Search

Silty lacustrine and paludal records spanning from the penultimate deglaciation (late Illinois Episode) to the present are preserved in kettles formed during marine isotope stage (MIS) 6 that lie adjacent to the MIS 2 Manito terrace along the valleys of the Sangamon and Illinois Rivers. Geochemical, mineralogical, particle-size, and chronological records from two basins, Smith Lake and Lake Ben, provide a unique opportunity to assess loess age and provenance of the southcentral Laurentide ice sheet. Here, I focused on deposits of redeposited loess of the last glaciation (Wisconsin Episode). Other than subtle laminations and fossils of emergent and aquatic plants and animals, the silty texture, chemical composition, and mineralogy of the lake sediment closely resemble Peoria Silt (loess). Sediment-core records sampled from Smith Lake and Lake Ben in Mason County, Illinois, confirm rapid deposition of proximal loess derived from the nearby outwash plain of the Illinois River from ca. 30.9–18.0 calibrated (cal) k.y. B.P. From 18.0–14.7 cal k.y. B.P., deposition of proximal loess was gradually replaced by far-traveled (distal) loess chiefly derived from Superior lobe provenance and global dust. The succession is capped by a cumulic gleysol that was truncated by recent plowing. Proximal loess deposits in lakes Smith and Ben are composed primarily of very fine sand, coarse silt, and medium silt (125– 16 µm) compared to the distal facies (fine silt and finer sediment: <16 µm). Proximal loess that originated from outwash of the Lake Michigan lobe contains abundant Ca, Zr, Sr, illite, and magnetic minerals. The distal facies contains relatively abundant Rb, Al, K, Si, Cu, and Pb, and it is enriched in smectite (relative to illite) and depleted in magnetic minerals. Radiocarbon ages on terrestrial gastropod aragonite in loess from the nearby New Cottonwood School and Thomas Quarry sites indicate that Peoria Silt deposition ended between 18 cal k.y. B.P. and 16 cal k.y. B.P. At these localities, the upper part of the record is complicated by the modern soil, which has altered soil organic matter and leached carbonate. At Lake Ben, the radiocarbon record of emergent aquatic vegetation indicates that sediment accumulation rates and the ratio of proximal to distal loess decreased at the correlative interval that yielded the youngest snail shell ages at Cottonwood School. The upward increase in element concentrations such as Al and decrease in coarse/fine silt ratios in the modern soil are prominent features in the Lake Ben record dating from 18–14 cal k.y. B.P. This correlation strongly suggests that many of the features of the modern soil were not developed into uniform parent material but, rather, were inherited from incremental addition of distal loess to proximal loess, collectively identified as Peoria Silt. The paucity of key elements (Cd, Hg, Se, Ag, Au) in this transition zone suggests limited contributions from western lobes (i.e., Des Moines and James lobes), but abundant Ti and Cu suggest a contribution from the Superior lobe. A new wrinkle in this interpretation is that the rise in elements such as Al, Rb, and Si suggests significant additions from generic dust from unknown sources. [ABSTRACT FROM AUTHOR]

Based on calibrated radiocarbon ages of terrestrial gastropod shells (Succineidae, Discus , Stenotrema , Webbhelix), the chronology of Peoria Silt (loess) deposition in the Central Lowlands is updated. These taxa provide reliable ages (within ~0.2 ka), based on historical shell dating, shell-organic age comparisons, and stratigraphic consistency. A compilation of 53 new and 36 published Peoria Silt shell ages (calibrated), from 12 localities, date from 30.0 to 17.4 ka. Proximal (fossiliferous) loess from 10 sections had mean loess accumulation rates of 0.6–2.2 mm/yr. Study sites along the upper Mississippi, Illinois, to mid-Mississippi, and Ohio-Wabash Valleys suggest Peoria loess accumulated from ~27 to 15 ka, ~29 to 18 ka, and ~30 to 18 ka, respectively. The cessation age for Peoria Silt, based on surface extrapolations, is ~1–6 ka earlier than some prior Illinois estimates, even assuming slower loess accumulation in the modern solum. Younger loess in northwestern Illinois likely reflects, in part, Superior and Des Moines Lobe glacial-meltwater sediment, and Iowan Erosion Surface inputs to the upper Mississippi Valley, after the Lake Michigan Lobe receded. Furthermore, stronger winds, drier conditions, and reduced vegetation cover in valley deflation areas may have favored higher accumulation rates and later loess deposition in northwestern relative to southeastern areas. [ABSTRACT FROM AUTHOR]

The Grand Canyon provides a deeply dissected view of the aquifers of the Colorado Plateau and its public and tribal lands. Stacked sandstone and karst aquifers are vertically connected by a network of faults and breccia pipes creating a complex groundwater network. Hydrochemical variations define structurally controlled groundwater sub-basins, each with main discharging springs. North Rim (N-Rim), South Rim (S-Rim), and far-west springs have different stable isotope fingerprints, reflecting different mean recharge elevations. Variation within each region reflects proportions of fast/slow aquifer pathways. Often considered perched, the upper Coconino (C) aquifer has a similar compositional range as the regional Redwall-Muav (R-M) karst aquifer, indicating connectivity. Natural and anthropogenic tracers show that recharge can travel 2 km vertically and tens of kilometers laterally in days to months via fracture conduits to mix with older karst baseflow. Six decades of piping N-Rim water to S-Rim Village and infiltration of effluent along the Bright Angel fault have sustained S-Rim groundwaters and likely induced S-Rim microseismicity. Sustainable groundwater management and uranium mining threats require better monitoring and application of hydrotectonic concepts. Hydrotectonic concepts include distinct structural sub-basins, fault fast conduits, confined aquifers, karst aquifers, upwelling geothermal fluids, and induced seismicity. N-Rim, S-Rim, and far-west springs have different stable isotope fingerprints reflecting different mean recharge elevations and residence times. The upper C and lower R-M aquifers have overlapping stable isotope fingerprints in a given region, indicating vertical connectivity between aquifers. S-Rim springs and groundwater wells are being sustained by ∼60 years of piping of N-Rim water to S-Rim, possibly inducing seismicity. [ABSTRACT FROM AUTHOR]

Slackwater deposits representing past flood events provide a robust means to extend systematic gage records further back in time, place historic floods in a longer-term context, and reduce uncertainties in flood hazard analysis. The identification and application of slackwater deposits in riverine paleoflood hydrology has traditionally been limited to arid bedrock-controlled environments and periglacial environments. In this study, we utilize methods developed in humid alluvial settings and apply them to slackwater deposits, one of the first studies to do so. This novel approach uses sediment texture and geochemistry to distinguish slackwater deposits from in situ material in a temperate alluvial setting. We identify multiple slackwater deposits in two rock shelters situated on bluffs adjacent to the lower Ohio River. Flood age estimates are based on optically stimulated luminescence (OSL) dating, and discharge estimates are based on a 1D HEC-RAS model. The uppermost slackwater deposit at both sites corresponds to the AD 1937 historic flood of record (∼31,400 m3/s), while another slackwater deposit identified only at the lower elevation site corresponds to a paleoflood that occurred around AD 1650 with a discharge of ∼23,900 m3/s. Our findings imply that the AD 1937 flood represents the largest magnitude flood on the lower Ohio River in at least the last 400 years. Inclusion of the paleoflood into a flood frequency analysis for the Ohio River at Louisville reduces uncertainties in large flood quantiles by ∼50%. [ABSTRACT FROM AUTHOR]

A case study of a South Wales manufacturer illustrates the need for companies to adopt an integrated strategy for computerization and information systems. Lack of management training blending computing and business skills can have a crippling effect on system development and organizational health. (SK)

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.ejor.2004.05.026 Byline: B. Curry, P.H. Morgan Keywords: Neural Networks; Network weights; Hidden layers; Backpropagation; Polytope Abstract: This paper considers two related issues regarding feedforward Neural Networks (NNs). The first involves the question of whether the network weights corresponding to the best fitting network are unique. Our empirical tests suggest an answer in the negative, whether using standard Backpropagation algorithm or our preferred direct (non-gradient-based) search procedure. We also offer a theoretical analysis which suggests that there will almost inevitably be functional relationships between network weights. The second issue concerns the use of standard statistical approaches to testing the significance of weights or groups of weights. Treating feedforward NNs as an interesting way to carry out nonlinear regression suggests that statistical tests should be employed. According to our results, however, statistical tests can in practice be indeterminate. It is rather difficult to choose either the number of hidden layers or the number of nodes on this basis. Author Affiliation: Cardiff Business School, Cardiff University, Aberconway Building, Colum Drive, Cardiff CF10 3EU, UK Article History: Received 26 February 2003; Accepted 25 May 2004