Improve Sampling Efficiency Research Articles

On the efficiency of stereologic volumetry as commonly implemented for three-dimensional digital images

The aim of the study was to demonstrate, for certain ideal shapes (right cylinders) and for representative neuroanatomical images, that stereologic volumetry of three-dimensional images is more efficient when the sampling grid is placed randomly on each cross-section rather than identically across sections. As an example, the special case of mathematical right cylinders is presented, and an informal proof is provided. For neuroanatomical images, a custom computer program estimated volume with either the fixed- or random-grid method, using the same cross-sectional slices and first-slice test grid position for each method. The slice spacing, grid size, and starting grid position were randomly varied within practical constraints for 100 000 trials in each image. For right cylinders, the random-grid method is always more efficient than the fixed-grid method. For the neuroanatomic images tested, relative variance was up to three times higher for the fixed-grid method than for the random-grid method, especially for test grids with few grid intersections (`hits') per section. With the random-grid method, relative variance is primarily dependent on the total number of hits rather than on the distribution of hits per section. Implementation of the random-grid method for stereologic volumetry in three-dimensional images should in general improve sampling efficiency.

Measuring Spider Richness: Effects of Different Sampling Methods and Spatial and Temporal Scales

Assessing the richness of invertebrate taxa to aid conservation and management requires a better understanding of the potential sources of error. Patterns of richness for heathland spiders at the species and family levels were compared across three sampling methods, four spatial scales, and monthly intervals (for 16 months). A total of 33 families and 130 species was collected: pitfall traps collected 94% of species, sweep net, 25%, and visual search, 41%. The sampling methods produced variable results. Pitfall trap and sweep net techniques identified significant, yet contrasting spatial differences in the number of families and species at one spatial scale. Pitfall trap data reflected strong temporal variation that influenced spatial patterns in richness (across one spatial scale for families and two for species). The use of broader temporal scales introduced a potential failure to detect significant differences in the richness of ground active spiders, and this risk varied spatially. The sweep net is not recommended for this habitat, although a method that targets the foliage is required for a more complete faunal assessment. Visual searches detected no significant patterns in richness, yet given its potential and increasing use for rapid biodiversity surveys, ways to improve sampling efficiency are suggested.

Temporal Variation in Activity of Bats and the Design of Echolocation-Monitoring Studies

I used Anabat II bat detectors to monitor echolocation calls of bats over two streams in the Oregon Coast Range for a total of 195 detector-nights. Activity of bats was positively correlated with biomass of insects and minimum nightly temperature, and was negatively correlated with length of night; activity levels at the two streams were positively correlated. Activity of bats was not significantly correlated with either hours of moonlight or with phase of moon. Level of activity within a night generally peaked shortly after sunset with a second, smaller peak in activity shortly before sunrise, but patterns varied substantially among nights. Total nightly activity at a site also varied substantially among nights, sometimes varying several-fold on consecutive nights. To assess the implications of temporal variation in activity of bats on sampling, I randomly sampled subsets of the data using from 2- to 12-night sample periods and calculated mean levels of activity for each subset. For subsets with seven or more nights, >60% of the subsets had means that were within 20% of the mean of the entire dataset. Less than 50% of the subsets had means within 10% of the mean of the entire dataset for any number of nights subsampled. When comparing activity between sites, use of blocked or paired designs improved sampling efficiency by 20%. Failure to account for temporal variation in activity of bats when designing research projects and monitoring programs could result in biased estimates of activity of bats.

Accelerated equilibration of polymer melts by time-coarse-graining

An effective-potential approach is presented for improving sampling efficiency in simulations of atomistically detailed models of dense long-chain liquids. The motion of atoms on short-time scales is described as rapid fluctuations about the slowly moving mean conformations of the chain molecules. The distribution of these fluctuations is approximated by that of isotropic elastic motion. The interactions between nonbonded pairs of atoms are preaveraged over this distribution and a much softer, effective interaction is obtained, allowing a correspondingly faster exploration of configuration space. The approximate sampling scheme is then tested on two model systems of united-atom liquid hydrocarbons—a melt of twenty C24 chains and one of ten C71 chains. In the C24 melt, where a comparison with fully equilibrated samples from rigorous algorithms is possible, the preaveraging is shown to produce an improvement in sampling efficiency of up to an order of magnitude at the expense of only a moderate loss in accuracy. The method is then applied to the C71 melt, where currently available, rigorous sampling algorithms fail to produce thorough equilibration. Also in this case, the preaveraging method proves to significantly enhance the sampling speed while producing satisfactory distributions of observables.

396 Percutaneous transvesical chorionic villus sampling: An alternative approach to the retroverted uterus

In 458 consecutive chorionic villus sampling (CVS) procedures, we observed a significant influence of uterine position upon sampling efficacy. Compared with anteverted (N = 243) or axial (N = 149) locations, the retroverted uterus (N = 66) was associated with a lower mean sample weight per aspiration (22, 18, and 15 mg, respectively; P less than .01) and a greater frequency of multiple-pass procedures (23, 31, and 52%, respectively; P less than .0001). To improve sampling efficiency in selected cases of uterine retroversion, we adopted a transvesical approach. When compared with transabdominal or transcervical techniques, transvesical CVS had the highest single-pass success rate (33, 33, and 60%, respectively). Only one in 30 transvesical cases required three placental passes, compared with nine of 36 retroverted uteri sampled by either transabdominal or transcervical techniques (P less than .05). The mean transvesical sample weight was 18.7 mg; at least 10 mg was retrieved in all cases. Post-procedure bleeding occurred in four instances and an additional patient suffered a spontaneous loss at 16 weeks' gestation. Aneuploidy was found in four of 30 biopsy specimens, and the remaining pregnancies either have delivered at term (N = 18) or are continuing (N = 7). Our preliminary experience suggests that selected use of this CVS method may improve sampling efficiency without increasing the incidence of complications.

Random sampling from hash files

In this paper we discuss simple random sampling from hash files on secondary storage. We consider both iterative and batch sampling algorithms from both static and dynamic hashing methods. The static methods considered are open addressing hash files and hash files with separate overflow chains. The dynamic hashing methods considered are Linear Hash files [Lit80] and Extendible Hash files [FNPS79]. We give the cost of sampling in terms of the cost of successfully searching a hash file and show how to exploit features of the dynamic hashing methods to improve sampling efficiency.

Electrostatic Effects in Asbestos Sampling I: Experimental Measurements

Electrostatic charge can cause errors during sampling of airborne asbestos fibers and other particles. The change in particle trajectories caused by charge effects during sampling can result in nonuniform deposits on the collecting filter surface and net loss of sample. The degree of these electrostatic effects depends on particle charge, sampler charge, sampler conductivity, and sampling flow rate and direction. The purpose of this research was to evaluate the dependence of sampling efficiency and sample uniformity on these variables. Humidity has been postulated as a primary determinant of particle charge during aerosol generation. Measurements of particle charge and concentration were made as a function of relative humidity with chrysotile fibers generated from a fluidized bed. A strong increase in charge and a decrease in concentration of fibers was noted as the relative humidity was decreased below 15%. The effects of conductive versus nonconductive samplers and sampling flow rate were measured as a function of particle and sampler charge levels. Nonconductive samplers can carry a large and variable charge distribution on their surfaces. This can result in a biased and highly variable particle deposit on the filter when sampling charged particles. Conductive cowls spread any acquired charge over the entire surface and produce a more symmetrical and less biased charged particle deposit. Increasing the sampling flow rate will improve sampling efficiency and decrease deposit variability because the charged particle has less time to interact with the field produced by the sampler. These results suggest that sampling problems caused by electrostatic charge interactions are most likely to occur under low humidity conditions of dust generation, that sampling should be done at as high a flow rate as possible to reduce these effects, and that analysts should select fields toward the center of the filter to minimize bias and variability.

New approaches to extensive forest inventory in Western Australia using large-scale aerial photography

Summary With the publication of its 1987 Timber Strategy, the Western Australian Government called for a new strategic-level inventory over 1.3-million hectares of the state's jarrah forests by 1991. The design for this inventory uses two phase or double sampling; the first phase involving intensive systematic sampling by large scale aerial photographs (1:1200) to obtain gross bole volume estimates for the forest and its various subunits, the second phase involving resampling a proportion of the first phase plots on the ground to obtain detailed information for adjusting the first phase photo-estimates and for partitioning gross bole volumes into product classes of various types. Large-scale photographs improve sampling efficiency partly because of their use to obtain precise tree measurements independent of ground control. One method available for such photography uses sequential exposures with a single camera and requires an accurate altimeter and tip/tilt recorder to determine scale and camera-orientat...

Use of On‐Site High Performance Liquid Chromatography to Evaluate the Magnitude and Extent of Organic Contaminants in Aquifers

AbstractScientists engaged in the appraisal of ground water contaminated by organic substances are faced with the problems of difficult sample collection and timely chemical analysis. High‐performance liquid chromatography was evaluated for on‐site determination of specific organic contaminants in ground water samples and was used at three study sites. Organic solutes were determined directly in water samples, with little or no preparation, and usually in less than an hour after collection. This information improved sampling efficiency and was useful in screening for subsequent laboratory analysis. On two occasions, on‐site analysis revealed that samples were undergoing rapid change, with major solutes being upgraded and alteration products being formed. In addition to sample stability, this technique proved valuable for monitoring other sampling factors such as compositional changes with respect to pumping, filtration, and cross contamination.

Delineating toxic areas by canine olfaction

A research project was undertaken to learn how the highly acute olfactory sensitivity of the canine could be applied with advantage to environmental problems. The objectives were to determine how dogs could be trained to detect hazardous and toxic pollutants in the environment and how the dogs' responses could be used by environmental workers to improve sampling efficiency and to help delineate contaminated sites that might be encountered in spills and improper disposal incidents. Three dogs were trained to recognize and locate chemicals selected from the toxic and hazardous chemical lists. One of these dogs was trained to respond upon detection of chemical scents at extremely low airborne concentrations such as those that might exist at the outer perimeter of a disposal site. Throughout the project. the safety and health of both dog and handler were carefully considered. Over a relatively short period, two dogs wore successfully trained to recognize toluene, and 2,4,5- and 2,4,6-trichlorophenol at levels that could not be detected as quickly or efficiently using conventional field instrumentation. These dogs were trained to seek out and retrieve chemically contaminated articles or to dig at the site of a simulated ground contamination. In a field experiment, both dogs successfully demonstrated their ability by locating as little as 0.2 g of chemical from distances as great as 50 ft. A third dog was acclimated to another chemical. 1,2.3-trichloropropane. to prepare for a field test at a nearby Superfund site contaminated with this material. This dog was trained to sit immediately when he detected the specified odor. This technique allows a dog to delineate the perimeter of a contaminated area without entering the dangerous zone defined by conventional instrumentation. A field experiment at the Superfund site was carried out under extremely adverse weather conditions; it provided encouraging but inconclusive results.

Wall sampling in slurry systems

AbstractAn experimental study of the errors arising in side wall sampling has been conducted. A two‐dimensional model provides qualitative explanation of these errors but, unlike probe sampling, quantitative prediction of sampling efficiency is not possible because the flow to the sampler is more complex. The experiments show that serious errors in concentration or particle size distribution can arise unless the particles are fine, the concentration is high and a very high sampling velocity is used. A projection extending from the pipe wall was found to improve sampling efficiency. However this improvement was of marginal value because the complex nature of the flow precluded a simple theoretical prediction.

A comparison of soil sampling procedures used to monitor soil fertility in permanent pastures

Soil testing programs are often brought in disrepute by unexplained variability in the data. The deposition of dung and urine onto grazed pasture brings about marked variation in the chemical status of soils which contributes to this variability. A study was undertaken to compare a range of sampling procedures to estimate Colwell-P, Bray-1 P, bicarbonate K and pH levels in adjacent low and high P status paddocks. The sampling strategies used consisted of 75 by 50 m grids; whole and stratified paddock zig-zag and cluster (monitor plot) samplings. Soil test means for the various parameters did not vary among sampling methods. The number of grid samples required to estimate within 10% of the mean varied from 121 for Bray-1 P down to 1 for soil pH. Sampling efficiencies were higher for cluster sampling than for whole paddock zig-zag path sampling. Stratification generally did not improve sampling efficiency in these paddocks. Soil test means declined as sampling depth increased, but the coefficient of variation remained constant for Colwell-P and pH. The results indicate that cluster sampling (monitor plots) is the most appropriate procedure for estimating the nutrient status of grazed pastures. This sampling method enables a more accurate measure to be taken of the nutrient status of a paddock and should allow more reasonable estimates to be made of the temporal variations in soil test.

A simple new way to help speed up Monte Carlo convergence rates: Energy-scaled displacement Monte Carlo

A method we call energy-scaled displacement Monte Carlo (ESDMC) whose purpose is to improve sampling efficiency and thereby speed up convergence rates in Monte Carlo calculations is presented. The method involves scaling the maximum displacement a particle may make on a trial move to the particle’s configurational energy. The scaling is such that on the average, the most stable particles make the smallest moves and the most energetic particles the largest moves. The method is compared to Metropolis Monte Carlo (MMC) and Force Bias Monte Carlo of (FBMC) by applying all three methods to a dense Lennard-Jones fluid at two temperatures, and to hot ST2 water. The functions monitored as the Markov chains developed were, for the Lennard-Jones case: melting, radial distribution functions, internal energies, and heat capacities. For hot ST2 water, we monitored energies and heat capacities. The results suggest that ESDMC samples configuration space more efficiently than either MMC or FBMC in these systems for the biasing parameters used here. The benefit from using ESDMC seemed greatest for the Lennard-Jones systems.

Staining and relative determination of elemental contents by the laser microprobe.

An improved laser microprobe procedure is developed and applied to the measurement of calcium content in microareas of right and left subepicardial muscles. The staining of canine cardiac muscles by Methylene Blue solution (1% w/v) was found to improve sampling efficiency. Elemental content is proportional to T-1/gamma, where T is the transmittance of the characteristic emission line of the element of the photographic plate and gamma is its contrast. In the present system, the calcium content is analyzed using T-3.2. We find that the staining of samples and the determination of T-1/gamma are useful procedures in the application of laser microprobe to the study of elemental content in biologic microareas.

AN INVESTIGATION INTO THE RELATIVE EFFICIENCY OF DIFFERENT SAMPLING PROCEDURES

Several procedures for the sampling of powders before particle-size analysis have been described, but little information is available as to the merits of the various techniques. In the present work the relative efficiencies of scoop sampling, cone and quartering, the chute riffler, and the spinning riffler have been investigated. Of these, scoop sampling is especially prone to error, since the whole of the powder does not pass through the sampling device and the previous history of the powder is important. The special problems of scoop sampling have been studied in some detail; it is concluded that training of the operator could improve sampling efficiency, but that, in general, this sampling technique should be avoided. The efficiency of the spinning-riffler technique proved far superior to any other, provided that the time for one revolution of the ring of containers was only a small fraction of the total time required for the powder to flow through the sampling device.