Robust design of rock slopes with multiple failure modes: modeling uncertainty of estimated parameter statistics with fuzzy number

The variability of shear characteristics of rock discontinuities is often difficult to ascertain. Thus, even with the reliability-based design (RBD) approach, which allows for consideration of the uncertainty of input parameters, the design of a rock slope system may be either cost-inefficient (overdesign) or unsafe (under-design), depending on whether the variation of input parameters is overestimated or underestimated. The uncertainty about the variation of input parameters is a critical issue in a RBD. This paper presents a feasible approach to addressing this problem using robust design concept. First, the uncertainty of the estimated statistics of input parameters (such as rock properties) is represented by fuzzy sets, which requires only the knowledge of lower and upper bounds of the estimated statistics. Then, the robust design concept is implemented to ensure that the final design is insensitive to, or robust against, the uncertainty of the estimated statistics of input parameters. The design methodology is demonstrated with an application to the design of a rock slope system with multiple failure modes. This design methodology, termed robust geotechnical design (RGD), aims to achieve a certain level of design robustness, in addition to meeting safety and cost requirements. In this paper, the RGD framework is realized through a multi-objective optimization, as it involves three requirements, safety, cost, and robustness. The significance of the design methodology is demonstrated with an example of rock slope design.     

Testing a conceptual hillslope recession model based on the storage–discharge relationship with the Richards equation

The conceptual recession model based on the storage–discharge relationship was proposed to account for the unsaturated–saturated water storage interaction. The recession model was formulated by combining the constitutive storage–discharge relationship with the integral balance equation for unsaturated and saturated water storage. The functional form of the constitutive storage–discharge relationship was determined from the spatial integration of the Richards equation. The performance of the recession model was tested by comparing with the solution of the Richards equation for different simulation geometric shapes and soil types. The conceptual recession model incorporating the unsaturated–saturated water storage interaction was in good agreement with the recession response of the Richards equation. However, the recession model that neglected the unsaturated–saturated water storage interaction was comparable to the Richards equation only for soils with the weak interaction between unsaturated water storage and saturated water storage. This result suggests the important role of the unsaturated–saturated water storage interaction in the formulation of the recession process when the derivative of the functional relationship between the unsaturated water storage and saturated water storage becomes significant. Copyright © 2007 John Wiley & Sons, Ltd.     

Predicting soil erosion hazard in Lattakia Governorate(W Syria)

The main objective of this study is to predict soil erosion in the Lattakia Governorate(WSyria)using the Water Erosion Prediction Project model(WEPP)and to compare the result with that of the RUSLE.Field survey and data collection were carried out,and 44 soil samples were analyzed.In addition,all the necessary input files were prepared for use in the WEPP model and RUSLE.Results show that more than of 80%of the locations studied experience slight to moderate erosion(less than 5 t/ha/y),whereas the rest of the locations experience severe soil erosion hazard.Moreover,the volume of runoff estimated by the WEPP model is in the range of 51e321 mm,and the R^2 between the simulated soil erosion and the predicted runoff reached 0.68.Interestingly,the R^2 between the WEPP model and RUSLE is 0.56,which indicates a good correlation between the two models.     

Widespread overspill from a saline density‐current channel and its interaction with topography on the south‐west Black Sea shelf

Seaward of the Bosphorus Strait, the south‐west Black Sea shelf is dominated by the world's largest channel network maintained by a quasi‐continuous saline (ca 35 → 31 psu) underflow. Calculations indicate that >85% of the initial discharge of ca 10⁴ m³s^?1 spills overbank before the shelf edge. This paper documents interaction of the overspill with sea bed topography using multibeam bathymetry, echo‐sounder images of the water column, conductivity–temperature–depth profiles and sediment cores. Overbank spill is widespread, particularly through crevasse channels and on the middle shelf where confinement by channel banks is negligible. Towards the outer shelf, the wind‐driven Rim Current advects mud along the shelf, contributing to levée successions and deposition on stoss sides of elongate transverse ridges. Echo‐sounder profiles reveal metre‐scale eddies over megaflutes, and breaking lee waves and internal hydraulic jumps over ridges. Megaflutes reach 600 m long and 7 m deep, yet form where the underflow, outside the flute, is no thicker than ca 2 to 5 m. Two types of elongate seaward‐facing ridges are recognized. Type 1 ridges, 2 to 5 m high, consist of bivalve‐rich muddy sand in low‐angle (3·5° to 6°) cross‐sets created by the underflow. Type 2 ridges, ca 5 m high, have crests up to 2 km long and a buried wedge‐shaped foundation (the ‘ridge‐core’) comprised of facies similar to Type 1 ridges. These ridge‐cores are blanketed on the landward side by stratified muds, and are capped by obliquely oriented ribs supporting a diverse benthic community. This facies distribution is interpreted to result from stoss‐side and lee‐side velocity and turbulence fluctuations induced by internal hydraulic jumps and breaking lee waves in overspilling portions of the underflow. Experimental results published by W.H. Snyder and co‐workers effectively explain ridge evolution and flow across the ridges, and therefore can be applied with confidence to less easily studied deep‐marine settings swept by turbidity currents.     

Hyper-parameter optimization of gradient boosters for flood susceptibility analysis

Properly choosing hyper-parameters improves machine learning models' performance and reduces training time and resource requirements. In this study, we investigated the uses of the Bayesian optimization algorithm for hyper-parameter searches of two classifiers, namely LightGBM and XGBoost. The models were verified with a dataset from Vietnam, including historical flood locations from satellite images and survey data, and 11 features from three groups, namely physical, hydrological, and human-related factors. The models' performance was evaluated using Area under Receiver Operating Characteristic curves (AUC-ROC). Several strategies were applied to avoid over-fitting, and the results show that two tuned Gradient boosters reached considerably high AUC values (approximately 0.98) compared with the previous study with a similar dataset. The model interpretation was also implemented using the Shapley (SHAP) values to understand better how models work and the interactions between features. The search for optimal hyper-parameters is worth investigating in the future, particularly when there is growing work for novel optimization algorithms. The verification of such an approach is scientifically sound, and the models can be used as an alternative solution for natural hazard analysis in countries prone to hazards.     

Monitoring precipitable water vapor in real-time using global navigation satellite systems

This paper addresses real-time monitoring of the precipitable water vapor (PWV) from GNSS measurements and presents some results obtained from 6-month long GNSS PWV experiments using international and domestic GNSS networks. In the real-time GNSS PWV monitoring system a server/client structure is employed to facilitate formation of PWV networks and single-differenced GNSS measurements are utilized to mitigate errors in GNSS satellites’ orbits and clocks. An issue relating to baseline length between the server and clients is discussed in detail and as a result the PWV monitor is configured to perform in two modes depending on the baseline length. The server estimates sequentially the zenith wet delay of the individual stations, which is then converted into the PWV of the stations. We evaluate system performance by comparing the real-time PWV solution with reference solutions including meteorological measurements obtained with radiosondes and deferred-time precision GNSS PWV solutions. Results showed that the standard deviation of difference between the real-time PWV and the reference solutions ranged from 2.1 to 3.4 mm in PWV for a 6-month long comparison, which was improved to 1.4 to 2.9 mm by reducing comparison period to 20 days in winter.     

Disparities in accessibility of public high schools,in metropolitan Baton Rouge,Louisiana 1990–2010

In this study, we use a Geographic Information System-based index to measure spatial accessibility of public high schools. The index is applied to the Baton Rouge Metropolitan Statistical Area in 1990, 2000, and 2010. Overall, urban areas had consistently worse accessibility scores compared with suburban and rural areas. High schools with more African American students tended to have poorer accessibility levels. Schools with lower accessibility tended to be associated with poorer average student performance, and the effect strengthened between 2000 and 2010. In developing a better understanding of the accessibility of public high schools, this paper reaffirms the significance of space, place, and race, and challenges the notion of declining relevance of accessibility in Louisiana public education policymaking.     

Minimising errors in estimating fish population and biomass densities using the ‘acoustic volume backscattering strength’ method

The basic acoustical methods for estimating the volume backscattering strength and the absolute biomass density are commonly known. However, the detailed techniques for minimising errors in fitting the calibrated parameters of an echo sounding system in these computations are still not adequately examined and discussed. This paper describes and discusses the following computing techniques involved in the absolute biomass estimation: (i) the fitting of the calibrated time varied gain (TVG) curve, the selection of its operating interval, and the correction of its response; (ii) the compensation for the limitation of the signal bandwidth; (iii) the computation of the equivalent beam width of a transducer using its calibrated beam patterns; and (iv) the accuracy in the conversion from the volume backscattering to the absolute fish density using individual fish target strength measurements.     

Inversion of inline and broadside marine controlled‐source electromagnetic data with constraints derived from seismic data

We present a structural smoothing regularization scheme in the context of inversion of marine controlled‐source electromagnetic data. The regularizing hypothesis is that the electrical parameters have a structure similar to that of the elastic parameters observed from seismic data. The regularization is split into three steps. First, we ensure that our inversion grid conforms with the geometry derived from seismic. Second, we use a seismic stratigraphic attribute to define a spatially varying regularization strength. Third, we use an indexing strategy on the inversion grid to define smoothing along the seismic geometry. Enforcing such regularization in the inversion will encourage an inversion result that is more intuitive for the interpreter to deal with. However, the interpreter should also be aware of the bias introduced by using seismic data for regularization. We illustrate the method using one synthetic example and one field data example. The results show how the regularization works and that it clearly enforces the structure derived from seismic data. From the field data example we find that the inversion result improves when the structural smoothing regularization is employed. Including the broadside data improves the inversion results even more, due to a better balancing between the sensitivities for the horizontal and vertical resistivities.