Assessment of Exploration Bias in Data-Driven Predictive Models and the Estimation of Undiscovered Resources

The spatial distribution of discovered resources may not fully mimic the distribution of all such resources, discovered and undiscovered, because the process of discovery is biased by accessibility factors (e.g., outcrops, roads, and lakes) and by exploration criteria. In data-driven predictive models, the use of training sites (resource occurrences) biased by exploration criteria and accessibility does not necessarily translate to a biased predictive map. However, problems occur when evidence layers correlate with these same exploration factors. These biases then can produce a data-driven model that predicts known occurrences well, but poorly predicts undiscovered resources. Statistical assessment of correlation between evidence layers and map-based exploration factors is difficult because it is difficult to quantify the “degree of exploration.” However, if such a degree-of-exploration map can be produced, the benefits can be enormous. Not only does it become possible to assess this correlation, but it becomes possible to predict undiscovered, instead of discovered, resources. Using geothermal systems in Nevada, USA, as an example, a degree-of-exploration model is created, which then is resolved into purely explored and unexplored equivalents, each occurring within coextensive study areas. A weights-of-evidence (WofE) model is built first without regard to the degree of exploration, and then a revised WofE model is calculated for the “explored fraction” only. Differences in the weights between the two models provide a correlation measure between the evidence and the degree of exploration. The data used to build the geothermal evidence layers are perceived to be independent of degree of exploration. Nevertheless, the evidence layers correlate with exploration because exploration has preferred the same favorable areas identified by the evidence patterns. In this circumstance, however, the weights for the “explored” WofE model minimize this bias. Using these revised weights, posterior probability is extrapolated into unexplored areas to estimate undiscovered deposits.     

Application of kinematic wave theory for predicting flash flood hazards on coupled alluvial fan–piedmont plain landforms

A rainstorm that caused a severe flash flood on the piedmont plain at the toe positions of two alluvial fans located to the west of the Organ Mountains in Dona Ana County, New Mexico, USA, is analysed. The space–time distributions of rainfall are evaluated from the Next Generation Weather Radar (NEXRAD) and overland flow is modelled as kinematic wave. The spatial distribution of rainfall shows a topographic control. The greatest rainfall depth, duration, and intensity occurred at the higher elevation mountain slopes and decreased with decreasing elevation from the alluvial fans to the piedmont plain. The alluvial fan–piedmont plain system is modelled by coupling divergent and rectangular overland flow planes. Explicit finite difference approximations, hybridized with the analytical method of characteristics, are made to the kinematic wave equations to account for the spatial and temporal distribution of the rainfall and variable boundary conditions. Simulation results indicate that sheet‐flow floodwater elevations rise (1) in a nonlinear fashion from the apex to toe positions of the alluvial fans, and (2) near linearly from the toe positions of the alluvial fans onto the piedmont plains with the formation of kinematic shocks near the middle to the upstream end of the plane at times between the initiation of the rainstorm and the time of concentration of the plane. Thus, the maximum flooding occurs at the middle or upstream sections of the piedmont plains regardless of the pattern of space–time variability of rainfall. These results are in agreement with observed geomorphologic features suggesting that piedmont plains are naturally flood‐prone areas. This case study demonstrates that flood hazards on piedmont plains can exceed those on alluvial fans. The models presented in this study suggest that the flood hazard zones on coupled alluvial fan–piedmont plain landforms should be delineated transverse to the flow directions, as opposed to the flood hazard zones with boundaries in the longitudinal direction of the axis of an alluvial fan. Copyright © 2003 John Wiley & Sons, Ltd.     

TESSIN VISLab—laboratory for scientific visualization

Scientific visualization is an integral part of the modeling workflow, enabling researchers to understand complex or large data sets and simulation results. A high-resolution stereoscopic virtual reality (VR) environment further enhances the possibilities of visualization. Such an environment also allows collaboration in work groups including people of different backgrounds and to present results of research projects to stakeholders or the public. The requirements for the computing equipment driving the VR environment demand specialized software applications which can be run in a parallel fashion on a set of interconnected machines. Another challenge is to devise a useful data workflow from source data sets onto the display system. Therefore, we develop software applications like the OpenGeoSys Data Explorer, custom data conversion tools for established visualization packages such as ParaView and Visualization Toolkit as well as presentation and interaction techniques for 3D applications like Unity. We demonstrate our workflow by presenting visualization results for case studies from a broad range of applications. An outlook on how visualization techniques can be deeply integrated into the simulation process is given and future technical improvements such as a simplified hardware setup are outlined.     

Methods of estimating allowable motion perturbations in side-scansonar systems

This document describes the effects of nonideal hydrophone array motion on sonar image quality and derives the allowable motion errors for vehicles carrying side-scan sonar systems. Estimates of the quantity of motion that causes just noticeable sensor performance degradation as well as those that render the sensors useless are derived using two different methods. Using a special resolution target, examples of simulated image distortion resulting from several types of motion are presented. A method for quantitatively measuring the actual image distortion in situ is described     

Bistatic radar cross section of ship targets

Data describing simultaneous bistatic (BRCS) and monostatic (MRCS) radar cross sections of freighters atXband near grazing incidence are presented. Details of the measurement system, calibration procedures, and target-radar geometry are discussed. Measured values of BRCS were typically less than the MRCS and decreased with increasing bistatic angle.