Artificial Neural Networks: A Solution to the Ambiguity in Prediction of Engineering Properties of Fine-Grained Soils

Determination of soaked california bearing ratio (CBR) and compaction characteristics of soils in the laboratory require considerable time and effort. To make a preliminary assessment of the suitability of soils required for a project, prediction models for these engineering properties on the basis of laboratory tests—which are quick to perform, less time consuming and cheap—such as the tests for index properties of soils, are preferable. Nevertheless researchers hold divergent views regarding the most influential parameters to be taken into account for prediction of soaked CBR and compaction characteristics of fine-grained soils. This could be due to the complex behaviour of soils—which, by their very nature, exhibit extreme variability. However this disagreement is a matter of concern as it affects the dependability of prediction models. This study therefore analyses the ability of artificial neural networks and multiple regression to handle different influential parameters simultaneously so as to make accurate predictions on soaked CBR and compaction characteristics of fine-grained soils. The results of simple regression analyses included in this study indicate that optimum moisture content (OMC) and maximum dry density (MDD) of fine-grained soils bear better correlation with soaked CBR of fine-grained soils than plastic limit and liquid limit. Simple regression analyses also indicate that plastic limit has stronger correlation with compaction characteristics of fine-grained soils than liquid limit. On the basis of these correlations obtained using simple regression analyses, neural network prediction models and multiple regression prediction models—with varying number of input parameters are developed. The results reveal that neural network models have more ability to utilize relatively less influential parameters than multiple regression models. The study establishes that in the case of neural network models, the relatively less powerful parameters—liquid limit and plastic limit can also be used effectively along with MDD and OMC for better prediction of soaked CBR of fine-grained soils. Also with the inclusion of less significant parameter—liquid limit along with plastic limit the predictions on compaction characteristics of fine-grained soils using neural network analysis improves considerably. Thus in the case of neural network analysis, the use of relatively less influential input parameters along with stronger parameters is definitely beneficial, unlike conventional statistical methods—for which, the consequence of this approach is unpredictable—giving sometimes not so favourable results. Very weak input parameters alone need to be avoided for neural network analysis. Consequently, when there is ambiguity regarding the most influential input parameters, neural network analysis is quite useful as all such influential parameters can be taken to consideration simultaneously, which will only improve the performance of neural network models. As soils by their very nature, exhibit extreme complexity, it is necessary to include maximum number of influential parameters—as can be determined easily using simple laboratory tests—in the prediction models for soil properties, so as to improve the reliability of these models—for which, use of neural networks is more desirable.     

Global Nitrogen Cycle: Pre-Anthropocene Mass and Isotope Fluxes and the Effects of Human Perturbations

We present a nitrogen cycle model for pre-industrial times based on an extensive literature database. The model consists of 18 reservoirs in the domains of the atmosphere, land, and ocean. The biotic reservoirs on land and in the ocean (N-fixing plants, non-N-fixing plants, and marine biota) interact with atmospheric N₂ and dissolved inorganic nitrogen (DIN, consisting of N₂, NO₃ ^?, and NH₄ ⁺) in the ocean and soil waters. Marine DIN is taken up by marine biota and transformed from ocean particulate organic matter to dissolved organic nitrogen and the ocean sediment. The atmosphere, the largest nitrogen reservoir, supplies N₂ to the system by N fixation, deposition, and dissolution, and these input fluxes are balanced by denitrification and volatilization back to the atmosphere. The land and ocean domains are linked by river transport, which carries both dissolved and particulate nitrogen to the oceanic coastal zone. The isotope–mass balances of the N reservoirs are calculated from the isotopic composition of the reservoirs and the fractionation factors accompanying the fluxes between the reservoirs based on reported values from different natural conditions. The model sensitivity was tested for different biouptake rates and was run with various human perturbations, including fertilization, nitrous oxide emissions, population-related sewage disposal, land-use changes, and temperature-dependent rate kinetics. The new N mass–isotope cycle model provides the basis for assessment of the impact of artificial fertilization between 1700 and 2050. The perturbation experiments in this study suggest that land-use change is the key factor altering the N mass cycle since industrialization.     

Vegetation structure controls on snow and soil moisture in restored ponderosa pine forests

Thinning of semi-arid forests to reduce wildfire risk is believed to improve forest health by increasing soil moisture. Increased snowpack, reduced transpiration and reduced rainfall interception are frequently cited mechanisms by which reduced canopy density may increase soil moisture. However, the relative importance of these factors has not been rigorously evaluated in field studies. We measured snow depth, snow water equivalent (SWE) and the spatial and temporal variation in soil moisture at four experimental paired treatment-control thinning sites in high elevation ponderosa pine forest northern Arizona, USA. We compared snow and soil moisture measurements with forest structure metrics derived from aerial imagery and 3-dimensional lidar data to determine the relationship between vegetation structure, snow and soil moisture throughout the annual hydrologic cycle. Soil moisture was consistently and significantly higher in thinned forest plots, even though the treatments were performed 8–11 years before this study. However, we did not find evidence that SWE was higher in thinned forests across a range of snow conditions. Regression tree analysis of soil moisture and vegetation structure data provided some evidence that localized differences in transpiration and interception of precipitation influence the spatial pattern of soil moisture at points in the annual hydrologic cycle when the system is becoming increasingly water limited. However, vegetation structure explained a relatively low amount of the spatial variance (R² < 0.23) in soil moisture. Continuous measurements of soil moisture in depth profiles showed stronger attenuation of soil moisture peaks in thinned sites, suggesting differences in infiltration dynamics may explain the difference in soil moisture between treatments as opposed to overlying vegetation alone. Our results show limited support for commonly cited relationships between vegetation structure, snow and soil moisture and indicate that future research is needed to understand how reduction in tree density alters soil hydraulic properties.     

Dynamical constraints on the Local Group from the CMB and 2MRS dipoles

We place constraints on the dynamics of the Local Group (LG) by comparing the dipole of the cosmic microwave background (CMB) with the peculiar velocity induced by the Two Micron All-Sky Redshift Survey galaxy sample. The analysis is limited by the lack of surveyed galaxies behind the zone of avoidance (ZoA). We therefore allow for a component of the LG velocity due to unknown mass concentrations behind the ZoA, as well as for an unknown transverse velocity of the Milky Way relative to the Andromeda galaxy. We infer extra motion along the direction of the Galactic Centre (where Galactic confusion and dust obscuration peaks) at the 95 per cent significance level. With a future survey of the ZoA it might be possible to constrain the transverse velocity of the Milky Way relative to Andromeda.     

Numerical study on basin-edge effects in the seismic response of the Gubbio valley,Central Italy

The Gubbio basin in Central Italy is a intermountain basin of extensional tectonic origin, typical of Central and Southern Apennines, characterized by moderate seismicity. The strongest recorded event within the area is a magnitude 5.7 earthquake which occurred on 29 April 1984 along the Gubbio fault, bordering the eastern side of the basin. The main objective of this study is to analyze the features of earthquake ground motion as related to basin-edge effects, by performing physics-based numerical simulations of the 1984 earthquake through a high-performance spectral element code. The simulated ground motions are found in reasonable agreement with the recorded motions when using the kinematic source model developed by Ameri et al. (Bull Seismol Soc Am 99:647–663, 2009), with a rise-time equal to 1 s and a nucleation point located in the middle of the fault. Pronounced differences were noted between records from the basin and adjacent sites at outcropping bedrock, owing to both the strong impedance contrast between soft alluvial sites and bedrock formations (lithostratigraphic amplification), as well as lateral discontinuities related to the 2D/3D geometry of the basin (generation of surface waves). Since the fault was located beneath the basin, 1D amplification effects were found to be more relevant than those associated with the generation of surface waves from the basin edge. Finally, an envelope delay spectrum was computed for the simulated ground motions, showing that surface waves are excited in the frequency band of 0.2–0.8 Hz with a significant increase of ground motion duration within the basin.     

The 21-cm power spectrum after reionization

We discuss the 21-cm power spectrum (PS) following the completion of reionization. In contrast to the reionization era, this PS is proportional to the PS of mass density fluctuations, with only a small modulation due to fluctuations in the ionization field on scales larger than the mean-free-path of ionizing photons. We derive the form of this modulation, and demonstrate that its effect on the 21-cm PS will be smaller than 1 per cent for physically plausible models of damped Lyα systems. In contrast to the 21-cm PS observed prior to reionization, in which H  ii regions dominate the ionization structure, the simplicity of the 21-cm PS after reionization will enhance its utility as a cosmological probe by removing the need to separate the PS into physical and astrophysical components. As a demonstration, we consider the Alcock–Paczynski test and show that the next generation of low-frequency arrays could measure the angular distortion of the PS at the per cent level for   z ∼ 3–5  .     

Evidence for merger-driven activity in the clustering of high-redshift quasars

Recently, a very large clustering length has been measured for quasars at a redshift of   z ∼ 4  . In combination with the observed quasar luminosity function, we assess the implications of this clustering for the relationship between quasar luminosity and dark matter halo mass. Our analysis allows for non-linearity and finite scatter in the relation between quasar luminosity and halo mass, as well as a luminosity dependent quasar lifetime. The additional novel ingredient in our modelling is the allowance for an excess in the observed bias over the underlying halo bias owing to the merger driven nature of quasar activity. We find that the observations of clustering and luminosity function can be explained only if both of the following conditions hold: (i) the luminosity to halo mass ratio increases with halo mass; (ii) the observed clustering amplitude is in excess of that expected solely from halo bias. The latter result is statistically significant at the 99 per cent level. Taken together, the observations provide compelling evidence for merger driven quasar activity, with a black-hole growth that is limited by feedback. In difference from previous analyses, we show that there could be scatter in the luminosity–halo mass relation of up to 1 dex, and that quasar clustering cannot be used to estimate the quasar lifetime.     

The distribution and behaviour of tar balls along the Israeli coast

A fortnightly sampling of tar balls on six beaches along the Israeli coastline between 14 April 1975 and 25 June 1976 showed that the mean content of tar during that period was 3625 g m^?1 of beach front. The northern and central parts of the coast were significantly more polluted than the southern part. Between July 1975 and February 1976 the mean tar quantity decreased continuously from 5635 to 1344 g m^?1.A comparison of tar quantities on the Israeli beaches with those of other beaches in the world showed that the Israeli beaches are more polluted than those of the west Atlantic coast, are as polluted as other beaches on the eastern coast of the Mediterranean Sea, and are less polluted than the beaches of Alexandria, Egypt, and Paphos, Cyprus. It is suggested that the tar content on the beach is related to the degree of oil pollution in the sea. The closer a beach is to an oil shipping lane or an oil dumping site, the heavier it is polluted.During storms, beach tar balls are pushed by the waves to the back of the beach or, in the case of a cliffed coast, are carried along the shore by the longshore current. When the tar balls reach a gap in the cliff (such as an estuary), they are carried inland by the storm waves. There the tar may become buried or dry, shrink and break into small particles which are then dispersed by the wind.     

The formation of natural cryogenic brines

The source of salts in the Ca-chloridic, hypersaline brines (up to 190 g Cl L⁻¹) occurring in crystalline basement rocks in the Canadian, Fennoscandian and Bohemian Shields and their evolution have been investigated and reported. The Cl-Br-Na relationship indicates that these waters have been concentrated from seawater, by freezing during glacial times. The Na/Cl ratio (0.25 to 0.35) in the more saline fluids is compatible with cooling down to −30°C, where the most saline waters have been concentrated by a factor of 25 to 30 relative to the parent seawater.The brines formed from seawater within cryogenic troughs, along the subarctic continental margins, around ice sheets. The depressions within which the brines formed are the cryogenic analogues of the classic, evaporitic lagoon. One million years suffice to saturate with brine a 2000km-radius by 1km-depth rock volume at an H₂O removal rate of only 2.8 mm/yr. Density-induced brine migration on a continental scale takes place via fissures below the ice.Our calculations, that were performed on a hypothetical ice sheet with dimensions compatible with the Laurentide ice sheet, demonstrate that during 1m.y., a 60m thick cryogenic sediment section could have formed. However, the precipitated minerals (mirabilite and hydrohalite) are repeatedly dispersed by the advance and retreat of the ice sheet, dissolved by melt water-seawater mixtures, and eroded during postglacial uplift, leaving almost no trace in the geological record.The cryogenic brines formed intermittently during and between glacial periods. The repeating advance and retreat of the ice sheets exerted a major control on the direction and intensity of brine flow. The cryogenic concentration of seawater and the migration of brine towards the center of the glaciostatic depression occurred mainly during the build up of the ice sheet, while reversal of the water flow from the center of the cryogenic basin outwards happened upon deglaciation. The flow of the waters in the subsurface was, inevitably, accompanied by significant dilution with melt water from the ice sheets.Using a “granitic” U concentration of 4 ppm and a (Ca-Mg mass balance based) rock/water ratio anywhere between 3.4 and 6.8 kg L⁻¹, a few hundred thousand years of brine-rock interaction are sufficient for the growth of ¹²⁹I in the most saline Canadian Shield brine to its present concentration (3.4×10⁸ atoms ¹²⁹I L⁻¹). Hence, both the formation of the saline fluids and their emplacement in their present sites occurred most likely within the Pleistocene.The young age calculated for cryogenic brines in crystalline shields and the dynamic water flow therein should raise concern about the planning and construction of high-grade nuclear waste repositories in such rocks, which are already under way.