Modelling evaporation using an artificial neural network algorithm

This paper investigates the prediction of Class A pan evaporation using the artificial neural network (ANN) technique. The ANN back propagation algorithm has been evaluated for its applicability for predicting evaporation from minimum climatic data. Four combinations of input data were considered and the resulting values of evaporation were analysed and compared with those of existing models. The results from this study suggest that the neural computing technique could be employed successfully in modelling the evaporation process from the available climatic data set. However, an analysis of the residuals from the ANN models developed revealed that the models showed significant error in predictions during the validation, implying loss of generalization properties of ANN models unless trained carefully. The study indicated that evaporation values could be reasonably estimated using temperature data only through the ANN technique. This would be of much use in instances where data availability is limited. Copyright © 2002 John Wiley & Sons, Ltd.     

An automated strategy for calculation of phase diagram sections and retrieval of rock properties as a function of physical conditions

We formulate an algorithm for the calculation of stable phase relations of a system with constrained bulk composition as a function of its environmental variables. The basis of this algorithm is the approximate representation of the free energy composition surfaces of solution phases by inscribed polyhedra. This representation leads to discretization of high variance phase fields into a continuous mesh of smaller polygonal fields within which the composition and physical properties of the phases are uniquely determined. The resulting phase diagram sections are useful for understanding the phase relations of complex metamorphic systems and for applications in which it is necessary to establish the variations in rock properties such as density, seismic velocities and volatile‐content through a metamorphic cycle. The algorithm has been implemented within a computer program that is general with respect to both the choice of variables and the number of components and phases possible in a system, and is independent of the structure of the equations of state used to describe the phases of the system.     

Structure sismique du socle paléozoı̈que du bassin des Doukkala,Môle côtier,Maroc occidental. Indication en faveur de l'existence d'une phase éo-varisqueSeismic structure of the Doukkala basin,Palaeozoic basement,western Morocco: a hint for an Eovariscan fold-and-thrust belt

Seismic profiles and well data from the Doukkala basin unravel the structure of the Palaeozoic basement and suggest that this coastal zone of western Morocco was affected by a compressive phase during the Frasnian. This resulted in the formation of upright, plurikilometric folds associated with reverse faults (North Doukkala), and of asymmetrical folds associated with mostly west verging ramps (South Doukkala). Folding involved all pre-Upper Frasnian formations and caused partial or total hiatus of Upper Frasnian–Strunian strata. This event can be correlated with the orogenic phase reported from more internal domains of the Morocco Hercynian belt, where it is referred to as the ‘Bretonne’ or ‘Eovariscan’ phase. To cite this article: H. Echarfaoui et al., C. R. Geoscience 334 (2002) 13–20     

ARTIFICIAL NEURAL NETWORK CORRECTION OF ROUGH ERRORS OF OBSERVATIONS*

In the context of tower measured radiation datasets.following the correction principle meeting a diagnostic equation in data quality control and in terms of a technique for model construction on data and ANN (artificial neural network) retrieval for BP correction of radiation measurements with rough errors available,a BP model is presented.Evidence suggests that the developed model works well and is superior to a convenient multivariate linear regression model,indicating its wide applications.     

Bedrock structure in Adapazari, Turkey—a possible cause of severe damage by the 1999 Kociaeli earthquake

The 1999 Kocaeli earthquake brought serious damage to downtown of Adapazari. To study why strong motions were generated at the town, a bedrock structure was investigated on the basis of Bouguer gravity anomaly, and SPAC and H/V analyses of microseisms. It was revealed that, the basin consists of three narrow depressions of bedrock with very steep edges, extending in E–W or NE–SW directions along the North Anatolia faults, and the depth to bedrock reaches 1000 m or more. Downtown of Adapazari is located 1–2 km apart from the basin-edge. It is considered that, the specific configuration of bedrock amplifies ground motions at the downtown area by focusing of seismic waves and/or interference between incident S-waves and surface-waves secondarily generated at the basin-edge. Studying 3D bedrock structure is an urgent issue for microzoning an urban area in a sedimentary basin.