A New Nonparametric Discriminant Analysis Algorithm Accounting for Bounded Data Errors

In a statistical pattern recognition context, discriminant analysis is designed to classify, when possible, objects into predefined categories. Because this method requires precise input data, uncertainties cannot be propagated in the classifying process. In real case studies, this could lead to drastic misinterpretations of objects. A new nonparametric algorithm based on interval arithmetic has thus been developed to propagate interval-form data. They consist in calculating interval conditional probability density functions and interval posterior probabilities. Objects are eventually assigned to a subset of classes, consistent with the data and their uncertainties. The classifying model is thus less precise, but more realistic than the standard one, which we prove on a real case study.     

Ab initio calculations of the Fe(II) and Fe(III) isotopic effects in citrates,nicotianamine, and phytosiderophore,and new Fe isotopic measurements in higher plants

Iron is one of the most abundant transition metal in higher plants and variations in its isotopic compositions can be used to trace its utilization. In order to better understand the effect of plant-induced isotopic fractionation on the global Fe cycling, we have estimated by quantum chemical calculations the magnitude of the isotopic fractionation between different Fe species relevant to the transport and storage of Fe in higher plants: Fe(II)-citrate, Fe(III)-citrate, Fe(II)-nicotianamine, and Fe(III)-phytosiderophore. The ab initio calculations show firstly, that Fe(II)-nicotianamine is ～3‰ (⁵⁶Fe/⁵⁴Fe) isotopically lighter than Fe(III)-phytosiderophore; secondly, even in the absence of redox changes of Fe, change in the speciation alone can create up to ～1.5‰ isotopic fractionation. For example, Fe(III)-phytosiderophore is up to 1.5‰ heavier than Fe(III)-citrate₂ and Fe(II)-nicotianamine is up to 1‰ heavier than Fe(II)-citrate. In addition, in order to better understand the Fe isotopic fractionation between different plant components, we have analyzed the iron isotopic composition of different organs (roots, seeds, germinated seeds, leaves and stems) from six species of higher plants: the dicot lentil (Lens culinaris), and the graminaceous monocots Virginia wild rye (Elymus virginicus), Johnsongrass (Sorghum halepense), Kentucky bluegrass (Poa pratensis), river oat (Uniola latifolia), and Indian goosegrass (Eleusine indica). The calculations may explain that the roots of strategy-II plants (Fe(III)-phytosiderophore) are isotopically heavier (by about 1‰ for the δ⁵⁶Fe) than the upper parts of the plants (Fe transported as Fe(III)-citrate in the xylem or Fe(II)-nicotianamine in the phloem). In addition, we suggest that the isotopic variations observed between younger and older leaves could be explained by mixing of Fe received from the xylem and the phloem.     

Comprehensive study of the drying behavior of Boom clay: Experimental investigation and numerical modeling

This paper presents a thermo‐hydro‐mechanical framework to model the drying behavior of Boom clay. First, the experimental campaign conducted Noémie Prime is briefly presented because it is used to validate the model. The data acquisition and processing is emphasized because of the use of X‐ray microtomography to be able to more accurately compare experimental and numerical strain fields. The different submodels are introduced. Numerical simulations are performed to illustrate the capability of the proposed model to reproduce the observed behavior. Finally, a comprehensive sensitivity study on several key model parameters associated with the water retention curve, and the permeability of the medium, is performed to get a better understanding of the physics behind the coupled model.     

Modelling wind-erosion risk in the Laghouat region (Algeria) using geomatics approach

Wind-erosion risk is a challenge that threatens land development in dry-land regions. Soil analysis, remote sensing, climatic, vegetal cover and topographic data were used in a geographic information system (GIS), using multi-criteria analysis (MCA) to map wind-erosion risk (R_we) in Laghouat, Algeria. The approach was based on modelling the risk and incorporating topographic and climatic effects. The maps were coded according to their sensitivity to wind erosion and to their socio-economic potential, from low to very high. By overlapping the effects of these layers, qualitative maps were drawn to reflect the potential sensitivity to wind erosion per unit area. The results indicated that severe wind erosion affects mainly all the southern parts and some parts in the north of Laghouat, where wind-erosion hazard (H_we) is very high in 43% of the total area, and which was affected mainly by natural parameters such as soil, topography and wind. The results also identified features vulnerable to R_we. The product of the hazard and the stake maps indicated the potential risk areas that need preventive measures; this was more than half of the study area, making it essential to undertake environmental management and land-use planning.     

Flow dynamics at the origin of thin clayey sand lacustrine turbidites: Examples from Lake Hazar,Turkey

Turbidity currents and their deposits can be investigated using several methods, i.e. direct monitoring, physical and numerical modelling, sediment cores and outcrops. The present study focused on thin clayey sand turbidites found in Lake Hazar (Turkey) occurring in eleven clusters of closely spaced thin beds. Depositional processes and sources for three of those eleven clusters are studied at three coring sites. Bathymetrical data and seismic reflection profiles are used to understand the specific geomorphology of each site. X‐ray, thin sections and CT scan imagery combined with grain‐size, geochemical and mineralogical measurements on the cores allow characterization of the turbidites. Turbidites included in each cluster were produced by remobilization of surficial slope sediment, a process identified in very few studies worldwide. Three types of turbidites are distinguished and compared with deposits obtained in flume studies published in the literature. Type 1 is made of an ungraded clayey silt layer issued from a cohesive flow. Type 2 is composed of a partially graded clayey sand layer overlain by a mud cap, attributed to a transitional flow. Type 3 corresponds to a graded clayey sand layer overlain by a mud cap issued from a turbulence‐dominated flow. While the published experimental studies show that turbulence is damped by cohesion for low clay content, type 3 deposits of this study show evidence for a turbulence‐dominated mechanism despite their high clay content. This divergence may in part relate to input variables, such as water chemistry and clay mineralogy, that are not routinely considered in experimental studies. Furthermore, the large sedimentological variety observed in the turbidites from one coring site to another is related to the evolution of a sediment flow within a field‐scale basin made of a complex physiography that cannot be tackled by flume experiments.     

Record of enhanced late Pliocene tectono‐volcanic activity in the Lesser Antilles forearc basin: Insights into the long‐term behaviour of the Lesser Antilles subduction zone

In this study, we investigate the Lesser Antilles forearc basin, focusing on the late Pliocene to Pleistocene sedimentary archives in order to track the occurrence of extreme events triggered by enhanced subduction‐related tectono‐volcanic activity. We identify late Piacenzian deposits covering a major regional erosional surface, displaying sedimentary dykes and large marine boulders embedded in a mixed continental–marine matrix, characteristic of tsunamites. We interpret this episode of platform emersion and the successive cataclysmic deposits as resulting from enhanced tectonic activity at the interface of the subduction zone, synchronous with the initiation of the Lesser Antilles volcanic arc. We then discuss the implications in terms of the mechanical behaviour of the Lesser Antilles subduction zone.     

Radon emanation of heterogeneous basin deposits in Kathmandu Valley,Nepal

Effective radium-226 concentration (EC_Ra) has been measured in soil samples from seven horizontal and vertical profiles of terrace scarps in the northern part of Kathmandu Valley, Nepal. The samples belong to the Thimi, Gokarna, and Tokha Formations, dated from 50 to 14 ky BP, and represent a diverse fluvio-deltaic sedimentary facies mainly consisting of gravelly to coarse sands, black, orange and brown clays. EC_Ra was measured in the laboratory by radon-222 emanation. The samples (n = 177) are placed in air-tight glass containers, from which, after an accumulation time varying from 3 to 18 days, the concentration of radon-222, radioactive decay product of radium-226 and radioactive gas with a half-life of 3.8 days, is measured using scintillation flasks. The EC_Ra values from the seven different profiles of the terrace deposits vary from 0.4 to 43 Bq kg^?1, with profile averages ranging from 12 ± 1 to 27 ± 2 Bq kg^?1. The values have a remarkable consistency along a particular horizon of sediment layers, clearly demonstrating that these values can be used for long distance correlations of the sediment horizons. Widely separated sediment profiles, representing similar stratigraphic positions, exhibit consistent EC_Ra values in corresponding stratigraphic sediment layers. EC_Ra measurements therefore appear particularly useful for lithologic and stratigraphic discriminations. For comparison, EC_Ra values of soils from different localities having various sources of origin were also obtained: 9.2 ± 0.4 Bq kg^?1 in soils of Syabru–Bensi (Central Nepal), 23 ± 1 Bq kg^?1 in red residual soils of the Bhattar-Trisuli Bazar terrace (North of Kathmandu), 17.1 ± 0.3 Bq kg^?1 in red residual soils of terrace of Kalikasthan (North of Trisuli Bazar) and 10 ± 1 Bq kg^?1 in red residual soils of a site near Nagarkot (East of Kathmandu). The knowledge of EC_Ra values for these various soils is important for modelling radon exhalation at the ground surface, in particular in the vicinity of active faults. Importantly, the study also reveals that, above numerous sediments of Kathmandu Valley, radon concentration in dwellings can potentially exceed the level of 300 Bq m^?3 for residential areas; a fact that should be seriously taken into account by the governmental and non-governmental agencies as well as building authorities.     

Do paleoclimate proxies agree? A test comparing 19 late Holocene climate and sea‐ice reconstructions from Icelandic marine and lake sediments

Geochemical, mineralogical and biological indicators preserved in sediments are widely used to reconstruct past climate change, but proxies differ in the degree to which their utility as climate indicators has been validated via laboratory experiments, modern spatial calibrations, or down‐core comparisons with instrumental climate data. Multi‐proxy studies provide another means of evaluating interpretations of proxies. This paper presents a multi‐proxy assessment comparing 19 sub‐centennially resolved late Holocene proxy records, covering the period 300–1900 AD, from seven Icelandic marine and lacustrine core sites. We employ simple statistical comparisons between proxy reconstructions to evaluate their correlations over time and, ultimately, their utility as proxies for regional climate. Proxies examined include oxygen isotopic composition of benthic and planktonic foraminifera, abundance of the sea‐ice biomarker IP₂₅, allochthonous quartz in marine sediments (a proxy for drift ice around Iceland), marine carbonate abundance, total organic carbon concentration, chironomid assemblages, lacustrine biogenic silica and carbon/nitrogen ratios in lake sediments. Most of the examined proxy records, including temperature and sea‐ice proxies, correlate strongly with each other over multi‐centennial timescales, and thus do appear to record changes in regional climate. Copyright © 2011 John Wiley & Sons, Ltd.     

Dispersion modeling of air pollutants in the atmosphere: a review

Modeling of dispersion of air pollutants in the atmosphere is one of the most important and challenging scientific problems. There are several natural and anthropogenic events where passive or chemically active compounds are emitted into the atmosphere. The effect of these chemical species can have serious impacts on our environment and human health. Modeling the dispersion of air pollutants can predict this effect. Therefore, development of various model strategies is a key element for the governmental and scientific communities. We provide here a brief review on the mathematical modeling of the dispersion of air pollutants in the atmosphere. We discuss the advantages and drawbacks of several model tools and strategies, namely Gaussian, Lagrangian, Eulerian and CFD models. We especially focus on several recent advances in this multidisciplinary research field, like parallel computing using graphical processing units, or adaptive mesh refinement.