Estimation of Hydraulic Conductivity and Its Uncertainty from Grain-Size Data Using GLUE and Artificial Neural Networks

Various approaches exist to relate saturated hydraulic conductivity (K _s) to grain-size data. Most methods use a single grain-size parameter and hence omit the information encompassed by the entire grain-size distribution. This study compares two data-driven modelling methods??multiple linear regression and artificial neural networks??that use the entire grain-size distribution data as input for K _s prediction. Besides the predictive capacity of the methods, the uncertainty associated with the model predictions is also evaluated, since such information is important for stochastic groundwater flow and contaminant transport modelling. Artificial neural networks (ANNs) are combined with a generalised likelihood uncertainty estimation (GLUE) approach to predict K _s from grain-size data. The resulting GLUE-ANN hydraulic conductivity predictions and associated uncertainty estimates are compared with those obtained from the multiple linear regression models by a leave-one-out cross-validation. The GLUE-ANN ensemble prediction proved to be slightly better than multiple linear regression. The prediction uncertainty, however, was reduced by half an order of magnitude on average, and decreased at most by an order of magnitude. This demonstrates that the proposed method outperforms classical data-driven modelling techniques. Moreover, a comparison with methods from the literature demonstrates the importance of site-specific calibration. The data set used for this purpose originates mainly from unconsolidated sandy sediments of the Neogene aquifer, northern Belgium. The proposed predictive models are developed for 173 grain-size K _s-pairs. Finally, an application with the optimised models is presented for a borehole lacking K _s data.     

Swift observations of GRB 050712

We present the results of X-ray and optical observations of GRB 050712 performed by Swift . The X-ray light curve of this burst exhibits episodes of flares in the first 1000 s, the same epoch at which the UVOT detected an optical counterpart. A shallow X-ray decay, with a decay slope of  α=−0.73  , followed and lasted ∼70 ks. This behaviour can be explained in terms of activity of the gamma-ray burst 'inner engine', with the possibility that the last flare is caused by the interaction of the ejecta with the surrounding medium.
We also find interesting spectral parameters for the X-ray emission. In particular, data suggest the presence of an intrinsic absorption in the first 1000 s, which can be explained if circumburst medium clouds lie along the line of sight.     

Runoff curve numbers for steep hillslopes with natural vegetation in semi‐arid tropical highlands,northern Ethiopia

Daily runoff from 27 plots (5 m × 2 m) recorded during two rainy seasons in the Tigray highlands (Ethiopia) were analysed together with daily rainfall to calculate runoff curve numbers for hillslopes covered by semi‐natural vegetation in varying stages of vegetation restoration. Curve number model parameters were derived using a least squares fitting procedure on the collected rainfall–runoff datasets. Curve numbers varied from 29 to 97. Land use type was an important explanatory factor for the variation in curve numbers, whereas hydrologic soil group was not. Curve numbers were negatively correlated with vegetation cover. Taking into account antecedent soil moisture conditions did not improve runoff prediction using the curve number method. As runoff prediction was less accurate in areas with low curve numbers, two separate regression functions relating curve numbers with vegetation cover were proposed for different land use types. Copyright © 2008 John Wiley & Sons, Ltd.     

The effect of dock length on harbour siltation

Density-driven exchange flows between estuaries and harbour docks are influenced by the length of the dock. As a result, increasing dock size through its lengthening, not necessarily results in an increase in sedimentation rates. The propagation of a low-salinity surface patch into the dock is blocked at the head of a relatively short dock, resulting in a reversal of density-driven flows, and a reduction of the hydrostatic pressure gradients in the entrance of the dock. A reduced hydrostatic pressure in the dock, in turn, promotes near-bed inflow. When this increased near-bed inflow coincides with a high sediment supply on the adjacent river, the sediment transport into the dock increases. This has been tested with an extensively validated high-resolution numerical model developed for the Deurganckdok in the Port of Antwerp. In the Deurganckdok, siltation rates are expected to decrease when the dock is fully excavated compared to the present half-opened dock.Whether exchange flows between estuaries and harbour docks are influenced by the length of the dock, depends on the tidal variation in salinity. For small tidal density variations (around 0.5 kg/m³), the dock length is expected to influence exchange flows in a short dock (approximately 1 km), whereas the dock should be much longer (4 km) when the tidal density variation is higher (around 5 kg/m³). Whether these changing exchange flow result in a lowering or increase of sediment import, depends on the phase difference between sediment concentration peaks on the adjacent river/estuary and the salinity variation, and on the vertical distribution of sediment.     

Satellite-based peatland mapping: Potential of the MODIS sensor

Peatlands play a major role in the global carbon cycle but are largely overlooked in current large-scale vegetation mapping efforts. In this study, we investigated the potential of the Moderate Resolution Imaging Spectroradiometer (MODIS) to capture the extent and distribution of peatlands in the St. Petersburg region of Russia by analyzing the relationships between peatland cover fractions derived from reference maps and  1-km resolution MODIS Nadir BRDF-Adjusted Reflectance (NBAR) data from year 2002.First, we characterized and mapped 50 peatlands from forest inventory and peat deposit inventory data. The peatlands represent three major nutritional types (oligotrophic, mesotrophic, eutrophic) and different sizes (0.6–7800 ha). In addition, parts of 6 peatlands were mined for peat and these were mapped separately. The reference maps provided information on peatland cover for 1105 MODIS pixels. We performed regression analysis on 50% of the pixels and reserved the remainder for model validation. Canonical correlation analysis on the MODIS reflectance bands and the peatland cover fractions produced a multi-spectral peatland cover index (PCI), which served as the predictor in a reduced major axis (RMA) regression model. The results suggest a high potential for mapping peatlands with MODIS. The RMA regression models explained much of the variance in the PCI (r² = 0.74 for mined and r² = 0.81 for unmined peatlands). Model validation showed high correlation between observed versus predicted peatland cover (mined: r = 0.87; unmined: r = 0.92). We used the models to derive peatland cover estimates for the St. Petersburg region and compared the results to current MODIS land cover maps.     

GRS evidence and the possibility of paleooceans on Mars

The Gamma Ray Spectrometer (Mars Odyssey spacecraft) has revealed elemental distributions of potassium (K), thorium (Th), and iron (Fe) on Mars that require fractionation of K (and possibly Th and Fe) consistent with aqueous activity. This includes weathering, evolution of soils, and transport, sorting, and deposition, as well as with the location of first-order geomorphological demarcations identified as possible paleoocean boundaries. The element abundances occur in patterns consistent with weathering in situ and possible presence of relict or exhumed paleosols, deposition of weathered materials (salts and clastic minerals), and weathering/transport under neutral to acidic brines. The abundances are explained by hydrogeology consistent with the possibly overlapping alternatives of paleooceans and/or heterogeneous rock compositions from diverse provenances (e.g., differing igneous compositions).     

Wave-induced topographic formstress in baroclinic channel flow

Large-scale zonal flow driven across submarine topography establishes standing Rossby waves. In the presence of stratification, the wave pattern can be represented by barotropic and baroclinic Rossby waves of mixed planetary topographic nature, which are locked to the topography. In the balance of momentum, the wave pattern manifests itself as topographic formstress. This wave-induced formstress has the net effect of braking the flow and reducing the zonal transport. Locally, it may lead to acceleration, and the parts induced by the barotropic and baroclinic waves may have opposing effects. This flow regime occurs in the circumpolar flow around Antarctica. The different roles that the wave-induced formstress plays in homogeneous and stratified flows through a zonal channel are analyzed with the BARBI (BARotropic-Baroclinic-Interaction ocean model, Olbers and Eden, J Phys Oceanogr 33:2719–2737, 2003) model. It is used in complete form and in a low-order version to clarify the different regimes. It is shown that the barotropic formstress arises by topographic locking due to viscous friction and the baroclinic one due to eddy-induced density advection. For the sinusoidal topography used in this study, the transport obeys a law in which friction and wave-induced formstress act as additive resistances, and windstress, the effect of Ekman pumping on the density stratification, and the buoyancy forcing (diapycnal mixing of the stratified water column) of the potential energy stored in the stratification act as additive forcing functions. The dependence of the resistance on the system parameters (lateral viscosity ε, lateral diffusivity κ of eddy density advection, Rossby radius λ, and topography height δ) as well as the dependence of transport on the forcing functions are determined. While the current intensity in a channel with homogeneous density decreases from the viscous flat bottom case in an inverse quadratic law ~δ ^–2 with increasing topography height and always depends on ε, a stratified system runs into a saturated state in which the transport becomes independent of δ and ε and is determined by the density diffusivity κ rather than the viscosity: κ/λ ² acts as a vertical eddy viscosity, and the transport is λ ²/κ times the applied forcing. Critical values for the topographic heights in these regimes are identified.     

Passive seismic source localization via common-reflection-surface attributes

The common-reflection-surface (CRS) stack can be viewed as a physically justified extension of the classical common-midpoint (CMP) stack, utilizing redundant information not only in a single, but in several neighboring CMP gathers. The zero-offset CRS moveout is parameterized in terms of kinematic attributes, which utilize reciprocity and raypath symmetries to describe the two-way process of the actual wave propagation in active seismic experiments by the propagation of auxiliary one-way wavefronts. For the diffraction case, only the attributes of a single one-way wavefront, originating from the diffractor are sufficient to explain the traveltime differences observed at the surface. While paraxial ray theory gives rise to a second-order approximation of the CRS traveltime, many higher-order approximations were subsequently introduced either by squaring the second-order expression or by employing principles of optics and geometry. It was recently discovered that all of these higher-order operators can be formulated either for the optical projection or in an auxiliary medium of a constant effective velocity. Utilizing this duality and the one-way nature of the CRS parameters, we present a simple data-driven stacking scheme that allows for the estimation of the a priori unknown excitation time of a passive seismic source. In addition, we demonstrate with a simple data example that the output of the suggested workflow can directly be used for subsequent focusing-based normal-incidence-point (NIP) tomography, leading to a reliable localization in depth.     

Determination of geometrical spreading from traveltimes

Geometrical spreading plays an important role for amplitude preserving migration, which is a very time-consuming process. In order to achieve efficiency in terms of computational time and, particularly, storage space, we propose a method to determine geometrical spreading from coarsely gridded traveltime tables. The method is based on a hyperbolic traveltime expansion and provides also a fast and accurate algorithm for the interpolation of traveltimes, including the interpolation of complete shots. Examples demonstrate the applicability of the method to isotropic and anisotropic media.