Use of Noise to Augment Training Data: A Neural Network Method of Mineral–Potential Mapping in Regions of Limited Known Deposit Examples     

Warick M. Brown  Tamás D. Gedeon  David I. Groves 《Natural Resources Research》2003,12(2):141-152

One of the main factors that affects the performance of MLP neural networks trained using the backpropagation algorithm in mineral-potential mapping isthe paucity of deposit relative to barren training patterns. To overcome this problem, random noise is added to the original training patterns in order to create additional synthetic deposit training data. Experiments on the effect of the number of deposits available for training in the Kalgoorlie Terrane orogenic gold province show that both the classification performance of a trained network and the quality of the resultant prospectivity map increasesignificantly with increased numbers of deposit patterns. Experiments are conducted to determine the optimum amount of noise using both uniform and normally distributed random noise. Through the addition of noise to the original deposit training data, the number of deposit training patterns is increased from approximately 50 to 1000. The percentage of correct classifications significantly improves for the independent test set as well as for deposit patterns in the test set. For example, using ±40% uniform random noise, the test-set classification performance increases from 67.9% and 68.0% to 72.8% and 77.1% (for test-set overall and test-set deposit patterns, respectively). Indices for the quality of the resultant prospectivity map, (i.e. D/A, D × (D/A), where D is the percentage of deposits and A is the percentage of the total area for the highest prospectivity map-class, and area under an ROC curve) also increase from 8.2, 105, 0.79 to 17.9, 226, 0.87, respectively. Increasing the size of the training-stop data set results in a further increase in classification performance to 73.5%, 77.4%, 14.7, 296, 0.87 for test-set overall and test-set deposit patterns, D/A, D × (D/A), and area under the ROC curve, respectively.  相似文献   

Use of Fuzzy Membership Input Layers to Combine Subjective Geological Knowledge and Empirical Data in a Neural Network Method for Mineral-Potential Mapping   总被引：3，自引：0，他引：3  

Brown  Warick  Groves  David  Gedeon  Tamás 《Natural Resources Research》2003,12(3):183-200

Use of GIS layers, in which the cell values represent fuzzy membership variables, is an effective method of combining subjective geological knowledge with empirical data in a neural network approach to mineral-prospectivity mapping. In this study, multilayer perceptron (MLP), neural networks are used to combine up to 17 regional exploration variables to predict the potential for orogenic gold deposits in the form of prospectivity maps in the Archean Kalgoorlie Terrane of Western Australia. Two types of fuzzy membership layers are used. In the first type of layer, the statistical relationships between known gold deposits and variables in the GIS thematic layer are used to determine fuzzy membership values. For example, GIS layers depicting solid geology and rock-type combinations of categorical data at the nearest lithological boundary for each cell are converted to fuzzy membership layers representing favorable lithologies and favorable lithological boundaries, respectively. This type of fuzzy-membership input is a useful alternative to the 1-of-N coding used for categorical inputs, particularly if there are a large number of classes. Rheological contrast at lithological boundaries is modeled using a second type of fuzzy membership layer, in which the assignment of fuzzy membership value, although based on geological field data, is subjective. The methods used here could be applied to a large range of subjective data (e.g., favorability of tectonic environment, host stratigraphy, or reactivation along major faults) currently used in regional exploration programs, but which normally would not be included as inputs in an empirical neural network approach.  相似文献   

Distribution and effectiveness of nivation in Mediterranean mountains: Peñalara (Spain)     

David Palacios  Nuria de Andrs  Enrique Luengo 《Geomorphology》2003,54(3-4):157-178

Snow accumulation is responsible for geomorphic and biogeographic processes taking place in the southern sector of the Peñalara massif in central Spain (40°51′N, 3°57′W; max. altitude 2428 m at Pico de Peñalara). This work compares the intensity of nivation on the eastern slope, leeward of the prevailing westerly winds and heavily eroded by glacial activity during the Pleistocene, to that of the western slope on the windward side, unaffected by glacial erosion and completely covered by a thick weathering mantle. On the eastern slope, nivation is effective only where the weathering mantle is exposed or on morainic formations. It does not occur on the landforms derived from glacial erosion. In contrast, the western side shows almost no evidence of snow action except where catastrophic mass movements have altered the regularity of the slope. During the post-glacial epoch, nivation cirques formed in the scars left by mass wasting. In the last 30 years, spring temperatures have increased and this activity has diminished. The pattern of evolution observed at Peñalara can be extrapolated to other Mediterranean mountains with similar characteristics such as marginal glacial activity during the Pleistocene, unconsolidated formations on the summits caused by chemical weathering, and dry, hot summers that can increase the effectiveness of nivation.  相似文献   

Permeability-porosity Relationships in Rocks Subjected to Various Evolution Processes     

Y. Bernabé  U. Mok  B. Evans 《Pure and Applied Geophysics》2003,160(5-6):937-960

— It is well known that there is no “universal” permeability-porosity relationship valid in all porous media. However, the evolution of permeability and porosity in rocks can be constrained provided that the processes changing the pore space are known. In this paper, we review observations of the relationship between permeability and porosity during rock evolution and interpret them in terms of creation/destruction of effectively and non-effectively conducting pore space. We focus on laboratory processes, namely, plastic compaction of aggregates, elastic-brittle deformation of granular rocks, dilatant and thermal microcracking of dense rocks, chemically driven processes, as a way to approach naturally occurring geological processes. In particular, the chemically driven processes and their corresponding evolution permeability-porosity relationships are discussed in relation to sedimentary rocks diagenesis.  相似文献   

Analytical Model for Permeability Evolution in Microcracking Rock     

G. D. H. Simpson  Y. Guéguen  F. Schneider 《Pure and Applied Geophysics》2003,160(5-6):999-1008

— Analytical expressions to predict the enhancement of permeability due to stress-induced microcracking in initially low porosity rock are presented. A fracture mechanical microcrack model is employed to derive integrated effective hydraulic variables as a function of stress, which are then used to calculate the evolution of permeability using the statistically-based Dienes model. The model enables determination of permeability enhancement as a function of two loading parameters and three material parameters. Results are in reasonable agreement with experimental measurements and indicate that appreciable increases in permeability can be anticipated during brittle failure. The analytical nature of the model makes it easily incorporatable into numerical models that require quantification of the permeability evolution as a function of stress, for which there is currently no law.  相似文献   

Extracting flow structures from tracer data     

Judit Schneider  Tamás Tél 《Ocean Dynamics》2003,53(2):64-72

A method of visualizing structures in closed chaotic flows out of homogenous particle distributions is presented in the example of models of a meandering jet. To this end, the system will be leaked or opened up by defining a region of the flow, so that a particle is considered to be escaped if it leaves this region. By applying this method to an ensemble of nonescaped tracers, we are able to characterize mixing processes by visualizing the converging and stretching filamentations (stable and unstable manifolds) in the flow without using additional mathematical tools. The possibility of applying the algorithm to analyze buoy data, and a comparison with the finite time manifolds are discussed.  相似文献   

Failure Mode and Spatial Distribution of Damage in Rothbach Sandstone in the Brittle-ductile Transition   总被引：4，自引：0，他引：4  

P. Bésuelle  P. Baud  T. Wong 《Pure and Applied Geophysics》2003,160(5-6):851-868

— To elucidate the spatial complexity of damage and evolution of localized failure in the transitional regime from brittle faulting to cataclastic ductile flow in a porous sandstone, we performed a series of triaxial compression experiments on Rothbach sandstone (20% porosity). Quantitative microstructural analysis and X-ray computed tomography (CT) imaging were conducted on deformed samples. Localized failure was observed in samples at effective pressures ranging from 5 MPa to 130 MPa. In the brittle faulting regime, dilating shear bands were observed. The CT images and stereological measurements reveal the geometric complexity and spatial heterogeneity of damage in the failed samples. In the transitional regime (at effective pressures between 45 MPa and 130 MPa), compacting shear bands at high angles and compaction bands perpendicular to the maximum compression direction were observed. The laboratory results suggest that these complex localized features can be pervasive in sandstone formations, not just limited to the very porous aeolian sandstone in which they were first documented. The microstructural observations are in qualitative agreement with theoretical predictions of bifurcation analyses, except for the occurrence of compaction bands in the sample deformed at effective pressure of 130 MPa. The bifurcation analysis with the constitutive model used in this paper is nonadequate to predict compaction band formation, may be due to the neglect of bedding anisotropy of the rock and multiple yield mechanisms in the constitutive model.  相似文献   

Dinosaur tracks from the Cretaceous of South Korea: Distribution, occurrences and paleobiological significance   总被引：2，自引：0，他引：2  

Min Huh  Koo Geun Hwang  In Sung Paik  Chull Hwan Chung  Bo Seong Kim 《Island Arc》2003,12(2):132-144

Abstract   Abundant dinosaur fossils including dinosaur footprints, eggs and nests, teeth and bones have been found from the Cretaceous non-marine deposits of Korea. Among them, dinosaur tracks are the most distinctive, and some track sites are among the most famous in the world. Until now, 27 dinosaur track localities have been discovered from the Cretaceous strata in the Gyeongsang Basin and several small basins. Ornithopod tracks are most abundant at most Korean track sites, and most of them are identified as Caririchnium ; that is, large ornithopod footprints with wide hoof impressions. Most theropod tracks are found in Neungju Basin and they consist of various types of small or medium-sized bird-like footprints, and other large footprints. Sauropod tracks are also abundant in the Gyeongsang Basin. The sauropod tracks vary in size, shape, and pattern of trackway, and suggest that diverse sauropods existed in this area. These diverse tracks in South Korea suggest that various dinosaurs flourished at the margins of lakes distributed in the southern part of the Korean Peninsula during the Cretaceous.  相似文献   

Coupled inverse modelling of groundwater flow and mass transport and the worth of concentration data   总被引：1，自引：0，他引：1  

Harrie-Jan Hendricks Franssen  Jaime Gómez-Hernández  Andrés Sahuquillo 《Journal of Hydrology》2003,281(4):281-295

This paper presents the extension of the self-calibrating method to the coupled inverse modelling of groundwater flow and mass transport. The method generates equally likely solutions to the inverse problem that display the variability as observed in the field and are not affected by a linearisation of the state equations. Conditioning to the state variables is measured by an objective function including, among others, the mismatch between the simulated and measured concentrations. Conditioning is achieved by minimising the objective function by gradient-based methods. The gradient contains the partial derivatives of the objective function with respect to: log conductivities, log storativities, prescribed heads at boundaries, retardation coefficients and mass sources. The derivatives of the objective function with respect to log conductivity are the most cumbersome and need the most CPU-time to be evaluated. For this reason, to compute this derivative only advective transport is considered. The gradient is calculated by the adjoint-state method. The method is demonstrated in a controlled, synthetic study, in which the worth of concentration data is analysed. It is shown that concentration data are essential to improve transport predictions and also help to improve aquifer characterisation and flow predictions, especially in the upstream part of the aquifer, even in the case that a considerable amount of other experimental data like conductivities and heads are available. Besides, conditioning to concentration data reduces the ensemble variances of estimated transmissivity, hydraulic head and concentration.  相似文献   

La vallée d'Ys : un paléoréseau hydrographique immergé en baie de Douarnenez (Finistère,France)     

Gwénaël Jouet  Claude Augris  Bernard Hallegouët  Pascal Le Roy  Joël Rolet 《Comptes Rendus Geoscience》2003,335(5):487-494

Interpretation of the recent high-resolution survey, CANADOU 2000, in the Bay of Douarnenez (Finistère, France) allowed us to restore the morphology of the substratum and the sedimentary filling of the bay. The Brioverian and Palaeozoic substratum reveals a well-defined network of incised valleys as results of successive emergence stages of the Bay during the Quaternary. Valleys join in a westward-widened mean valley, called Ys Valley. The present-day sedimentary fill of the bay of Douarnenez appears mainly controlled by the Holocene rise and the consecutive highstand. It comprises fluvial and estuarine deposits filling up incised valleys and marine sedimentation extending out of the incised valleys. To cite this article: G. Jouet et al., C. R. Geoscience 335 (2003).To cite this article: G. Jouet et al., C. R. Geoscience 335 (2003).  相似文献