Contaminants in sediments: Using robust regression for grain-size normalization

Sediment contaminant concentrations usually show an inverse correlation with grain size. This can cause difficulties in distinguishing real differences in contamination from artifacts caused by variations in sediment texture. To overcome this, regression analysis is frequently used to remove the dependency of concentrations on grain size. However, least squares regression lines can be affected markedly by the presence of a small number of unusual samples in the dataset. These outliers may represent samples which are more severely contaminated or which were derived from areas with different underlying geology. They can be removed semi-manually, but robust regression methods such as least absolute values provide a convenient and objective alternative. The methods are illustrated using an example dataset of metal contaminants in sediments from the Humber Estuary, United Kingdom. Least squares regression on the complete dataset yields a rather poor grain size normalization for several elements. By contrast, least absolute values regression produces results very similar to those obtained by least squares regression after careful manual removal of outliers, but it avoids the need for subjective judgments of which data points to omit from the analysis. The intercepts of several of the fitted regression lines were non-zero, indicating that regression-based normalization is preferable to methods based on ratios.     

Erosion of the eastern United States observed with 10Be

The problem of identifying areas of accelerated erosion in a dynamic landscape is complicated. The limited history of sediment yield measurements makes this task difficult even if geomorphic evidence is available. Beryllium-10, a cosmogenic isotope produced by cosmic rays interacting with the earth's atmosphere and surface, has chemical and physical properties that make it useful as a tracer for erosion and sediment transport processes. The rarity of the stable isotope, ⁹Be, allows ¹⁰Be to be detected with accelerator mass spectrometry in natural materials at extremely low levels. Backgrounds for rocks and sediments below 10⁵ atom per g are now attainable, a value to be compared with an average deposition rate of 1.3 × 10⁶ atom cm^?2 yr^?1. The affinity of Be for the components of soil and sediment is sufficiently high that it is effectively immobilized on contact, thereby allowing ¹⁰Be to function as a tracer of sediment transport. To a good approximation all the ¹⁰Be transport out of a drainage basin is on the sediment leaving it. The number of ¹⁰Be atoms passing the gauging station can be determined by measuring the concentration of the isotope in the sediment, if the annual sediment load is known. The ratio of the ¹⁰Be carried from the basin by the sediment to that incident upon it, called the erosion index, has been determined for 48 drainage basins within the same physiographic province, which allows them to be reasonably compared, all of which have sediment yield data. Basins located in the Atlantic coastal plain have an average index of 0.3 with the maximum observed being 0.9. Basins located between the fall line and the mountains, a region called the Piedmont, have an average value of 2.2 with individual values ranging from 0.6 to 6.7; this marked difference is thought to result from two centuries of farming on land of moderate gradient. Basins in the highland regions reflect local conditions with low indices for those in grass and timber and high indices associated with destructive land use. The data allow an estimate of the erosion index for the pre-colonial Piedmont, which then allows the pre-colonial sediment yield to be calculated. A number of basins have also been examined world wide with similar conclusions derived. An important deviation from the rule is noted for rivers that erode large regions of loess, such as the Mississippi, Hwang Ho, and Yangtze. Large aeolian deposits were laid down during the ice age in these basins, deposits that brought inherited ¹⁰Be with them and that are easily eroded.     

Display of acoustic reflection profiles using a dot matrix plotter

Acoustic reflection profiling data display is traditionally done with the aid of a facsimile type of recorder. It is not uncommon to record the unprocessed acoustic data on a tape recorder for subsequent playback through a laboratory computer. This still involves the use of some sort of facsimile recorder for the ultimate display of profiles. This paper presents the results of a study to adapt a high-speed digital dot matrix plotter for the ultimate display in place of the conventional facsimile recorder. Because a minicomputer drives the display directly, a host of signal conditioning procedures are permitted, with the final display being generated in real time. Algorithms are developed to control the marking density, allow adaptive threshold control, bottom tracking, automatic gain control, and de-emphasis of water column boundary reverberation. These techniques are just a few of the many that can be employed since the computer can readily be carried on a large ship in deep water, or a small vessel in a harbour. Shallow water is the difficult case for high energy acoustic sources because the water column boundaries behave much like an excited acoustic cavity. For this reason, a section of seismic profile is shown which was obtained with a 7·5 kHz pinger in only 8 m of water in Narragansett Bay. This research was partiallysupported by the Division of Computer Research of the National Science Foundation.     

Unification Of Non-Dimensional Solutions To Asymptotic Equations For Plumes Of Different Shape

A model of buoyant fluid rise is developed in aLagrangian framework. Results from the model arevalidated by comparison with laboratory and fieldexperiments. The model is sufficiently general toinclude geometries of any cross-sectional shape. Exact solutions to the asymptotic equations forcontinuous discharges from line and point sources andfor an instantaneous discharge from a point source areconsidered. Prismatic, cylindrical andspherical shapes approximate these three geometries,respectively. Accommodation of these shapes withinthe same general model allows for direct comparison. It is shown that, for discharges into a cross-flowthat may be stratified or unstratified, thenon-dimensional trajectory, thickness and dilution canbe uniquely specified using three parameters. Theseare the non-dimensional size of the source, therelative importance of the initial fluxes of momentumand buoyancy and the number of orthogonal axes throughwhich entrainment can occur. Such non-dimensionalresults are particularly useful for examining thosescenarios for which there are limited experimentaldata.     

Stable nickel isotopes and cosmogenic beryllium-10 and aluminum-26 in metallic spheroids from Meteor Crater,Arizona

Abstract— The isotopic abundances of Ni in 17 metallic spheroids from Meteor Crater, Arizona, were determined by inductively coupled plasma mass spectrometry (ICP-MS). Sixteen spheroids have normal isotopic abundances. A 17th shows a marginally detectable mass fractionation of 0.40 ± 0.14 %/AMU in favor of the heavier isotopes. The general absence of mass fractionation indicates that open system evaporation caused little loss of Ni. Variable activities of the cosmogenic radionuclides ¹⁰Be and ²⁶Al were measured by accelerator mass spectrometry in separate suites of spheroids. Activities of ²⁶Al in most samples and of ¹⁰Be in metal cores separated from spheroids indicate that they either (1) come from greater depths in the parent meteoroid than do hand specimens, or (2) lost Al and Be during the process of spheroid formation. One individual spheroid has ¹⁰Be and ²⁶Al activities comparable to those of bulk specimens. This result suggests that spheroid formation may occasionally include material from the outermost meter or so of the impactor. Relatively high activities of ¹⁰Be, ～3 dpm/kg, in the siliceous shells of Canyon Diablo spheroids very likely have a meteoric origin.     

Exposure history of glass and breccia phases of lunar meteorite EET87521

Abstract— Glass-rich separates were prepared from a sample of the basaltic lunar meteorite EET87521 rich in dark glass. Noble gas isotopic abundances and ²⁶Al and ¹⁰Be activities were measured to find out whether shock effects associated with lunar launch helped to assemble these phases. Similar ¹⁰Be and ²⁶Al activities indicate that all materials in EET87521 had a common exposure history in the last few million years before launch. However, the glass contains much higher concentrations of trapped gases and records a much longer cosmic-ray exposure, 100 Ma–150 Ma, in the lunar regolith than does the bulk sample. The different histories show that the glass existed long before the ejection of EET87521. The trapped ⁴⁰Ar/³⁶Ar ratio of 1.6 ± 0.1 implies that the lunar exposure that produced most of the stable cosmogenic noble gases began 500 Ma ago. Cosmogenic and trapped noble gas components correlate strongly in various temperature-release fractions and phases of EET87521, which is probably because the glass contains most of the gas. The trapped solar ratios, ²⁰Ne/²²Ne = 12.68 ± 0.20 and ³⁶Ar/³⁸Ar = 5.24 ± 0.05 can be understood as resulting from a mixture consisting of ～60% solar wind and 40% solar energetic particles (SEP). All EET87521 phases show a ⁴⁰K-⁴⁰Ar gas retention age of ～3300 Ma, which is in the range of typical lunar mare basalts.