Shallow Dialogue Processing Using Machine Learning Algorithms (or not)

This paper presents a shallow dialogue analysis model, aimed at human-human dialogues in the context of staff or business meetings. Four components of the model are defined, and several machine learning techniques are used to extract features from dialogue transcripts: maximum entropy classifiers for dialogue acts, latent semantic analysis for topic segmentation, or decision tree classifiers for discourse markers. A rule-based approach is proposed for solving cross-modal references to meeting documents. The methods are trained and evaluated thanks to a common data set and annotation format. The integration of the components into an automated shallow dialogue parser opens the way to multimodal meeting processing and retrieval applications.

Fate and effects of drilling wastes in a shallow estuarine mesocosm

To address concerns expressed about the possible effect of drilling mud discharges on shallow, low-energy estuarine ecosystems, a 12 month study was designed to detect alterations in water quality and sediment geochemistry. Each drilling mud used in the study and sediments from the study site were analyzed in the laboratory for chemical and physical characteristics. Potential water quality impacts were simulated by the EPA-COE elutriation test procedure. Mud toxicity was measured by acute and chronic bioassays with Mysidopsis bahia, Mercenaria mercenaria, and Nereis virens.^ For the field study, a relatively pristine, shallow (1.2 m) estuary (Christmas Bay, TX) without any drilling activity for the last 30 years was chosen for the study site. After a three month baseline study, three stations were selected. Station 1 was an external control. At each treatment station (2, 3), mesocosms were constructed to enclose a 3.5 m$\sp3$ water column. Each treatment station included an internal control site also. Each in situ mesocosm, except the controls, was successively dosed at a mesocosm-specific dose (1:100; 1:1,000; or 1:10,000 v/v) with 4 field collected drilling muds (spud, nondispersed, lightly-treated, and heavily-treated lignosulfonate) in sequential order over 1.5 months. Twenty-four hours after each dose, water exchange was allowed until the next treatment. Station 3 was destroyed by a winter storm. After the last treatment, the enclosures were removed and the remaining sites monitored for 6 months. One additional site was similarly dosed (1:100 v/v) with clean dredged sediment from Christmas Bay for comparison between dredged sediments and drilling muds.^ Results of the analysis of the water samples and field measurements showed that water quality was impacted during the discharges, primarily at the highest dose (1:100 v/v), but that elevated levels of C, Cr (T,F), Cr$\sp{+3}$ (T, F), N, Pb, and Zn returned to ambient levels before the end of the 24 hour exposure period or immediately after water exchange was allowed (Al, Ba(T), Chlorophyll ABC, SS, %T). Barium, from the barite, was used as a geochemical tracer in the sediments to confirm estimated doses by mass balance calculations. Barium reached a maximum of 166x background levels at the high dose mesocosm. Barium levels returned to ambient or only slightly elevated levels at the end of the 6 month monitoring period due to sediment deposition, resuspension, and bioturbation. QA/QC results using blind samples consisting of lab standards and spiked samples for both water and sediment matrices were within acceptable coefficients of variation.^ In order to avoid impacts on water quality and sediment geochemistry in a shallow estuarine ecosystem, this study concluded that a minimal dilution of 1:1,000 (v/v) would be required in addition to existing regulatory constraints. ^

Physicochemical properties and 3D geostatistical simulations of the Herten and the Descalvado aquifer analogs

This data set presents a comprehensive characterisation of the sedimentary structures from important groundwater hosting formations in Germany (Herten aquifer analog) and Brazil (Descalvado aquifer analog). Multiple 2-D outcrop faces are described in terms of hydraulic, thermal and chemical properties and interpolated in 3D using stochastic techniques. For each aquifer analog, multiple 3D realisations of the facies heterogeneity are provided using different stochastic simulations settings. These are unique analogue data sets that can be used by the wider community to implement approaches for characterising aquifer formations.

Stable isotope composition of carbonate components of Rhaetian shallow-water limestones of the Wombat Plateau

Stable oxygen- and carbon-isotope ratios of Rhaetian (upper Triassic) limestone samples from the Wombat Plateau, northwest Australia, were measured to explore possible diagenetic pathways that the material underwent after deposition in a shallow-water environment, before plateau submergence in the Early Cretaceous. Host sediment isotopic values cluster near typical marine carbonate values (d18O ranging from -2.57 per mil to +1.78 per mil and d13C, from +2.45 per mil to +4.01 per mil). Isotopic values of equant clear calcite lining or filling rock pores also plot in the field of marine cements (d18O = +1.59 per mil to -2.24 per mil and d13C = +4.25 per mil to +2.57 per mil), while isotopic values for neomorphic calcites replacing skeletal (megalodontid shell) carbonate material show a wider scatter of oxygen and carbon values, d18O ranging from +2.73 per milo to -6.2 per mil and d13C, from +5.04 per mil to +1.22 per mil. Selective dissolution of metastable carbonate phases (aragonite?) and neomorphic replacement of skeletal material probably occurred in a meteoric phreatic environment, although replacement products (inclusion-rich microspar, clear neomorphic spar, etc.) retained the original marine isotopic signature because transformation probably occurred in a closed system dominated by the composition of the dissolving phases (high rock/water ratio). The precipitation of late-stage equant (low-Mg?) calcite cement in the pores occurred in the presence of normal marine waters, probably in a deep-water environment, after plateau drowning. Covariance of d18O and d13C toward negative values indeed suggests influence of meteorically modified fluids. However, none of the samples shows negative carbon values, excluding the persistence of organic-rich soils on subaerial karstic surfaces (Caribbean-style diagenesis). Petrographical and geochemical data are consistent with the sedimentological evidence of plateau drowning in post-Rhaetian times and with a submarine origin of the >70-m.y.-long Jurassic hiatus.

(Table 1) Boron-isotope measurements of shallow-dwelling planktonic foraminifera of ODP Holes 144-817A, 144-872C and 143-865C

Knowledge of the evolution of atmospheric carbon dioxide concentrations throughout the Earth's history is important for a reconstruction of the links between climate and radiative forcing of the Earth's surface temperatures. Although atmospheric carbon dioxide concentrations in the early Cenozoic era (about 60 Myr ago) are widely believed to have been higher than at present, there is disagreement regarding the exact carbon dioxide levels, the timing of the decline and the mechanisms that are most important for the control of CO2 concentrations over geological timescales. Here we use the boron-isotope ratios of ancient planktonic foraminifer shells to estimate the pH of surface-layer sea water throughout the past 60 million years, which can be used to reconstruct atmospheric CO2 concentrations. We estimate CO2 concentrations of more than 2,000 p.p.m. for the late Palaeocene and earliest Eocene periods (from about 60 to 52 Myr ago), and find an erratic decline between 55 and 40 Myr ago that may have been caused by reduced CO2 outgassing from ocean ridges, volcanoes and metamorphic belts and increased carbon burial. Since the early Miocene (about 24 Myr ago), atmospheric CO2 concentrations appear to have remained below 500 p.p.m. and were more stable than before, although transient intervals of CO2 reduction may have occurred during periods of rapid cooling approximately 15 and 3 Myr ago.

Succession of benthic hard-bottom communities in the shallow sublitoral of Comau Fjord, Chile

Succession was already studied over decades. The present thesis investigated the succession on hard substrate at two different study sites within the fjord Comau, Chile. Nine plates were installed at both sites (mouth of fjord and inner fjord) and photographed over three years. Additionally the natural community was recorded and a ground truthing was carried out to verify the analyzed species. Respectively at both sites over 50 different species were identified. Abundance data decreased with only one exception continuously, whereas the percentage cover increased. But the communities on the recruitment plates do still not reach the community structure of the natural environment. The present data showed that the hard-bottom succession in the fjord Comau is best described by the TOLERANCE MODEL (Connell & Slatyer, 1977). An important species of the natural community is the stony coral Desmophyllum dianthus, which normally (outside the fjord) grows beneath 1000 m water depth. The results of this work indicate that the mature community is not reached after 36 months.