Automatic revision of rules used to guide the generalisation process in systems based on a trial and error strategy

Automating the generalisation process, a major issue for national mapping agencies, is extremely complex. Several works have proposed to deal with this complexity using a trial and error strategy. The performance of systems based on such a strategy is directly dependent on the quality of the control knowledge (i.e. heuristics) used to guide the trials. Unfortunately, most of the time, the definition and updation of knowledge is a fastidious task. In this context, automatic knowledge revision can not only improve the performance of the generalisation, but also allow it to automatically adapt to various usages and evolve when new elements are introduced. In this article, an offline knowledge revision approach is proposed, based on a logging of the system and on the analysis of outcoming logs. This approach is dedicated to the revision of control knowledge expressed by production rules. We have implemented and tested this approach for the automated generalisation of groups of buildings within a generalisation model called AGENT, from initial data that reference a scale of approximately 1:15,000 compared with the target map's scale of 1:50,000. The results show that our approach improves the quality of the control knowledge and thus the performance of the system. Moreover, the approach proposed is generic and can be applied to other systems based on a trial and error strategy, dedicated to generalisation or not.     

Coarse carbonate breccias as a result of water-wave cyclic loading (uppermost Jurassic – South-East Basin, France)

In the uppermost Jurassic of the central part of the South-East Basin of France, an association of lime mudstone beds, calcarenite beds and coarse carbonate breccia bodies form an informal stratigraphical unit called the 'Barre Tithonique'. In the 'Barre Tithonique', gradual transitions from lime mudstone or calcarenite to breccia show different stages of deformation leading to progressive brecciation of the original lithologies. The study of the breccia facies, and the observed gradual transitions as a whole, document a new early diagenetic process in carbonate environments, resulting from water-wave and seabed interaction. Water-wave induced brecciation and its abundance in the 'Barre Tithonique' indicate that sea–seabed interaction was significant. Comparison with modern studies of the mechanics of wave–seabed interaction suggests that water depth was less than 200 m. It is demonstrated that sedimentary features such as channel-like structures, previously interpreted as being the result of erosion and deposition of mud-flows, were in fact produced by wave-induced, in situ reworking of lime mud, without any significant unidirectional flow or gravity induced displacement.     

Selenium speciation and partitioning within Burkholderia cepacia biofilms formed on α-Al2O3 surfaces

The distribution and speciation of Se within aerobic Burkholderia cepacia biofilms formed on α-Al₂O₃ (1-102) surfaces have been examined using grazing-angle X-ray spectroscopic techniques. We present quantitative information on the partitioning of 10⁻⁶ M to 10⁻³ M selenate and selenite between the biofilms and underlying alumina surfaces derived from long-period X-ray standing wave (XSW) data. Changes in the Se partitioning behavior over time are correlated with microbially induced reduction of Se(VI) and Se(IV) to Se(0), as observed from X-ray absorption near edge structure (XANES) spectroscopy.Selenite preferentially binds to the alumina surfaces, particularly at low [Se], and is increasingly partitioned into the biofilms at higher [Se]. When B. cepacia is metabolically active, B. cepacia rapidly reduces a fraction of the SeO₃²⁻ to red elemental Se(0). In contrast, selenate is preferentially partitioned into the B. cepacia biofilms at all [Se] tested due to a lower affinity for binding to the alumina surface. Rapid reduction of SeO₄²⁻ by B. cepacia to Se(IV) and Se(0) subsequently results in a vertical segregation of Se species at the B. cepacia/α-Al₂O₃ interface. Elemental Se(0) accumulates within the biofilm with Se(VI), whereas Se(IV) intermediates preferentially sorb to the alumina surface.B. cepacia/α-Al₂O₃ samples incubated with SeO₄²⁻ and SeO₃²⁻ when the bacteria were metabolically active result in a significant reduction in the mobility of Se vs. X-ray treated biofilms. Remobilization experiments show that a large fraction of the insoluble Se(0) produced within the biofilm is retained during exchange with Se-free solutions. In addition, Se(IV) intermediates generated during Se(VI) reduction are preferentially bound to the alumina surface and do not fully desorb. In contrast, Se(VI) is rapidly and extensively remobilized.     

Earthquake mechanisms of the Adriatic Sea and Western Greece: implications for the oceanic subduction-continental collision transition

We present 21 focal solutions (magnitude > 5.5) reliably computed by body-wave modelling for the western Hellenic arc from Yugoslavia to the southern Peloponnese. Mechanisms located within the Aegean show normal faulting, the T-axis trending N-S in the centre and parallel to the active boundary in the external part. Mechanisms associated with the Keffalinia fault are consistent with dextral strike-slip motion. Reverse mechanisms located along the active boundary are remarkably consistent and do not depend on the nature of the active boundary (continental collision or oceanic subduction). The consistency in azimuth of the slip vectors and of the GPS velocity relative to Africa, all along the active boundary, suggests that the deformation is related to the same motion. The discrepancy between seismic-energy release and the amount of shortening confirms that the continental collision is achieved by seismic slip on faults but the oceanic subduction is partially aseismic. The northward decrease in velocity between continental collision and oceanic subduction suggests the continental collision to be a recent evolution of the active subduction.     

Enrichment of redox-sensitive trace metals (U,V, Mo,As) associated with the late Hauterivian Faraoni oceanic anoxic event

The Faraoni Level is a short-lived oxygen-deficient event that took place during the latest Hauterivian. In order to improve our understanding of the palaeoenvironmental conditions that occurred during this event, we have analysed the contents of several redox-sensitive trace elements (U, V, Mo, As, Co, Cd, Cu, Zn, Ni, Pb, Cr) from bulk limestone samples of late Hauterivian–early Barremian age from three reference sections. U, V, Mo and As show consistent and significant enrichments during the Faraoni event whereas the other redox-sensitive trace elements analysed here are not systematically enriched. In order to explain this discrepant behaviour, we propose that the Faraoni Level was deposited during a period of anoxic conditions near the sediment–water interface. The distinctive peaks in U, V, Mo and As contents are traceable throughout the three studied sections and represent a good correlation tool which helps to identify the Faraoni Level and its equivalents in the western Tethyan realm and outside of the Tethys. For example, a peak in U contents in upper Hauterivian sediments of the northwestern Pacific realm (ODP leg 185, site 1149) may well be an expression of the Faraoni event in this particular basin.     

Ancient subsurface structure beneath crater Clavius:constraint by recent high-precision gravity and topography data

With the increasing precision of the GRAIL gravity field models and topography from LOLA, it is possible to investigate the substructure beneath crater Clavius. An admittance between gravity and topography data is commonly used to estimate selenophysical parameters, including load ratio, crustal thickness and density, and elastic thickness. Not only a surface load, but also a subsurface load is considered in estimation. The algorithm of particle swarm optimization(PSO) with a swarm size of 400 is employed as well.Results indicate that the observed admittance is best-fitted by the modeled admittance based on a spherical shell model, which was proved to be unsatisfactory in the previous study. The best-fitted load ratio f is around-0.194. Such a small load ratio conforms to the direct proportion between the nearly uncompensated topography and its corresponding negative gravity anomaly. It also indicates that a surface load dominates all the loads. Constrained within 2σSTD, a small crustal thickness(～30 km) and a crustal density of ～2587 kg m-3are found, quite close to the results from previous GRAIL research. Considering the well constrained crustal thickness and density, the best-fitted elastic thickness(～7 km) is rational. This result is slightly smaller than the previous study(～12 km). Such difference can be attributed to the difference in crustal density used and the precision of gravity and topography data. Considering that the small difference between the modeled gravity anomaly and observations is quite small, a parameter inversed here could be an indicator of the subsurface structure beneath Clavius.