Compile-time partitioning of iterative parallel loops to reducecache coherency traffic

Adaptive data partitioning (ADP) which reduces the execution time of parallel programs by reducing interprocessor communication for iterative parallel loops is discussed. It is shown that ADP can be integrated into a communication-reducing back end for existing parallelizing compilers or as part of a machine-specific partitioner for parallel programs. A multiprocessor model to analyze program execution factors that lead to interprocessor communication and a model for the iterative parallel loop to quantify communication patterns within a program are defined. A vector notation is chosen to quantify communication across a global data set. Communication parameters are computed by examining the indexes of array accesses and are adjusted to reflect the underlying system architecture by compensating for cache line sizes. These values are used to generate rectangular and hexagonal partitions that reduce interprocessor communication     

Nitrate and chloride concentrations in the high plains aquifer,Texas

Groundwater nitrate and chloride concentrations were compiled for 122 wells in a rural, three‐county area of northwest Texas. The counties are located on the High Plains aquifer, a major source of groundwater in the region. Cropland/pasture is the predominant land use in the study area. The area also contains numerous cattle feedlots. Fertilizer and manure associated with those land uses are potential sources of ground‐water contamination. Although locally elevated above background levels, none of the chemical concentrations exceeded the primary drinking water standard of 44.27 mg/L for nitrate (10mg/L for NO₃ — N) or secondary standard of 250mg/L for chloride. Rank correlations between nitrate and chloride were statistically significant in two of the three counties, where the solutes may have originated from a common surface source. Denitrification and scant precipitation recharge may account for an absence of nitrate levels above the drinking water standard.     

Arsenic,Nitrate, Chloride And Bromide Contamination In The Gulf Coast Aquifer,South-Central Texas,USA

Arsenic, nitrate, chloride, and bromide concentrations in the Gulf Coast Aquifer of south-central Texas, USA, were compiled, mapped, and evaluated in the context of local land use and geology. Agriculture and oil production are predominant land uses and potential sources of groundwater contamination in the study area. Data were compiled from 69 wells with a median depth of 160.5 v m. Eight observations surpassed the 44.3 v mg/L standard for nitrate (10 v mg/L NO 3 -N), and 24 observations exceeded the 10 v µg/L standard for arsenic. There was a statistically significant, inverse correlation between nitrate and well depth, and a direct correlation between nitrate and arsenic. Arsenic concentrations were significantly higher in a uranium-bearing sand formation compared to other formations in the study area. Chloride concentrations were also high relative to the (secondary) drinking water standard (250 v mg/L), with a median of 342 v mg/L and maximum of 6840 v mg/L. Most chloride/bromide ratios were near 300, but there were four significantly lower values, consistent with oilfield brine or evaporite dissolution. Results of this study suggest that (1) geology exerts a major control on arsenic concentrations in groundwater, (2) agricultural activity contributes substantially to nitrate and chloride and, to a lesser extent, arsenic concentrations in groundwater, and (3) oilfield brine has locally impacted groundwater in the study area.     

On evaluating Weibull fits to mechanical testing data

The use of coefficient of determination, R², and the Anderson–Darling (A²) hypothesis test to evaluate the goodness-of-fit to the two-parameter Weibull distribution was investigated. Results of Monte Carlo simulations for sample sizes between 5 and 100 indicated that guidelines provided previously in the literature are too conservative for sample sizes up to 80. New guidelines for the use of R² and A² for sample sizes between 5 and 100 have been developed. The two measures of goodness-of-fit were found to agree more than 95% of the time, regardless of sample size. The use of the new guidelines has been demonstrated on two datasets from the casting literature.     

Microorganism selection and biosurfactant production in a continuously and periodically operated bioslurry reactor

A continuous-flow reactor (CSTR) and a soil slurry-sequencing batch reactor (SS-SBR) were maintained in 8l vessels for 180 days to treat a soil contaminated with diesel fuel (DF). Concentrations of Candida tropicalis, Brevibacterium casei, Flavobacterium aquatile, Pseudomonas aeruginosa, and Pseudomonas fluorescens were determined using fatty acid methyl ester (FAME) analysis. DF removal (biological and volatile) and biosurfactant concentrations were measured. The SS-SBR encouraged the growth of biosurfactant-producing species relative to the CSTR. Counts of biosurfactant-producing species (C. tropicalis, P. aeruginosa, P. fluorescens) relative to total microbial counts were 88% in the SS-SBR and 23% in the CSTR. Biosurfactants were produced in the SS-SBR to levels of nearly 70 times the critical micelle concentration (CMC) early in the cycle, but were completely degraded by the end of each cycle. No biosurfactant production was observed in the CSTR. DF biodegradation rates were over 40% greater and DF stripping was over five times lower in the SS-SBR than the CSTR. However, considerable foaming occurred in the SS-SBR. Reversing the mode of operation in the reactors on day 80 caused a complete reversal in microbial consortia and reactor performance by day 120. These results show that bioslurry reactor operation can be manipulated to control overall reactor performance.     

Evaluation of linear spectral unmixing and ΔNBR for predicting post‐fire recovery in a North American ponderosa pine forest

The Differenced Normalized Burn Ratio (ΔNBR) is widely used to map post‐fire effects in North America from multispectral satellite imagery, but has not been rigorously validated across the great diversity in vegetation types. The importance of these maps to fire rehabilitation crews highlights the need for continued assessment of alternative remote sensing approaches. To meet this need, this study presents a first preliminary comparison of immediate post‐fire char (black ash) fraction, as measured by linear spectral unmixing, and ΔNBR, with two quantitative one‐year post‐fire field measures indicative of canopy and sub‐canopy conditions: % live tree and dry organic litter weight (gm^?2). Image analysis was applied to Landsat 7 Enhanced Thematic Mapper (ETM+) imagery acquired both before and immediately following the 2000 Jasper Fire, South Dakota. Post‐fire field analysis was conducted one‐year post‐fire. Although the immediate post‐fire char fraction (r ² = 0.56, SE = 28.03) and ΔNBR (r ² = 0.55, SE = 29.69) measures produced similarly good predictions of the % live tree, the standard error in the prediction of litter weight with the char fraction method (r ² = 0.55, SE = 4.78) was considerably lower than with ΔNBR (r ² = 0.52, SE = 8.01). Although further research is clearly warranted to evaluate more field measures, in more fires, and across more fire regimes, the char fraction may be a viable approach to predict longer‐term indicators of ecosystem recovery and may potentially act as a surrogate retrospective measure of the fire intensity.     

Viscoplastic-dynamic crack propagation: Experimental and analysis research for crack arrest applications in engineering structures

The many practical situations in which the initiation of rapid crack propagation cannot be absolutely precluded and where the consequences of a large scale fracture would be catastrophic mandate the development and use of crack arrest fracture mechanics technology. Most currently applied procedures utilize linear elastic analyses and corresponding material fracture property characterizations. Even though the theoretical basis for the use of this level of approach is incomplete, selected practical applications are described in this paper that show that these can still be effectively made. In addition, to address the current deficiencies and to treat those conditions where small-scale yielding conditions are clearly not satisfied, more advanced viscoplastic-dynamic analyses are also being developed. An approach is described in this paper that performs well-instrumented fracture propagation experiments in concert with viscoplastic dynamic finite-element simulations. This approach, currently used to quantify critical values of the T ^* crack tip characterizing parameter developed by Atluri, is being pursued in order to provide transferable material crack arrest toughness values in the regime where viscoplastic-dynamic conditions dominate.

---

Résumé Les nombreuses situations pratiques où l'amorçage d'une croissance rapide d'une fissure ne peut être formellement prévue, et où les conséquences d'une rupture à grande échelle seraient catastrophiques, requièrent le développement et l'usage de la technique de mécanique de la rupture du point de vue de l'arrêt de fissuration.La plupart des procédures actuelles recourent à des analyses linéaires et élastique et aux caractérisations correspondantes des propriétés de résistance à la rupture des matériaux. Même si la base théorique pour une approche de ce niveau est encore incomplète, des applications pratiques spécifiques peuvent en être faites, ainsi qu'on le décrit dans l'étude. En outre, pour répondre aux déficiences actuelles, et pour traiter les conditions qui ne peuvent s'accomoder des conditions de plastification à petite échelle, on développe une analyse dynamique viscoplastique plus évoluée. Dans l'étude, on décrit une approche qui associe des essais de propagation de fissure dûment instrumentés et des simulations dynamiques viscoplastiques par éléments finis. Cette approche, qui est déjà utilisée pour établir les valeurs critiques du paramètres T ^* développé par Atluri pour caractériser l'extrémité de la fissure, est approfondie en vue de fournir des valeurs transférables de la ténacité du matériau vis-à-vis de l'arrêt de propagation d'une fissure, là où prédominent des conditions de viscoplasticité dynamique.

     

Sodium adsorption ratios and salinity levels in nine texas aquifers: implications for irrigated agriculture

This study evaluated sodium and salinity irrigation hazards for Texas’ nine major aquifers. Sodium adsorption ratios (SAR's) and electrical conductivity levels (EC's) were calculated at 5,442 water wells. Aquifer median SAR's varied widely, from 0.3 (Edwards Balcones Fault Zone) to 8.5 (Carrizo‐Wilcox). The percentage of SAR observations exceeding 17.5 ranged from 0.0 (Seymour Aquifer) to 36.3 (Carrizo‐Wilcox Aquifer). There was a significant positive correlation between SAR and well depth in four aquifers and a significant negative correlation in four aquifers. Median EC's ranged from 545 umhos/cm (Edwards Balcones Fault Zone Aquifer) to 2,280 umhos/cm (Cenozoic Pecos Alluvium Aquifer). Respectively, 12.1% and 95.1% of the observations exceeded 750 umhos/cm in the Edwards Balcones Fault Zone and Cenozoic Pecos Alluvium Aquifers. EC and well depth were significantly positively correlated in two aquifers and significantly negatively correlated in five aquifers. Several factors control SAR's and EC's in Texas aquifers, including rock/sediment composition, groundwater chemical evolution, and seepage from nearby formations. Probable human controls include irrigation return flow, oil field brine, and pumping‐induced saltwater intrusion. The Trinity, Carrizo‐Wilcox, Gulf Coast, Cenozoic Pecos Alluvium, Seymour, and Hueco‐Mesilla Bolson Aquifers have the greatest potential for sodium or salinity irrigation problems.     

Consideration of non-steady-state crack growth in materials evaluation and design

Fracture mechanics is being used increasingly in the design and reliability assurance of reactor pressure vessels. This technology enables a quantitative assessment of the potential influence of environment-enhanced crack growth on vessel integrity to be made. The basis of such an assessment is a one-to-one relationship between the rate of crack growth and K, the applied value of the crack-tip stress intensity factor. The purpose of the work described in this paper is to examine the uniqueness of this relationship and also to elucidate and assess the significance of any deviations from such a relationship.Data on a variety of low alloy steels, exposed to aqueous environments, are presented which demonstrate that, under certain conditions, the K versus growth rate relationship appears not to be unique—that is, for a given value ofK, several distinctly different rates can occur. Tests conducted under both static and cyclic loading are shown to exhibit this behaviour following test start-up, test interruptions, or changes in loading variables such as cyclic frequency. In order to understand this phenomenon, experiments were conducted on a quenched and tempered low alloy steel exposed to distilled water. Using specimens in which K remains constant, as the crack grows under constant load, data have been obtained which demonstrate that this behaviour is due, at least in part, to non-steady state crack growth. Initially, rates are significantly less than the steady-state rate and, although crack growth eventually accelerates to the steady-state rate, significant transient periods occur. A unique, geometry independent relationship only exists between crack growth rates and the crack-tip stress intensity factor provided these non steady-state effects are eliminated.For a given material, the duration of this transient period appears to be dependent on environment chemistry, thereby suggesting that the phenomenon is controlled by the kinetics of underlying chemical processes which lead to the material's degradation. A kinetic model based on the concept of a hydrogen embrittled volume of material at the crack tip provides a unified understanding of the various transient behaviours in the steels considered in this paper.The practical significance of this time-dependent behaviour is that it can cause laboratory data to misrepresent the growth rates which occur under operating conditions in reactor pressure vessels. Recommendations are provided for identifying these non-steady-state effects when acquiring laboratory data. Approaches to dealing with transient crack growth rates in design and reliability assessments are also outlined.