Hemisphere, high-accuracy VLBI geodesy between Southern Hemisphere sites, and subarcsecond radio astrometry of celestial sources south of declination -45 deg. The goals and implementation of the array are discussed, the methods of modeling and hybrid image production are explained, and the VLBI structure of the sources that were observed is summarized.

Description: A key step in the application of formal automated design techniques to structures under transient loading is the calculation of sensitivities of response quantities to the design parameters. This paper considers response quantities to the design parameters. This paper considers structures with general forms of damping acted on by general transient loading and addresses issues of computational errors and computational efficiency. The equations of motion are reduced using the traditional basis of vibration modes and then integrated using a highly accurate, explicit integration technique. A critical point constraint formulation is used to place constraints on the magnitude of each response quantity as a function of time. Three different techniques for calculating sensitivities of the critical point constraints are presented. The first two are based on the straightforward application of the forward and central difference operators, respectively. The third is based on explicit differentiation of the equations of motion. Condition errors, finite difference truncation errors, and modal convergence errors for the three techniques are compared by applying them to a simple five-span-beam problem. Sensitivity results are presented for two different transient loading conditions and for both damped and undamped cases.

Description: A new and simple method of finite-element grid improvement is presented. The objective is to improve the accuracy of the analysis. The procedure is based on a minimization of the trace of the stiffness matrix. For a broad class of problems this minimization is seen to be equivalent to minimizing the potential energy. The method is illustrated with the classical tapered bar problem examined earlier by Prager and Masur. Identical results are obtained.

Description: Assuming a single-mode transverse displacement, a simple formula is derived for the transverse load-displacement relationship of a plate under in-plane compression. The formula is used to derive a simple analytical expression for the nonlinear dynamic response of postbuckled plates under sinusoidal or random excitation. The highly nonlinear motion of snap-through can be easily interpreted using the single-mode formula. Experimental results are obtained with buckled and cylindrical aluminum panels using discrete frequency and broadband excitation of mechanical and acoustic forces. Some important effects of the snap-through motion on the dynamic response of the postbuckled plates are described. Static tests were used to identify the deformation shape during snap-through.

Description: An experimental investigation of composite to composite bonded joints was undertaken to study the effect of bond thickness on debond growth rate under cyclic loading and critical strain energy release rate under static loading. Double cantilever beam specimens of graphite/epoxy adherends bonded with EC 3445 were tested under mode I loading. A different behavior of fracture and fatigue strength was observed with variation of bondline thickness.

Description: Results from IRAS and the technical details of the IRAS satellite are reviewed. The IRAS telescope operation is briefly described. The observation of the Vega phenomenon, or the orbiting clouds of dust which encircle main sequence stars, is discussed. Also, the discovery that a large number of solar-type stars have excess IR radiation due to orbiting shells or disks of solid material is examined. In addition, the prospects for future telescopes are considered.

Description: Multidecadal time series of surface winds from central tropical Pacific islands are used to compute trends in the trade winds between the end of WWII and 1985. Over this period, averaged over the whole region, there is no statistically significant trend in speed or zonal or meridional wind (or pseudostress). However, there is some tendency, within a few degrees of the equator, toward weakening of the easterlies and increased meridional flow toward the equator. Anomalous conditions subsequent to the 1972-73 ENSO event make a considerable contribution to the long-term trends. The period 1974-80 has been noted previously to have been anomalous, and trends over that period are sharply greater than those over the longer records.

Description: The feasibility of using a viscoplastic model developed by Robinson to perform a nonlinear structural analysis was investigated. The paper presents analytical solutions for three classical problems: (1) a pressurized thick-walled cylinder, (2) a thin rotating disk, and (3) a pressurized thick-walled sphere. The analytical expressions derived for the stress and the strain rates for these components are general in nature as they consider both the mechanical and thermal loadings to be time-dependent. A computer program VISTAN (VIscoplastic STructural ANalyzer) was developed to obtain the stress and strain distributions. The finite element solutions for these problems are also presented. The numerical results pertaining to isothermal loading conditions are provided. The results obtained demonstrate the feasibility of using the viscoplastic model developed by Robinson to perform nonlinear structural analyses.

Description: This paper presents an analysis of the uncertainties expected in vertical velocities using a vertically pointing airborne Doppler radar which has a nadir or zenith-pointing beam. To examine the expected uncertainty, the Doppler velocity equation for a moving platform is derived and it is applied to cases of nadir-fixed and stabilized beams. The main emphasis of the paper is on the effect of platform stability on the deduced vertical air motions and it is shown that the antenna must be stabilized to obtain desired accuracy in the vertical velocity measurements.

Description: Thirteen lightcurves of asteroid 3 Juno from three different oppositions are given. The pole of Juno is less than 10 deg from ecliptic longitude 104 deg and latitude + 36 deg (or 316 and + 62 deg). The sidereal period is 0.3003969 + or - 0.0000003 (1 sigma) day and the rotation is prograde. There is little similarity between the Juno lightcurves from 8 oppositions which is unexplained at this time.

Description: A combined experimental and analytical study of bearing and bypass loading on single-fastener specimens of a 16-ply, quasi-isotropic T300/5208 graphite/epoxy laminate with a centrally located hole is reported. The specimens were loaded in either tension or compression, and onset damage, ultimate strengths, and corresponding failure modes were determined. The tension data showed the expected linear interaction for combined bearing-bypass loading with damage developing in the net-section tension mode. However, the bearing-onset strengths showed an unexpected interaction of the bearing and compressive bypass loads in which the latter reduced the bearing-onset strength. A linear finite element analysis showed that bearing-bypass loading had a marked influence on the bolt-hole contact which in turn had a significant effect on local stresses.

Description: Monthly fields of shortwave radiation (SR) and latent heat flux (LE) over the central and eastern tropical Pacific between 1980 and 1983 have been computed using satellite data. They are the dominant variable components of surface thermal forcing on the ocean in this time scale. During the 1982-1983 ENSO episode, surface-wind convergence and cloudiness associated with the displacement of equatorial organized convection caused a reduction in both the SR into the ocean and the LE out of the ocean. The lag-correlation coefficients between the forcing (SR-LE) and the sea surface temperature are found to be significantly high outside the equatorial region, showing that surface thermal forcing is the dominant factor in sea surface temperature change. In the narrow equatorial wave guide, ocean dynamics play a more important role, and surface heat flux is a consequence rather than the cause of sea surface temperature change.

Description: An aircraft experiment has been conducted with a dual-frequency (10 GHz and 35 GHz) radar/radiometer system and an 18-GHz radiometer to test various rain-rate retrieval algorithms from space. In the experiment, which took place in the fall of 1988 at the NASA Wallops Flight Facility, VA, both stratiform and convective storms were observed. A ground-based radar and rain gauges were also used to obtain truth data. An external radar calibration is made with rain gauge data, thereby enabling quantitative reflectivity measurements. Comparisons between path attenuations derived from the surface return and from the radar reflectivity profile are made to test the feasibility of a technique to estimate the raindrop size distribution from simultaneous radar and path-attenuation measurements.

Description: Azimuthal response of a scatterometer to radiation scattered by the sea surface was studied in a wind-wave tank. The variation of the normalized radar cross section with the azimuth angle is fitted by a three-term series. Results show that the upwind-downwind asymmetry decreases as the wind speed increases. The crosswind modulation depends on the wind velocity. The results show that the evolution of the long-wind-crosswind ratio evolves with wind speed in a manner similar to the evolution of the isotropy of short capillary-gravity waves. The maximum of the isotropy of the short wind waves is obtained for wind velocities close to 4 m/s. For the same value of the velocity, the variations of radar response between long-wind and crosswind directions is minimum. For lower or higher values of wind velocities the directional accuracy of the radar increases, since the wind-wave field tends to align in the wind direction.

Description: The present interpretation of the radar cross section sigma exp 0 measured by satellite altimeters implies that the rms wave slope gamma is controlled solely by the local wind. However, parameters of wave spectra, including the exponent in the power law for the equilibrium range, depend on sea maturity. The latter is characterized by the nondimensional fetch, x = gX/U-squared. Consequently, gamma and sigma exp 0 are controlled by both U and the wind fetch X. Geosat data for one year are used jointly with in-situ wind and wave observations to assess the fetch-related error trend in altimeter wind speeds. This trend results in overestimated winds in the regions and seasons characterized by a high x, and vice versa. A procedure for wind speed retrieval based on processing sigma exp 0 jointly with the significant wave height information contained in the altimeter wave forms is proposed.

Description: A finite element formulation is presented for determining the large-amplitude free and steady-state forced vibration response of arbitrarily laminated anisotropic composite rectangular thin plates. The nonlinear stiffness and harmonic force matrices of an arbitrarily laminated composite rectangular plate element are developed for nonlinear free and forced vibration analyses. The linearized updated-mode method with nonlinear time function approximation is employed for the solution of the system nonlinear eigenvalue equations. The amplitude-frequency relations for convergence with gridwork refinement, different boundary conditions, aspect ratios, lamination angles and number of plies are presented. The finite element results are compared with available approximate continuum solutions.

Description: The dynamics of the Pluto-Charon system are reviewed from a historical perspective. Although Pluto's orbit crosses Neptune's, an intricate system of nested resonances keeps these planets apart. Pluto's orbit is apparently chaotic as well. Pluto always keeps the same face turned toward Charon, and vice versa. Tides also damp Charon's orbital eccentricity and inclination. Precession of Pluto's orbital plane causes Pluto's obliquity to vary periodically from formally prograde to retrograde. Pluto is probably an original member of the Solar system, but not an escaped satellite of Neptune.

Description: Seasat scatterometer data over the Arabian Sea are used to build wind-stress fields during July and August 1978. They are first compared with 3-day wind analyses from ship data along the Somali coast. Seasat scatterometer specifications of 2-m/s and 20-deg accuracy are fulfilled in almost all cases. The exceptions are for winds stronger than 14 m/s, which are underestimated by the scatterometer by 15 percent. Wind stress is derived from these wind data using a bulk formula with a drag coefficient depending on the wind intensity. A successive-correction objective analysis is used to build the wind-stress field over the Arabian Sea with 2 x 2-deg and 6-day resolution. The final wind-stress fields are not significantly dependent on the objective analysis because of the dense coverage of the scatterometer. The combination of scatterometer and coastal ship data gives the best coverage to resolve monsoon wind structures even close to the coast. The final wind stress fields show wind features consistent with other monthly mean wind stress field. However, a high variability is observed on the 6-day time scale.

Description: The objectives are to present the derivation of the new virtual crack closure technique, evaluate the accuracy of the technique, and finally to present the results of a limited parametric study of laminates with a postbuckled delamination. Although the new virtual crack closure technique is general, only homogeneous, isotropic laminates were analyzed. This was to eliminate the variation of flexural stiffness with orientation, which occurs even for quasi-isotropic laminates. This made it easier to identify the effect of geometrical parameters on G. The new virtual crack closure technique is derived. Then the specimen configurations are described. Next, the stress analyses is discussed. Finally, the virtual crack closure technique is evaluated and then used to calculate the distribution of G along the delamination front of several laminates with a postbuckled delamination.

Description: A description of the finite element implementation of Robinson's unified viscoplastic model into the General Purpose Finite Element Program (MARC) is presented. To demonstrate its application, the implementation is applied to some uniaxial and multiaxial problems. A comparison of the results for the multiaxial problem of a thick internally pressurized cylinder, obtained using the finite element implementation and an analytical solution, is also presented. The excellent agreement obtained confirms the correct finite element implementation of Robinson's model.

Description: It is shown that gamma-ray burst detectors on a network of microspacecraft can be used to directly determine burst directions to a few arcsec in both angular coordinates. This level of accuracy is sufficient to conduct searches for the quiescent counterpart of the burster at other wavelengths. It is maintained that the strawman instrument for the microspacecraft network, consisting of two 1 kg Na I detectors on opposite ends of the spacecraft, is capable of providing a nearly omnidirectional FOV with adequate sensitivity to gamma-ray bursts.

Description: Cumulus processes involved in the interaction and merging of clouds under the influence of different imposed conditions (including large-scale lifting forcing, environmental wind shear, and cloud microphysical processes) were studied using simulations with a three-dimensional model. The design of the study was to generate several convective clouds randomly inside the model domain, and then to observe and analyze the interactions and merging between the simulated clouds. Ten merged clouds were identified. Seven of these, each involving two previously separated clouds, generally lie along a line parallel to the initial environmental wind shear vector, while one (also a two-cloud system) lies along a line perpendicular to the wind shear vector prior to merging. The remaining two merging systems involve three parent clouds each; they are a combination of parallel and perpendicular cells. The merging mechanisms associated with three-cloud merging cases are studied by examining the temperature, pressure, and wind fields prior to, during, and following the merging.

Description: In the dynamic formulation of holonomic and nonholonomic systems based on D'Alembert-Lagrange equation, the forces of constraints are maintained in the dynamic equations by introducing auxiliary variables, called Lagrange multipliers. This approach introduces a set of generalized reaction forces associated with the system generalized coordinates. Different sets of variables can be used as generalized coordinates and accordingly, the generalized reactions associated with these generalized coordinates may not be the actual reaction forces at the joints. In rigid body dynamics, the generalized reaction forces and the actual reaction forces at the joints represent equipollent systems of forces since they produce the same total forces and moments at and about any point on the rigid body. This is not, however, the case in deformable body analyses wherein the generalized reaction forces depend on the system generalized reference and elastic coordinates. In this paper, a method for determining the actual reaction forces at the joints from the generalized reaction forces in deformable multibody systems is presented.

Description: The possibility of initiating the growth of ice sheets by solar insolation variations is examined. The study is conducted using a climate model with three different orbital configurations corresponding to 116,000 and 106,000 yr before the present and a modified insolation field with greater reductions in summer insolation at high northern latitudes. Despite the reduced summer and fall insolation, the model fails to maintain snow cover through the summer at locations of suspected ice sheet initiation. The results suggest that there is a discrepancy between the model's response to Milankovitch perturbations and the geophysical evidence of ice sheet initiation. If the model results are correct, the growth of ice shown by geophysical evidence would have occurred in an extremely ablative environment, demanding a complicated strategy.

Description: The Boundary Force Method (BFM), a form of indirect boundary element method, is used to analyze composite laminates with cracks. The BFM uses the orthotropic elasticity solution for a concentrated horizontal and vertical force and a moment applied at a point in a cracked, infinite sheet as the fundamental solution. The necessary stress functions for this fundamental solution were formulated using the complex variables theory of orthotropic elasticity. The current method is an improvement over a previous method using only forces and no moment. The improved method was verified by comparing it to accepted solutions for a finite-width, center-crack specimen subjected to uniaxial tension. Four graphite/epoxy laminates were used: (0 + or - 45/90)sub s, (0), (+ or - 45)sub s, and (+ or - 30)sub s. The BFM results agreed well with accepted solutions. Convergence studies showed that with the addition of the moment in the fundamental solution, the number of boundary elements required for a converged solution was significantly reduced. Parametric studies were done for two configurations for which no orthotropic solutions are currently available; a single edge crack and an inclined single edge crack.

Description: Thirty-three Landsat TM scenes of California stratocumulus cloud fields were acquired as part of the FIRE Marine Stratocumulus Intensive Field Observations in July 1987. They exhibit a wide variety of stratocumulus structures. Analysis has so far focused upon the July 7 scene, in which aircraft from NASA, NCAR, and the British Meteorological Office repeatedly gathered data across a stratocumulus-fair weather cumulus transition. The aircraft soundings validate the cloud base temperature threshold determined by spatial coherence analysis of the TM thermal band. Brightness variations in the stratocumulus region exhibit a -5/3 power-law decrease of the wavenumber spectra for scales larger than the cloud thickness, about 200 m, changing to a -3 power at smaller scales. Observations by an upward-looking three-channel microwave radiometer on San Nicolas Island also show the -5/3 power-law in total integrated liquid water, suggesting that the largest-scale TM brightness variations are primarily due to variations in the liquid water. The Kolmogorov 5/3 power suggests that for some purposes liquid water in turbulent stratocumulus clouds may be treated as a passive scalar, simply reflecting variations in vertical velocity. This may be tested using the velocities measured by the aircraft.

Description: The fundamental relationship between the morphology of a composite laminate and the resulting free edge effects is explored and related to delamination failures. Cross-ply, angle-ply, and quasi-isotropic laminates are discussed in detail. It is shown that the local mismatch in elastic properties of adjacent layers and the global stacking sequence of a laminate both have a significant influence on the interlaminar stresses and delamination failures.

Description: An analysis is provided for the capabilities of the Spacewatch Camera operated at Kitt Peak National Observatory by the University of Arizona for the detection of the large flux of small comets in the vicinity of earth that is proposed by Frank et al. (1986). The characteristics of this telescope are used to determine its optimum operational mode for the detection of these small comets. It is concluded that the Spacewatch Camera can detect these small comets near earth if the proposed large fluxes in prograde orbits are present. It is unlikely that previous optical telescope surveys would have detected these objects. A heretofore undetected, large flux of small objects was detected. The characteristics of these objects appear to be in general agreement with those of the comets proposed by Frank et al. (1986). No other known population of objects or other effects that could provide the observed optical signatures has been identified at this time. Because these initial results are of such great importance to our perception of the solar system, they should be conservatively treated as an initial result until confirmation is made with similar observations from other telescopes.

Description: New nonlinear partial differential equations containing terrain curvature and its rate of change are derived that describe the flow of an atmospheric density current. Unlike the classical hydraulic-type equations for density currents, the new equations are valid for two-dimensional, gradually varied flow over highly curved terrain, hence suitable for computing unsteady (or steady) flows over arbitrary mountain/valley profiles. The model assumes the atmosphere above the density current exerts a known arbitrary variable pressure upon the unknown interface. Later this is specialized to the varying hydrostatic pressure of the atmosphere above. The new equations yield the variable velocity distribution, the interface position, and the pressure distribution that contains a centrifugal component, often significantly larger than its hydrostatic component. These partial differential equations are hyperbolic, and the characteristic equations and characteristic directions are derived. Using these to form a characteristic mesh, a hypothetical unsteady curved-flow problem is calculated, not based upon observed data, merely as an example to illustrate the simplicity of their application to unsteady flows over mountains.

Description: A method is developed for the optimal design of composite links based on dynamic performance criteria directly related to structural modal damping and dynamic stiffness. An integrated mechanics theory correlates structural composite damping to the parameters of basic composite material systems, laminate parameters, link shape, and modal deformations. The inclusion of modal properties allows the selective minimization of vibrations associated with specific modes. Ply angles and fiber volumes are tailored to obtain optimal combinations of damping and stiffness. Applications to simple composite links indicate wide margins for trade-offs and illustrate the importance of various design variables to the optimal design.

Description: The IRAS survey detectors show nonlinear characteristics in their gains and baselines. This behavior can be described by an exponentially decaying gain enhancement, dependent on previously observed flux. The baseline behavior has been modeled by a decaying function composed of the sum of two exponential functions. The scale lengths are 10 arcmin and 1.1 deg for 12 microns, and 8 arcmin and 43 arcmin for 25 microns. Amplitude changes due to this type of hysteresis are less than 1 percent. In addition there is a considerable response of all detectors to particles associated with the South Atlantic Anomaly. Attempts to correct for this effect were performed on the 12- , 60- and 100-micron detectors in the form of bias boosts. This note presents a correction procedure for the 25-micron detectors.

Description: Local buckling can cause large interlaminar stresses along the delamination front, which can lead to delamination growth. This paper examines several methods of calculating strain-energy release rates, which are often used to predict delamination growth. The thin-film plate analysis, which was least expensive, calculated the total strain-energy release rate G(T) quite accurately. However, the stress field along the delamination front is highly mixed-mode and has no fixed ratio of G(I) to G(II). Since plate analysis can only calculate G(T), it would not be useful for accurate predictions of delamination growth if mode mix is important.

Description: New optical spectroscopic and UBV photometric observations of the bright Be/shell star V644 Mon (HD 51480) are presented. The object, which has been described as an interacting binary system, exhibits strong, variable Balmer emission as well as numerous metallic emission features in the blue. No signs of absorption features due to any late-type companion are seen at wavelengths below 6500 A.

Description: The sources of sub-Saharan precipitation are studied using diagnostic procedures integrated into the code of the GISS climate model. Water vapor evaporating from defined source regions is 'tagged', allowing the determination of the relative contributions of each evaporative source to the simulated July rainfall in the Sahel. Two June-July simulations are studied to compare the moisture sources, moisture convergence patterns and the spatial variations of precipitation for rainy and drought conditions. Results for this case study indicate that patterns of moisture convergence and divergence over northern Africa had a stronger influence on model rainfall over the sub-Sahara than did evaporation rates over the adjacent oceans or moisture advection from ocean to continent. While local continental evaporation contributed significant amounts of water to Sahelian precipitation in the'rainy' simulation, moisture from the Indian Ocean did not precipitate over the Sahel in either case.

Description: The long-term climatic evolution of the earth has been studied on the basis of one-dimensional, globally-averaged climate models yielding only a qualitative understanding of climatic history, and in any case proceeding under a series of potentially invalid assumptions. One such major assumption, which invites comparison with a three-dimensional GCM, is that of fixed relative humidity. A GCM may also be used to study the problem of water loss from both the earth and Venus.

Description: Earlier airborne studies of the infrared bands between 5 and 8 microns have now been extended to a sample of southern sources selected from the IRAS Low Resolution Spectra (LRS) atlas. The correlation between the strongest bands at 6.2 and 7.7 microns is now based on a total sample of 40 sources and is very strong. A new emission band at 5.2 microns, previously predicted for polycyclic aromatic hydrocarbons (PAHs), is recognized in 27 sources; it too correlates with the dominant 7.7 micron band, showing that the 5.2 micron feature also belongs to the generic spectrum of PAH features at 3.3, 5.6, 6.2, 6.2, 7.7, 8.7, 11.3, and 12.7 microns. Sufficient sources are had now to define the relative strengths of most of these bands in three separate nebular environments: planetaries, H II regions, and reflection nebulae. Significant variations are detected in the generic spectra of PAHs in these different environments which are echoed by variations in the exact wavelength of the strong 7.7 micron peak. The earlier suggestion that, in planetaries, the fraction of total emission observed by IRAS that is carried by the PAH emissions is correlated with nebular gas-phase C/O ratio is supported by the addition of newly-observed southern planetaries, including the unusually carbon-rich (WC10) nebular nuclei. These (WC10) nuclei also exhibit a strong plateau of emission linking the 6.2 and 7.7 micron features.

Description: High resolution spectra of the 11.3 micron emission band in M82 and NGC 7027 were obtained using the University of Texas IR echelle spectrometer on the IRTF in April 1988. The spectral resolution was 0.004 micron, with coverage from 11.0 to 11.6 microns. Spectra were measured at ten positions along a 10 min. long slit. Analysis of the data is still in progress, but initial results show no clear evidence of narrow structure within the feature. The analysis will involve comparison of the observed spectra to laboratory and predicted spectra of Polycylic Aromatic Hydrocarbons (PAHs) and Quenched Carbonaceous Composite (QCCs) to determine which may be responsible for the emission. The spectra will be examined with a goal of determining whether the emission is caused by molecular or solid state material. The data are also examined for evidence of variations in the shape and strength of the 11.3 micron feature with position on the sky. In NGC 7027 the 10 min. long slit went across the edge of the ionized nebulae, allowing comparison of emission from both ionized and neutral regions.

Description: Images were obtained of the (fluorescent) molecular hydrogen 1-0 S(1) line, and of the 3.3 micron emission feature, in Orion's Bar and three reflection nebulae. The emission from these species appears to come from the same spatial locations in all sources observed. This suggests that the 3.3 micron feature is excited by the same energetic UV-photons which cause the molecular hydrogen to fluoresce.

Description: Over the years, NASA has been conducting the Advanced Earth-to-Orbit (AETO) Propulsion Technology Program to provide the knowledge, understanding, and design methodology that will allow the development of advanced Earth-to-orbit propulsion systems with high performance, extended service life, automated operations, and diagnostics for in-flight health monitoring. The objective of the Aerothermodynamic Loads Definition Study is to develop methods to more accurately predict the operating environment in AETO propulsion systems, such as the Space Shuttle Main Engine (SSME) powerhead. The approach taken consists of 2 parts: to modify, apply, and disseminate existing computational fluid dynamics tools in response to current needs and to develop new technology that will enable more accurate computation of the time averaged and unsteady aerothermodynamic loads in the SSME powerhead. The software tools are detailed. Significant progress was made in the area of turbomachinery, where there is an overlap between the AETO efforts and research in the aeronautical gas turbine field.

Description: The purpose of the Probabilistic Structural Analysis Method (PSAM) project is to develop structural analysis capabilities for the design analysis of advanced space propulsion system hardware. The boundary element method (BEM) is used as the basis of the Probabilistic Advanced Analysis Methods (PADAM) which is discussed. The probabilistic BEM code (PBEM) is used to obtain the structural response and sensitivity results to a set of random variables. As such, PBEM performs analogous to other structural analysis codes such as finite elements in the PSAM system. For linear problems, unlike the finite element method (FEM), the BEM governing equations are written at the boundary of the body only, thus, the method eliminates the need to model the volume of the body. However, for general body force problems, a direct condensation of the governing equations to the boundary of the body is not possible and therefore volume modeling is generally required.

Description: A major research and technology program in Probabilistic Structural Analysis Methods (PSAM) is currently being sponsored by the NASA Lewis Research Center with Southwest Research Institute as the prime contractor. This program is motivated by the need to accurately predict structural response in an environment where the loadings, the material properties, and even the structure may be considered random. The heart of PSAM is a software package which combines advanced structural analysis codes with a fast probability integration (FPI) algorithm for the efficient calculation of stochastic structural response. The basic idea of PAAM is simple: make an approximate calculation of system response, including calculation of the associated probabilities, with minimal computation time and cost, based on a simplified representation of the geometry, loads, and material. The deterministic solution resulting should give a reasonable and realistic description of performance-limiting system responses, although some error will be inevitable. If the simple model has correctly captured the basic mechanics of the system, however, including the proper functional dependence of stress, frequency, etc. on design parameters, then the response sensitivities calculated may be of significantly higher accuracy.

Description: Two-dimensional optical strain measurements on high temperature test specimens are presented. This two-dimensional capability is implemented through a rotatable sensitive strain axis. Three components of surface strain can be measured automatically, from which the first and second principal strains are calculated. One- and two-dimensional strain measurements at temperatures beyond 750 C with a resolution of 15 microstrain are demonstrated. The system is based on a one-dimensional speckle shift technique. The speckle shift technique makes use of the linear relationship between surface strain and the differential shift of laser speckle patterns in the diffraction plane. Laser speckle is a phase effect that occurs when spatially coherent light interacts with an optically rough surface. Since speckle is generated by any diffusely reflecting surface, no specimen preparation is needed to obtain a good signal. Testing was done at room temperature on a flat specimen of Inconel 600 mounted in a fatigue testing machine. A load cell measured the stress on the specimen before and after acquiring the speckle data. Strain components were measured at 0 C (parallel to the load axis) and at plus or minus 45 C, and plots indicate the calculated values of the first and second principal strains. The measured values of Young's modulus and Poisson's ratio are in good agreement with handbook values. Good linearity of the principal strain moduli at high temperatures indicate precision and stability of the system. However, a systematic error in the high-temperature test setup introduced a scale factor in the slopes of the two-dimensional stress-strain curves. No high temperature effects, however, have been observed to degrade speckle correlation.

Description: The fusion of the probabilistic finite element method (PFEM) and reliability analysis for probabilistic fracture mechanics (PFM) is presented. A comprehensive method for determining the probability of fatigue failure for curved crack growth was developed. The criterion for failure or performance function is stated as: the fatigue life of a component must exceed the service life of the component; otherwise failure will occur. An enriched element that has the near-crack-tip singular strain field embedded in the element is used to formulate the equilibrium equation and solve for the stress intensity factors at the crack-tip. Performance and accuracy of the method is demonstrated on a classical mode 1 fatigue problem.

Description: The probabilistic structural analysis method (PSAM) was developed to analyze the effects of fluctuating loads, variable material properties, and uncertain analytical models especially for high performance structures such as the Space Shuttle Main Engine turbopump blades. Risk is calculated after expensive service experience. However, probabilistic structural analysis provides a rational alternative method to quantify uncertainties in the structural performance and durability. NESSUS (Numerical Evaluation of Stochastic Structures Under Stress) was developed as a probabilistic structural analysis computer code which integrates finite element methods and reliability algorithms, capable to predicting the probability distributions of structural response variables such as stress, displacement, natural frequencies, and buckling loads. This computer code is detailed.

Description: The Probabilistic Structural Analysis Methods (PSAM) project developed at the Southwest Research Institute integrates state-of-the-art structural analysis techniques with probability theory for the design and analysis of complex large-scale engineering structures. An advanced efficient software system (NESSUS) capable of performing complex probabilistic analysis has been developed. NESSUS contains a number of software components to perform probabilistic analysis of structures. These components include: an expert system, a probabilistic finite element code, a probabilistic boundary element code and a fast probability integrator. The NESSUS software system is shown. An expert system is included to capture and utilize PSAM knowledge and experience. NESSUS/EXPERT is an interactive menu-driven expert system that provides information to assist in the use of the probabilistic finite element code NESSUS/FEM and the fast probability integrator (FPI). The expert system menu structure is summarized. The NESSUS system contains a state-of-the-art nonlinear probabilistic finite element code, NESSUS/FEM, to determine the structural response and sensitivities. A broad range of analysis capabilities and an extensive element library is present.

Description: Constitutive and life prediction models are developed and verified for materials typically used in hot gas path components of reusable space propulsion systems over the range of relevant operating environments. The efforts were centered on the development of crack initiation life prediction methods, while the efforts of a counterpart group were centered on the development of cyclic crack propagation life prediction methods. The complexion of the active tasks are presented. A significant new task started this year will incorporate the various material constitutive and life prediction models developed in this program into a comprehensive creep-fatigue damage analysis and life assessment computer code. The program will function as a postprocessor to general structural analysis programs (such as finite element or boundary element codes) using the output of such analyses (stress, strain, and temperature fields as functions of time) as the input to the damage analysis and life assessment code. The code will be designed to execute on engineering/scientific workstations and will feature a windowing, mouse-driven user interface. Current plans call for the code to be finished and made available for use in mid 1991.

Description: An analysis of the relationship between the IMF section boundary crossing, solar flares, the sunspot 11 year cycle variation and the thunderstorm index is given, using the superposition epoch method, for data from more than 13,000 thunderstorms from 10 meteorological stations in the Beijing area and the Northeast region during 1957 to 1978. The results show that for some years a correlation exists between the thunderstorm index and the positive IMF section boundary crossing. The thunderstorm index increases obviously within three days near the crossing and on the seventh day after the crossing. The influence of the crossing on thunderstorms is stronger in the first half year than the latter half year. For different classes of solar flares, the influences are not equally obvious. The solar flares which appeared on the west side, especially in the western region (from 0 to 30 deg) have the most obvious influence. There is no discernible correlation between the thunderstorm index and the sunspot eleven-year cycle.

Description: The possibility that solar variations associated with the 11-year solar cycle may be the cause of the changes in tropospheric weather and climate has been the subject to scientific investigation for several decades. Meteorologists are greatly concerned with the changes in tropospheric phenomena. An attempt was made to find solar activity related changes in tropospheric weather, by the modulation of the quasi-biennial oscillation (QBO) of zonal wind at 50 mb. Rainfall and surface temperature data for a period of about three solar cycles, 1953 to 1988, from various stations in the Indian subcontinent were utilized. By extension, a possible teleconnection was looked for between the temperature changes in middle atmospheric levels and surface temperature when the data are stratified according to east or west phase of the QBO. The temperature data were averaged for January and February to represent the winter temperature and for July and August to represent the summer temperature.

Description: The part of energy of the planetary waves which enters the stratosphere depends on conditions of planetary wave generation and propagation through the tropopause, and the part of planetary wave energy which enters the mesosphere depends on conditions of planetary wave propagation through the stratopause. An attempt is made to estimate connections between extratropical middle atmosphere temperature long term variations and portions of energy of planetary waves which enter the mesosphere and stratosphere during winter seasons in Northern and Southern Hemispheres. Interannual variations of temperatures at the 30 km and 70 km levels are investigated for the central winter months of the period 1970 to 1986. This period includes the descending branch of the 20th solar cycle and the whole 21st cycle. Calculations are made on the basis of measurements at Heiss Island and Molodezhnaya.

Description: The direct modulation of temperature of the mid-latitude mesosphere by the solar-cycle EUV variation, which leads to greater heat input at higher solar activity, is well established. Middle atmosphere temperature modulation by the solar cycle is independently confirmed by the variation of reflection heights of low frequency radio waves in the lower ionosphere, which are regularly monitored over about 30 years. As explained elsewhere in detail, these reflection heights depend on the geometric altitude of a certain isobaric surface (near 80 k), and on the solar ionizing Lyman-alpha radiation flux. Knowing the solar cycle variation of Lyman-alpha how much the measured reflection heights would be lowered with the transition from solar minimum to maximum can be calculated, if the vertical baric structure of the neutral atmosphere would remain unchanged. An discrepancy between expected and observed height change must be explained by an uplifting of the isobaric level from solar minimum to maximum, caused by the temperature rise in the mesosphere. By integrating the solar cycle temperature changes over the height region of the middle atmosphere, and assuming that the lower boundary (tropopause) has no solar cycle variation, the magnitude of this uplifting can be estimated. It is given for the Lidar-derived and for the rocket-measured temperature variations. Comparison suggests that the real amplitude of the solar cycle temperature variation in the mesosphere is underestimated when using the rocket data, but probably overestimated with the Lidar data.

Description: Before the introduction of the Quasi Biennial Oscillation (Q.B.O.) in the study of the solar atmosphere relationship by Labitzke (1987) and Labitzke and Van Loon (1988), the only region of the atmosphere where an effect of a change in solar activity was generally admitted was the mesosphere. The response of the mesosphere, in phase with the solar activity, was found to be about one order of magnitude above model expectancy (around 10 to 20 Kelvin). It was observed independently of the season and maximized around 70 km (Chanin et al. 1987). However, from the same study, it was shown that the response of the stratosphere of opposite sign, clearly seen during winter and autumn, was at the threshold of detection in spring and summer. In the stratosphere, it was shown later that the separation of the data taking into account the sign of the Q.B.O. amplifies the negative correlation of the stratospheric temperature with solar activity in winter; it then becomes more significantly negative for the East phase of the Q.B.O. than when the data are all mixed (Labitzke and Chanin 1988). The studies of the seasonal response of the atmosphere to solar effect is crucial to understand the possible mechanism responsible of such a solar activity Q.B.O. relationship, knowing that the global dynamic circulation is quite different according to the seasons. The question is examined as to whether such separation of the data according to the phase of the Q.B.O. has any impact on the solar response of the middle atmosphere for seasons other than winter.

Description: There is no doubt that the antropogenic effect play an important role in the effects of corpuscular radiation on weather and climate. The task, however, is to distinguish between antropogenic effect in the atmosphere due to human activities and natural climatic fluctuations influencing biological systems. The increase in global temperature during the past 100 years is in relatively good coincidence with the increase in geomagnetic (corpuscular) activity. It is concluded that it could have been the increase in temperature on the Northern Hemisphere, due to the processes occurring in the auroral oval under enhanced corpuscular radiation which led to an increased atmospheric concentration of CO2 in the past. Both processes, i.e., antropogenic and solar activity effects, should be therefore intensively studied due to their important role for elucidating the past and present global change mainly in temperature, climate and biological systems.

Description: A blocking pattern which formed over eastern North America following the landfall of Hurricane Juan during November 1985 was investigated. It is hypothesize that latent heat released in the Hurricane's rainfall was either directly or indirectly responsible for the large observed 500 mb height rises over eastern Canada during the formation of this block. This idea is evaluated with a diagnostic model for the height tendency field which includes latent heat release as a forcing function. The total column heating is calculated using satellite-derived precipitation estimates. These estimates are qualitatively congruent with observations, but overestimate light rainfall and underestimate heavy rainfall. The calculations reveal that the direct contribution of the heating to the 500 mb height tendency field is small relative to the quasigeostrophic forcing. However, maxima in heating coincide with regions where anticyclonic potential vorticity is generated. Once such region is just upstream of the location of large 500 mb height rises in the incipient block. An indirect role is proposed for the heating in this case. Specifically, anticyclonic potential vorticity is generated near the heating maxima; this vorticity is then advected downstream, forcing the 500 mb heights to rise and the block to develop.

Description: Simple models are being developed to simulate interaction of planetary and synoptic-scale waves incorporating the effects of large-scale topography; eddy heat and momentum fluxes (or nonlinear dynamics); radiative heating/cooling; and latent heat release (precipitation) in synoptic-scale waves. The importance of latent heat release is determined in oceanic storm tracks for temporal variability and time-mean behavior of planetary waves. The model results were compared with available observations of planetary and synoptic-scale wave variability and time-mean circulation. The usefulness of monitoring precipitation in oceanic storm tracks by satellite observing systems was ascertained. The modeling effort includes two different low-order quasi-geostrophic models-time-dependent version and climatological mean version. The modeling also includes a low-order primitive equation model. A time-dependent, multi-level version will be used to validate the two-level Q-G models and examine effects of spherical geometry.

Description: In structural analysis the amount of computational time necessary for a solution is proportional to the number of degrees of freedom times the bandwidth squared. In implicit time analysis, this must be done at each discrete point in time. If, in addition, the problem is nonlinear, then this solution must be iterated at each point in time. If the bandwidth is large, the size of the problem that can be analyzed is severely limited. The multi-grid method is a possible algorithm which can make this solution much more computationally efficient. This method has been used for years in computational fluid mechanics. It works on the fact that relaxation is very efficient on the high frequency components of the solution (nearest neighbor interactions) and not very good on low frequency components of the solution (far interactions). The multi-grid method is then to relax the solution on a particular model until the residual stops changing. This indicates that the solution contains the higher frequency components. A coarse model is then generated for the lower frequency components to the solution. The model is then relaxed for the lower frequency components of the solution. These lower frequency components are then interpolated to the fine model. In computational fluid mechanics the equations are usually expressed as finite differences.

Description: The deployment, assembly and mission-oriented maneuvering of space structures in orbit will trigger large motions of flexible, truss-type structures. In addition, the presence of on-board controls both for attitude stabilization and specified vibration tolerance requirements may further complicate the dynamic behavior of the orbiting structures. Because of safety and cost considerations, the dynamic response of the combined structural and control systems must be predicted reliably. This need can only be met through the development of reliable and efficient simulation capabilities, since there is general agreement that on-orbit experiments should be limited because of cost, time and facility constraints. The long-term objective of this research effort is to develop a next-generation computer simulator for the dynamics and control of large space structures. The simulator will be based on integrating four research thrusts: a new multibody dynamics formulation methodology, modeling capabilities in long/slender truss-beam components with realistic joints, efficient computational procedures that can be implemented either in sequential or concurrent computers, and prototype simulation modules that can be easily processed into a modern large-scale engineering software system such as the NASA/Computational Structural Mechanics (CSM) testbed.

Description: The 3-D inelastic analysis method is a focused program with the objective to develop computationally effective analysis methods and attendant computer codes for three-dimensional, nonlinear time and temperature dependent problems present in the hot section of turbojet engine structures. Development of these methods was a major part of the Hot Section Technology (HOST) program over the past five years at Lewis Research Center.

Description: A summary of the status of this five-year project which is now in its third year of research and development is presented. The goal of the project is the development of several methodologies for probabilistic structural modeling. Probabilistic structural modeling consists of stochastic models of material properties, part geometries, boundary conditions, as well as loading conditions. The current presentation focuses on one methodology - coupling of an advanced finite element structural analysis code with probabilistic modeling strategies. The essential algorithm developments for combining the finite element and probabilistic analysis methods are reported. The validity of the resulting probabilistic structural analysis method is confirmed through a series of test problems with exact results based on Monte Carlo simulations. Additionally, the applicability of the method to a Space Propulsion System (a turbine blade) is demonstrated for static stresses.

Description: This project supports the conversion of codes in Computational Structural Mechanics (CSM) to a parallel form which will efficiently exploit the computational power available from multiprocessors. The work is a part of a comprehensive, FORTRAN-based system to form a basis for a parallel version of the NICE/SPAR combination which will form the CSM Testbed. The software is macro-based and rests on the force methodology developed by the principal investigator in connection with an early scientific multiprocessor. Machine independence is an important characteristic of the system so that retargeting it to the Flex/32, or any other multiprocessor on which NICE/SPAR might be imnplemented, is well supported. The principal investigator has experience in producing parallel software for both full and sparse systems of linear equations using the force macros. Other researchers have used the Force in finite element programs. It has been possible to rapidly develop software which performs at maximum efficiency on a multiprocessor. The inherent machine independence of the system also means that the parallelization will not be limited to a specific multiprocessor.

Description: A new numerical algorithm for the solution of large-order eigenproblems typically encountered in linear elastic finite element systems is presented. The architecture of parallel processing is utilized in the algorithm to achieve increased speed and efficiency of calculations. The algorithm is based on the frontal technique for the solution of linear simultaneous equations and the modified subspace eigenanalysis method for the solution of the eigenproblem. Assembly, elimination and back-substitution of degrees of freedom are performed concurrently, using a number of fronts. All fronts converge to and diverge from a predefined global front during elimination and back-substitution, respectively. In the meantime, reduction of the stiffness and mass matrices required by the modified subspace method can be completed during the convergence/divergence cycle and an estimate of the required eigenpairs obtained. Successive cycles of convergence and divergence are repeated until the desired accuracy of calculations is achieved. The advantages of this new algorithm in parallel computer architecture are discussed.

Description: The overall objective of this research is to develop efficient methods for the solution of linear and nonlinear systems of equations on parallel and supercomputers, and to apply these methods to the solution of problems in structural analysis. Attention has been given so far only to linear equations. The methods considered for the solution of the stiffness equation Kx=f have been Choleski factorization and the conjugate gradient iteration with SSOR and Incomplete Choleski preconditioning. More detail on these methods will be given on subsequent slides. These methods have been used to solve for the static displacements for the mast and panel focus problems in conjunction with the CSM testbed system based on NICE/SPAR.

Description: The background for the CSM Testbed Architecture and a description of the three architecture related tasks under the NASA/Langley CSM contract (NAS1-18444) are given. For each task the objectives, subtasks chosen to achieve the objectives, and the accomplishments are presented. For Task 2, Near-Term Enhancements to the CSM Testbed Architecture, the primary objectives are to quickly correct: (1) inefficiency in the GAL data manager, and (2) deficiency in the CLIP-interpreted command language, CLAMP. The corresponding modifications should preserve upward-compatibility to a reasonable extent, while at the same time increasing the flexibility and extensibility of the CSM Testbed. Researchers have increased the efficiency of GAL by a factor of 2+ and have made several improvements in CLIP. For Task 5, Matrix Algebra Methods and Utilities, the goal is to investigate the current capabilities of the CSM Testbed and the required improvements to the CSM Testbed for performing the matrix algebraic operations required for the analysis of present and future CSM focal problems using advanced methods. Researchers have made an extensive study of the matrix algebra functions in SPAR that includes documenting these routines and supplying inline comments.

Description: The PISCES project started in 1984 under the sponsorship of the NASA Computational Structural Mechanics (CSM) program. A PISCES 1 programming environment and parallel FORTRAN were implemented in 1984 for the DEC VAX (using UNIX processes to simulate parallel processes). This system was used for experimentation with parallel programs for scientific applications and AI (dynamic scene analysis) applications. PISCES 1 was ported to a network of Apollo workstations by N. Fitzgerald.

Description: In January 1987, SPARTA received a Phase I SBIR award from the NASA Lewis Research Center to investigate the feasibility of a finite element solver implemented on multiple VLSI processors. The transputer was chosen as the processor for the feasibility study since it combined low cost with high performance and was specifically designed to directly link with other transputers to form networks of multiple processors. A brief description of transputers, a summary of the SBIR feasibility study, and a discussion of issues concerning a large scale transputer based finite element solver (TBFES) and the performance levels which can be expected are discussed.

Description: Damage assessment of structural assemblies is treated as an identification problem. A brief review of identification methods is first presented with particular focus on the output error approach. The use of numerical optimization methods in identifying the location and extent of damage in structures is studied. The influence of damage on eigenmode shapes and static displacements is explored as a means of formulating a measure of damage in the structure. Preliminary results obtained in this study are presented and special attention is directed at the shortcomings associated with the nonlinear programming approach to solving the optimization problem.

Description: Within the past several years, the Jet Propulsion Laboratory has designed and built major ground antenna structures in Spain, Australia, and California. One of the antennas at each location is a 70 meter-diameter structure that is a retrofit of the existing 64 meter antenna. The 64 meter existing antennas were first stripped back to a 34 meter interior and then completely new construction with deeper trusses was added to extend the interior to 70 meters. The 70 meter project included the rare opportunity to collect field data to compare with predictions of the finite-element analytical models. The new quadripod design was tested for its lower mode natural frequencies and the main reflector was measured by theodolite to determine deflections of subsets of the backup-structure deformations under load. The emphasis here is to examine measurement results and possibly provide some appreciation of the relationship of predictions made from the design model to actual measurements.

Description: The application of a generalized optimality criteria to framed structures is presented. The optimality conditions, Lagrangian multipliers, resizing algorithm, and scaling procedures are all represented as a function of the objective and constraint functions along with their respective gradients. The optimization of two plane frames under multiple loading conditions subject to stress, displacement, generalized stiffness, and side constraints is presented. These results are compared to those found by optimizing the frames using a nonlinear mathematical programming technique.

Description: Recent developments in multilevel optimization are briefly reviewed. The general nature of the multilevel design task, the use of approximations to develop and solve the analysis design task, the structure of the formal multidiscipline optimization problem, a simple cantilevered beam which demonstrates the concepts of multilevel design and the basic mathematical details of the optimization task and the system level are among the topics discussed.

Description: A two-level design optimization metholology is described. A progress report of its application to Printed Wiring Board (PWB) assembly examples is given. The design of PWB assemblies is a complex task which is generally conducted as a sequential process. Individual PWBs are usually designed first, followed by the composition of the PWBs into an assembly. As a result, optimizing design considerations such as assembly reliability cannot be accomplished. This study showed that a two-level decomposition method can be employed to optimize for reliability at both the PWB- and the assembly-level in a coupled manner. The two-level decomposition method also resolved the mixed-integer nonlinear programming nature of the problem rather easily.

Description: Independent Modal Space Control (IMSC) is a technique that is often used for the control of large order structural systems. The pertinent optimization problem in the simultaneous design and control of structures is a min - min problem that minimizes with respect to the structural design variables, the minimum value of the performance index with respect to the control forces obtained using the IMSC technique. The minimization process requires derivatives of eigenvalues and eigenvectors with respect to the design variables. These derivatives can be computed by a rather involved analytical procedure or a relatively simple finite difference procedure. The computer cost effectiveness of these two procedures for the derivative calculations is examined.

Description: The objective of this presentation is to introduce the attendees to the DYSCO program. The emphasis will be on the features which make it multidisciplinary. DYSCO is a very general and versatile software program which couples and solves dynamic systems. It was initiated in the late 1970's in response to a helicopter analysis requirement. The system development, however, resulted in an executive which was completely separated from any particular area of technology, except that of second order ordinary differential equations. During the course of its development, it was funded by the Army Aviation Applied Technology Directorate, the Air Force Wright Aeronautical Laboratories, and by the Kaman Aerospace Corporation. It is completely written in FORTRAN and is operational on IBM and VAX computers.

Description: A numerical method is presented for design sensitivity analysis, using an iterative-method reanalysis of the structure generated by a small perturbation in the design variable; a forward-difference scheme is then employed to obtain the approximate sensitivity. Algorithms are developed for displacement and stress sensitivity, as well as for eignevalues and eigenvector sensitivity, and the iterative schemes are modified so that the coefficient matrices are constant and therefore decomposed only once.

Description: The considerations and the resultant approach used to implement design sensitivity capability for grids into a large scale, general purpose finite element system (MSC/NASTRAN) are presented. The design variables are grid perturbations with a rather general linking capability. Moreover, shape and sizing variables may be linked together. The design is general enough to facilitate geometric modeling techniques for generating design variable linking schemes in an easy and straightforward manner. Test cases have been run and validated by comparison with the overall finite difference method. The linking of a design sensitivity capability for shape variables in MSC/NASTRAN with an optimizer would give designers a powerful, automated tool to carry out practical optimization design of real life, complicated structures.

Description: Structural optimization has been available to the structural analysis community as a tool for many years. The popular use of displacement method finite-element techniques to analyze linearly elastic structures has resulted in an ability to calculate the weight and constraint gradients inexpensively for numerical optimization of structures. Here, recent experiences in the investigation and use of structural optimization are discussed. In particular, experience with the commercially available ADS/NASOPT code is addressed. An overview of the ADS/NASOPT procedure and how it was implemented is given. Two example problems are also discussed.

Description: ASTROS (Automated Structural Optimization System) is a finite-element-based multidisciplinary structural optimization procedure developed under Air Force sponsorship to perform automated preliminary structural design. The design task is the determination of the structural sizes that provide an optimal structure while satisfying numerous constraints from many disciplines. In addition to its automated design features, ASTROS provides a general transient and frequency response capability, as well as a special feature to perform a transient analysis of a vehicle subjected to a nuclear blast. The motivation for the development of a single multidisciplinary design tool is that such a tool can provide improved structural designs in less time than is currently needed. The role of such a tool is even more apparent as modern materials come into widespread use. Balancing conflicting requirements for the structure's strength and stiffness while exploiting the benefits of material anisotropy is perhaps an impossible task without assistance from an automated design tool. Finally, the use of a single tool can bring the design task into better focus among design team members, thereby improving their insight into the overall task.

Description: Large space structures (LSSs) and other dynamical systems of current interest are often extremely complex assemblies of rigid and flexible bodies subjected to kinematical constraints. A formulation is presented for the governing equations of constrained multibody systems via the application of singular value decomposition (SVD). The resulting equations of motion are shown to be of minimum dimension.

Description: A recently developed finite-element capability for general nonlinear shell analysis, featuring the use of three-dimensional constitutive equations within an efficient resultant-oriented framework, is employed to simulate the postbuckling response of an axially compressed composite cylindrical panel with a circular cutout. The problem is a generic example of modern composite aircraft components for which postbuckling strength (i.e., fail-safety) is desired in the presence of local discontinuities such as holes and cracked stiffeners. While the computational software does a reasonable job of predicting both the buckling load and the qualitative aspects of postbuckling (compared both with experiment and another code) there are some discrepancies due to: (1) uncertainties in the nominal layer material properties, (2) structural sensitivity to initial imperfections, and (3) the neglect of dynamic and local material delamination effects in the numerical model. Corresponding refinements are suggested for the realistic continuation of this type of analysis.