Search NASA⌕ Search

SEARCH · Search NASA

Results for “stochastic sampling”

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 955 records · Page 53

Stochastic simulation of predictive space–time scenarios of wind speed using observations and physical model outputs

We propose a statistical space-time model for predicting atmospheric wind speed based on deterministic numerical weather predictions and historical measurements. We consider a Gaussian multivariate space-time framework that combines multiple sources of past physical model outputs and measurements in order to produce a probabilistic wind speed forecast within the prediction window. We illustrate this strategy on wind speed forecasts during several months in 2012 for a region near the Great Lakes in the United States. The results show that the prediction is improved in the mean-squared sense relative to the numerical forecasts as well as in probabilistic scores. Moreover, the samples are shown to produce realistic wind scenarios based on sample spectra and space-time correlation structure.

54 ENVIRONMENTAL SCIENCES↗

A Machine‐Learning‐Assisted Stochastic Cloud Population Model as a Parameterization of Cumulus Convection

Abstract A machine‐learning‐assisted stochastic cloud population model is coupled with the Advanced Research Weather Research and Forecasting (WRF) model to represent fluctuations in the cloud‐base mass flux associated with the life cycles and interactions among cumulus convection cells. In this cloud population model, the size distribution and the associated cloud‐base mass flux of the convective cells are related to their previous state and to the change in the total convective area via a transition function. The convective area tendency in turn is assumed to depend on the cloud‐base mass flux that is resolved by the host WRF model. The transition function is represented by a single hidden‐layer neural network trained by the evolution of convective cell size distributions in a 1‐km grid‐spacing WRF simulation run over the Australian Monsoon region. At every grid point of the host model, the cloud population model predicts the cell size and cloud‐base mass flux distributions from which a random sample of cells is fed to an entraining parcel model that calculates precipitation as well as the associated liquid water potential temperature and total moisture tendencies. These tendencies are averaged over the cells and provided to the host model. Several regional simulations are performed over tropical and midlatitude domains to test this as a potential approach to scale‐aware parameterization. It is shown that such an approach could be a new promising path to simulating realistic precipitation statistics and propagation of precipitation associated with the Madden‐Julian Oscillation while maintaining realistic depictions of the diurnal cycle over both land and ocean.

54 ENVIRONMENTAL SCIENCES↗

Nature and origin of basin-forming projectiles

The formation of the observed lunar multi-ring basins is discussed in the context of current theories of terrestrial planet formation, particularly those in which these planets formed by the accumulation of large planetesimals. The observed number, size, and timing of lunar basin-forming impacts is in the range expected for such theories. Tidal disruption during close encounters to earth and Venus can provide a single mechanism that explains a number of details concerning the number, size distribution, and stochastic nature of the timing of these impacts. A basin time scale is suggested in which Nectaris is associated with the 4.1 b.y. age of the Apollo 16 light matrix breccias. In accordance with the present consensus, Serenitatis is 3.86 b.y., Imbrium and Orientale 3.80-3.82 b.y. in age. Other nearside circular basins (e.g., Humorum and Crisium) are intermediate in age between 3.86 and 4.12 b.y. The large number of 3.8-3.9 b.y. ages is attributed primarily to the magnitude of the Imbrium and Serenitatis impacts, and sampling bias resulting from concentration of collection sites in proximity to these basins.

Wetherill, G. W.↗

Stabilization of the 81-channel coherent beam combination using machine learning

We develop a rapidly converging algorithm for stabilizing a large channel-count diffractive optical coherent beam combination. An 81-beam combiner is controlled by a novel, machine-learning based, iterative method to correct the optical phases, operating on an experimentally calibrated numerical model. A neural-network is trained to detect phase errors based on interference pattern recognition of uncombined beams adjacent to the combined one. Due to the non-uniqueness of solutions in the full space of possible phases, the network is trained within a limited phase perturbation/error range. This also reduces the number of samples needed for training. Simulations have proven that the network can converge in one step for small phase perturbations. When the trained neural-network is applied to a realistic case of 360 degree full range, an iterative scheme exploits random walking at the beginning, with the accuracy of prediction on phase feedback direction, to allow the neural-network to step into the training range for fast convergence. This neural-network-based iterative method of phase detection works tens of times faster than the commonly used stochastic parallel gradient descent approach (SPGD) using a single-detector and random dither when both are tested with random phase perturbations.

Wang, Dan↗

Diffusion-Model-Assisted Supervised Learning of Generative Models for Density Estimation

Here, we present a supervised learning framework of training generative models for density estimation. Generative models, including generative adversarial networks (GANs), normalizing flows, and variational auto-encoders (VAEs), are usually considered as unsupervised learning models, because labeled data are usually unavailable for training. Despite the success of the generative models, there are several issues with the unsupervised training, e.g., requirement of reversible architectures, vanishing gradients, and training instability. To enable supervised learning in generative models, we utilize the score-based diffusion model to generate labeled data. Unlike existing diffusion models that train neural networks to learn the score function, we develop a training-free score estimation method. This approach uses mini-batch-based Monte Carlo estimators to directly approximate the score function at any spatial-temporal location in solving an ordinary differential equation (ODE), corresponding to the reverse-time stochastic differential equation (SDE). This approach can offer both high accuracy and substantial time savings in neural network training. Once the labeled data are generated, we can train a simple, fully connected neural network to learn the generative model in the supervised manner. Compared with existing normalizing flow models, our method does not require the use of reversible neural networks and avoids the computation of the Jacobian matrix. Compared with existing diffusion models, our method does not need to solve the reverse-time SDE to generate new samples. As a result, the sampling efficiency is significantly improved. We demonstrate the performance of our method by applying it to a set of 2D datasets as well as real data from the University of California Irvine (UCI) repository.

97 MATHEMATICS AND COMPUTING↗

High temperature ion channels and pores

The present invention includes an apparatus, system and method for stochastic sensing of an analyte to a protein pore. The protein pore may be an engineer protein pore, such as an ion channel at temperatures above 55.degree. C. and even as high as near 100.degree. C. The analyte may be any reactive analyte, including chemical weapons, environmental toxins and pharmaceuticals. The analyte covalently bonds to the sensor element to produce a detectable electrical current signal. Possible signals include change in electrical current. Detection of the signal allows identification of the analyte and determination of its concentration in a sample solution. Multiple analytes present in the same solution may also be detected.

Kang, Xiaofeng↗

High temperature ion channels and pores

The present invention includes an apparatus, system and method for stochastic sensing of an analyte to a protein pore. The protein pore may be an engineer protein pore, such as an ion channel at temperatures above 55° C. and even as high as near 100° C. The analyte may be any reactive analyte, including chemical weapons, environmental toxins and pharmaceuticals. The analyte covalently bonds to the sensor element to produce a detectable electrical current signal. Possible signals include change in electrical current. Detection of the signal allows identification of the analyte and determination of its concentration in a sample solution. Multiple analytes present in the same solution may also be detected.

Kang, Xiaofeng↗

Towards a detailed soot model for internal combustion engines

In this work, we present a detailed model for the formation of soot in internal combustion engines describing not only bulk quantities such as soot mass, number density, volume fraction, and surface area but also the morphology and chemical composition of soot aggregates. The new model is based on the Stochastic Reactor Model (SRM) engine code, which uses detailed chemistry and takes into account convective heat transfer and turbulent mixing, and the soot formation is accounted for by SWEEP, a population balance solver based on a Monte Carlo method. In order to couple the gas-phase to the particulate phase, a detailed chemical kinetic mechanism describing the combustion of Primary Reference Fuels (PRFs) is extended to include small Polycyclic Aromatic Hydrocarbons (PAHs) such as pyrene, which function as soot precursor species for particle inception in the soot model. Apart from providing averaged quantities as functions of crank angle like soot mass, volume fraction, aggregate diameter, and the number of primary particles per aggregate for example, the integrated model also gives detailed information such as aggregate and primary particle size distribution functions. In addition, specifics about aggregate structure and composition, including C/H ratio and PAH ring count distributions, and images similar to those produced with Transmission Electron Microscopes (TEMs), can be obtained. The new model is applied to simulate an n-heptane fuelled Homogeneous Charge Compression Ignition (HCCI) engine which is operated at an equivalence ratio of 1.93. In-cylinder pressure and heat release predictions show satisfactory agreement with measurements. Furthermore, simulated aggregate size distributions as well as their time evolution are found to qualitatively agree with those obtained experimentally through snatch sampling. It is also observed both in the experiment as well as in the simulation that aggregates in the trapped residual gases play a vital role in the soot formation process. (author)

33 ADVANCED PROPULSION SYSTEMS↗

TU-C-17A-08: Improving IMRT Planning and Reducing Inter-Planner Variability Using the Stochastic Frontier Method: Validation Based On Clinical and Simulated Data

Purpose: Intensity modulated radiation therapy always requires compromises between PTV coverage and organs at risk (OAR) sparing. We previously developed metrics that correlate doses to OAR to specific patients’ morphology using stochastic frontier analysis (SFA). Here, we aim to examine the validity of this approach using a large set of realistically simulated dosimetric and geometric data. Methods: SFA describes a set of treatment plans as an asymmetric distribution with respect to a frontier defining optimal plans. Eighty head and neck IMRT plans were used to establish a metric predicting the mean dose to parotids as a function of simple geometric parameters. A database of 140 parotids was used as a basis distribution to simulate physically plausible data of geometry and dose. Distributions comprising between 20 and 5000 were simulated and the SFA was applied to obtain new frontiers, which were compared to the original frontier. Results: It was possible to simulate distributions consistent with the original dataset. Below 160 organs, the SFA could not always describe distributions as asymmetric: a few cases showed a Gaussian or half-Gaussian distribution. In order to converge to a stable solution, the number of organs in a distribution must ideally be above 100, but in many cases stable parameters could be achieved with as low as 60 samples of organ data. Mean RMS value of the error of new frontiers was significantly reduced when additional organs are used. Conclusion: The number of organs in a distribution showed to have an impact on the effectiveness of the model. It is always possible to obtain a frontier, but if the number of organs in the distribution is small (< 160), it may not represent de lowest dose achievable. These results will be used to determine number of cases necessary to adapt the model to other organs.

60 APPLIED LIFE SCIENCES↗

Randomized Algorithms for Scientific Computing (RASC)

Randomized algorithms have propelled advances in artificial intelligence (AI) and represent a foundational research area in advancing AI for Science. Future advancements in DOE Office of Science priority areas such as climate science, astrophysics, fusion, advanced materials, combustion, and quantum computing all require randomized algorithms for surmounting challenges of complexity, robustness, and scalability. Advances in data collection and numerical simulation have changed the dynamics of scientific research and motivate the need for randomized algorithms. For instance, advances in imaging technologies such as X-ray ptychography, electron microscopy, electron energy loss spectroscopy, or adaptive optics lattice light-sheet microscopy collect hyperspectral imaging and scattering data in terabytes, at breakneck speed enabled by state-of-the-art detectors. The data collection is exceptionally fast compared with its analysis. Likewise, advances in high-performance architectures have made exascale computing a reality and changed the economies of scientific computing in the process. Floating-point operations that create data are essentially free in comparison with data movement. Thus far, most approaches have focused on creating faster hardware. Ironically, this faster hardware has exacerbated the problem by making data still easier to create. Under such an onslaught, scientists often resort to heuristic deterministic sampling schemes (e.g., low-precision arithmetic, sampling every nth element) and sacrifice potentially valuable accuracy. Dramatically better results can be achieved via randomized algorithms, reducing the data size as much as or more than naive deterministic subsampling can achieve, while retaining the high accuracy of computing on the full data set. By randomized algorithms we mean those algorithms that employ some form of randomness in internal algorithmic decisions to accelerate time to solution, increase scalability, or improve reliability. Examples include matrix sketching for solving large-scale least-squares problems (see Figure 1) and stochastic gradient descent for training machine learning models. We are not recommending heuristic methods but rather randomized algorithms that have certificates of correctness and probabilistic guarantees of optimality and near-optimality. Such approaches can be useful beyond acceleration, for example, in understanding how to avoid measure zero worst-case scenarios that plague methods such as QR matrix factorization.

97 MATHEMATICS AND COMPUTING↗

Particle-laden swirling free jets: Measurements and predictions

A theoretical and experimental investigation of single-phase and particle-laden weakly swirling jets was conducted. The jets were injected vertically downward from a 19 mm diameter tube with swirl numbers ranging from 0 to 0.33. The particle-laden jets had a single loading ratio (0.2) with particles having a SMD of 39 microns. Mean and fluctuating properties of both phases were measured using nonintrusive laser based methods while particle mass flux was measured using an isokinetic sampling probe. The continuous phase was analyzed using both a baseline kappa-epsilon turbulence model and an extended version with modifications based on the flux Richardson number to account for effects of streamline curvature. To highlight effects of interphase transport rates and particle/turbulence interactions, effects of the particles were analyzed as follows: (1) locally homogeneous flow (LHF) analysis, where interphase transport rates are assumed to be infinitely fast; (2) deterministic separated flow (DSF) analysis, where finite interphase transport rates are considered but particle/turbulence interactions are ignored; and (3) stochastic separated flow (SSF) analysis, where both effects are considered using random-walk computations.

Bulzan, D. L.↗

Particle-laden swirling free jets - Measurements and predictions

A theoretical and experimental investigation of single-phase and particle-laden weakly swirling jets was conducted. The jets were injected vertically downward from a 19 mm diameter tube with swirl numbers ranging from 0 to 0.33. The particle-laden jets had a single loading ratio (0.2) with particles having an SMD of 39 micrometers. Mean and fluctuating properties of both phases were measured using nonintrusive laser based methods while particle mass flux was measured using an isokinetic sampling probe. The continuous phase was analyzed using both a baseline k-epsilon turbulence model and an extended version with modifications based on the flux Richardson number to account for effects of streamline curvature. To highlight effects of interphase transport rates and particle/turbulence interactions, effects of the particles were analyzed as follows: (1) locally homogeneous flow (LHF) analysis, where interphase transport rates are assumed to be infinitely fast; (2) deterministic separated flow (DSF) analysis, where finite interphase transport rates are considered but particle/turbulence interactions are ignored; and (3) stochastic separated flow (SSF) analysis, where both effects are considered using random-walk computations.

Bulzan, D. L.↗

Multi-level methods and approximating distribution functions

Biochemical reaction networks are often modelled using discrete-state, continuous-time Markov chains. System statistics of these Markov chains usually cannot be calculated analytically and therefore estimates must be generated via simulation techniques. There is a well documented class of simulation techniques known as exact stochastic simulation algorithms, an example of which is Gillespie’s direct method. These algorithms often come with high computational costs, therefore approximate stochastic simulation algorithms such as the tau-leap method are used. However, in order to minimise the bias in the estimates generated using them, a relatively small value of tau is needed, rendering the computational costs comparable to Gillespie’s direct method. The multi-level Monte Carlo method (Anderson and Higham, Multiscale Model. Simul. 10:146–179, 2012) provides a reduction in computational costs whilst minimising or even eliminating the bias in the estimates of system statistics. This is achieved by first crudely approximating required statistics with many sample paths of low accuracy. Then correction terms are added until a required level of accuracy is reached. Recent literature has primarily focussed on implementing the multi-level method efficiently to estimate a single system statistic. However, it is clearly also of interest to be able to approximate entire probability distributions of species counts. We present two novel methods that combine known techniques for distribution reconstruction with the multi-level method. We demonstrate the potential of our methods using a number of examples.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

CONSTRAINING SUB-PARSEC BINARY SUPERMASSIVE BLACK HOLES IN QUASARS WITH MULTI-EPOCH SPECTROSCOPY. I. THE GENERAL QUASAR POPULATION

We perform a systematic search for sub-parsec binary supermassive black holes (BHs) in normal broad-line quasars at z < 0.8, using multi-epoch Sloan Digital Sky Survey (SDSS) spectroscopy of the broad Hβ line. Our working model is that (1) one and only one of the two BHs in the binary is active; (2) the active BH dynamically dominates its own broad-line region (BLR) in the binary system, so that the mean velocity of the BLR reflects the mean velocity of its host BH; (3) the inactive companion BH is orbiting at a distance of a few R{sub BLR}, where R{sub BLR} ∼ 0.01-0.1 pc is the BLR size. We search for the expected line-of-sight acceleration of the broad-line velocity from binary orbital motion by cross-correlating SDSS spectra from two epochs separated by up to several years in the quasar rest frame. Out of ∼700 pairs of spectra for which we have good measurements of the velocity shift between two epochs (1σ error ∼40 km s{sup –1}), we detect 28 systems with significant velocity shifts in broad Hβ, among which 7 are the best candidates for the hypothesized binaries, 4 are most likely due to broad-line variability in single BHs, and the rest are ambiguous. Continued spectroscopic observations of these candidates will easily strengthen or disprove these claims. We use the distribution of the observed accelerations (mostly non-detections) to place constraints on the abundance of such binary systems among the general quasar population. Excess variance in the velocity shift is inferred for observations separated by longer than 0.4 yr (quasar rest frame). Attributing all the excess to binary motion would imply that most of the quasars in this sample must be in binaries, that the inactive BH must be on average more massive than the active one, and that the binary separation is at most a few times the size of the BLR. However, if this excess variance is partly or largely due to long-term broad-line variability, the requirement of a large population of close binaries is much weakened or even disfavored for massive companions. Future time-domain spectroscopic surveys of normal quasars can provide vital prior information on the structure function of stochastic velocity shifts induced by broad-line variability in single BHs. Such surveys with improved spectral quality, increased time baseline, and more epochs can greatly improve the statistical constraints of this method on the general binary population in broad-line quasars, further shrink the allowed binary parameter space, and detect true sub-parsec binaries.

79 ASTRONOMY AND ASTROPHYSICS↗

Wide-field Hubble space telescope observations of the globular cluster system in NGC 1399

We present a comprehensive high spatial resolution imaging study of globular clusters (GCs) in NGC 1399, the central giant elliptical cD galaxy in the Fornax galaxy cluster, conducted with the Advanced Camera for Surveys (ACS) aboard the Hubble Space Telescope (HST). Using a novel technique to construct drizzled point-spread function libraries for HST/ACS data, we accurately determine the fidelity of GC structural parameter measurements from detailed artificial star cluster experiments and show the superior robustness of the GC half-light radius, r{sub h} , compared with other GC structural parameters, such as King core and tidal radius. The measurement of r{sub h} for the major fraction of the NGC 1399 GC system reveals a trend of increasing r{sub h} versus galactocentric distance, R {sub gal}, out to about 10 kpc and a flat relation beyond. This trend is very similar for blue and red GCs, which are found to have a mean size ratio of r {sub h,} {sub red}/r {sub h,} {sub blue} = 0.82 ± 0.11 at all galactocentric radii from the core regions of the galaxy out to ∼40 kpc. This suggests that the size difference between blue and red GCs is due to internal mechanisms related to the evolution of their constituent stellar populations. Modeling the mass density profile of NGC 1399 shows that additional external dynamical mechanisms are required to limit the GC size in the galaxy halo regions to r{sub h} ≈ 2 pc. We suggest that this may be realized by an exotic GC orbit distribution function, an extended dark matter halo, and/or tidal stress induced by the increased stochasticity in the dwarf halo substructure at larger galactocentric distances. We compare our results with the GC r{sub h} distribution functions in various galaxies and find that the fraction of extended GCs with r{sub h} ≥ 5 pc is systematically larger in late-type galaxies compared with GC systems in early-type galaxies. This is likely due to the dynamically more violent evolution of early-type galaxies. We match our GC r{sub h} measurements with radial velocity data from the literature and split the resulting sample at the median r{sub h} value into compact and extended GCs. We find that compact GCs show a significantly smaller line-of-sight velocity dispersion, (σ{sub cmp}) = 225 ± 25 km s{sup –1}, than their extended counterparts, (σ{sub ext}) = 317 ± 21 km s{sup –1}. Considering the weaker statistical correlation in the GC r{sub h} color and the GC r{sub h} -R {sub gal} relations, the more significant GC size-dynamics relation appears to be astrophysically more relevant and hints at the dominant influence of the GC orbit distribution function on the evolution of GC structural parameters.

79 ASTRONOMY AND ASTROPHYSICS↗

The Galah Survey: Classification and Diagnostics with t-SNE Reduction of Spectral Information

Galah is an ongoing high-resolution spectroscopic survey with the goal of disentangling the formation history of the Milky Way using the fossil remnants of disrupted star formation sites that are now dispersed around the Galaxy. It is targeting a randomly selected magnitude-limited ( V ≤ 14) sample of stars, with the goal of observing one million objects. To date, 300,000 spectra have been obtained. Not all of them are correctly processed by parameter estimation pipelines, and we need to know about them. We present a semi-automated classification scheme that identifies different types of peculiar spectral morphologies in an effort to discover and flag potentially problematic spectra and thus help to preserve the integrity of the survey results. To this end, we employ the recently developed dimensionality reduction technique t-SNE ( t -distributed stochastic neighbor embedding), which enables us to represent the complex spectral morphology in a two-dimensional projection map while still preserving the properties of the local neighborhoods of spectra. We find that the majority (178,483) of the 209,533 Galah spectra considered in this study represents normal single stars, whereas 31,050 peculiar and problematic spectra with very diverse spectral features pertaining to 28,579 stars are distributed into 10 classification categories: hot stars, cool metal-poor giants, molecular absorption bands, binary stars, H α /H β emission, H α /H β emission superimposed on absorption, H α /H β P-Cygni, H α /H β inverted P-Cygni, lithium absorption, and problematic. Classified spectra with supplementary information are presented in the catalog, indicating candidates for follow-up observations and population studies of the short-lived phases of stellar evolution.

79 ASTRONOMY AND ASTROPHYSICS↗

SELECTION OF BURST-LIKE TRANSIENTS AND STOCHASTIC VARIABLES USING MULTI-BAND IMAGE DIFFERENCING IN THE PAN-STARRS1 MEDIUM-DEEP SURVEY

We present a novel method for the light-curve characterization of Pan-STARRS1 Medium Deep Survey (PS1 MDS) extragalactic sources into stochastic variables (SVs) and burst-like (BL) transients, using multi-band image-differencing time-series data. We select detections in difference images associated with galaxy hosts using a star/galaxy catalog extracted from the deep PS1 MDS stacked images, and adopt a maximum a posteriori formulation to model their difference-flux time-series in four Pan-STARRS1 photometric bands g {sub P1}, r {sub P1}, i {sub P1}, and z {sub P1}. We use three deterministic light-curve models to fit BL transients; a Gaussian, a Gamma distribution, and an analytic supernova (SN) model, and one stochastic light-curve model, the Ornstein-Uhlenbeck process, in order to fit variability that is characteristic of active galactic nuclei (AGNs). We assess the quality of fit of the models band-wise and source-wise, using their estimated leave-out-one cross-validation likelihoods and corrected Akaike information criteria. We then apply a K-means clustering algorithm on these statistics, to determine the source classification in each band. The final source classification is derived as a combination of the individual filter classifications, resulting in two measures of classification quality, from the averages across the photometric filters of (1) the classifications determined from the closest K-means cluster centers, and (2) the square distances from the clustering centers in the K-means clustering spaces. For a verification set of AGNs and SNe, we show that SV and BL occupy distinct regions in the plane constituted by these measures. We use our clustering method to characterize 4361 extragalactic image difference detected sources, in the first 2.5 yr of the PS1 MDS, into 1529 BL, and 2262 SV, with a purity of 95.00% for AGNs, and 90.97% for SN based on our verification sets. We combine our light-curve classifications with their nuclear or off-nuclear host galaxy offsets, to define a robust photometric sample of 1233 AGNs and 812 SNe. With these two samples, we characterize their variability and host galaxy properties, and identify simple photometric priors that would enable their real-time identification in future wide-field synoptic surveys.

79 ASTRONOMY AND ASTROPHYSICS↗

Spatio-Temporal Data Analysis at Scale Using Models Based on Gaussian Processes

Gaussian processes are the most commonly used statistical model for spatial and spatio-temporal processes that vary continuously. They are broadly applicable in the physical sciences and engineering and are also frequently used to approximate the output of complex computer models, deterministic or stochastic. We undertook research related to theory, computation, and applications of Gaussian processes as well as some work on estimating extremes of distributions for which a Gaussian process assumption might be inappropriate. Our theoretical contributions include the development of new classes of spatial-temporal covariance functions with desirable properties and new results showing that certain covariance models lead to predictions with undesirable properties. To understand how Gaussian process models behave when applied to deterministic computer models, we derived what we believe to be the first significant results on the large sample properties of estimators of parameters of Gaussian processes when the actual process is a simple deterministic function. Finally, we investigated some theoretical issues related to maxima of observations with varying upper bounds and found that, depending on the circumstances, standard large sample results for maxima may or may not hold. Our computational innovations include methods for analyzing large spatial datasets when observations fall on a partially observed grid and methods for estimating parameters of a Gaussian process model from observations taken by a polar-orbiting satellite. In our application of Gaussian process models to deterministic computer experiments, we carried out some matrix computations that would have been infeasible using even extended precision arithmetic by focusing on special cases in which all elements of the matrices under study are rational and using exact arithmetic. The applications we studied include total column ozone as measured from a polar-orbiting satellite, sea surface temperatures over the Pacific Ocean, and annual temperature extremes at a site in New York City. In each of these applications, our theoretical and computational innovations were directly motivated by the challenges posed by analyzing these and similar types of data.

97 MATHEMATICS AND COMPUTING↗