Forecasting ozone concentrations in the east of Croatia using nonparametric Neural Network Models     

ELVIRA KOVAČ-ANDRIĆ  ALAA SHETA  HOSSAM FARIS  MARTINA ŠRAJER GAJDOŠIK 《Journal of Earth System Science》2016,125(5):997-1006

Ozone is one of the most significant secondary pollutants with numerous negative effects on human health and environment including plants and vegetation. Therefore, more effort is made recently by governments and associations to predict ozone concentrations which could help in establishing better plans and regulation for environment protection. In this study, we use two Artificial Neural Network based approaches (MPL and RBF) to develop, for the first time, accurate ozone prediction models, one for urban and another one for rural area in the eastern part of Croatia. The evaluation of actual against the predicted ozone concentrations revealed that MLP and RBF models are very competitive for the training and testing data in the case of Kopa?ki Rit area whereas in the case of Osijek city, MLP shows better evaluation results with 9% improvement in the correlation coefficient. Furthermore, subsequent feature selection process has improved the prediction power of RBF network.  相似文献   

Carbonate stromatolites from a Messinian hypersaline setting in the Caltanissetta Basin, Sicily: petrographic evidence of microbial activity and related stable isotope and rare earth element signatures     

ELVIRA OLIVERI  RODOLFO NERI  ADRIANA BELLANCA  ROBERT RIDING†  Associate Editor: Tracy Frank 《Sedimentology》2010,57(1):142-161

Lower Messinian stromatolites of the Calcare di Base Formation at Sutera in Sicily record periods of low sea‐level, strong evaporation and elevated salinity, thought to be associated with the onset of the Messinian Salinity Crisis. Overlying aragonitic limestones were precipitated in normal to slightly evaporative conditions, occasionally influenced by an influx of meteoric water. Evidence of bacterial involvement in carbonate formation is recorded in three dolomite‐rich stromatolite beds in the lower portion of the section that contain low domes with irregular crinkly millimetre‐scale lamination and small fenestrae. The dominant microfabrics are: (i) peloidal and clotted dolomicrite with calcite‐filled fenestrae; (ii) dolomicrite with bacterium‐like filaments and pores partially filled by calcite or black amorphous matter; and (iii) micrite in which fenestrae alternate with dark thin wispy micrite. The filaments resemble Beggiatoa‐like sulphur bacteria. Under scanning electron microscopy, the filaments consist of spherical aggregates of dolomite, interpreted to result from calcification of bacterial microcolonies. The dolomite crystals are commonly arranged as rounded grains that appear to be incorporated or absorbed into developing crystal faces. Biofilm‐like remains occur in voids between the filaments. The dolomite consistently shows negative δ¹³C values (down to ?11·3‰) and very positive δ¹⁸O (mean value 7·9‰) that suggest formation as primary precipitate with a substantial contribution of organic CO₂. Very negative δ¹³C values (down to ?31·6‰) of early diagenetic calcite associated with the dolomite suggest contribution of CO₂ originating by anaerobic methane oxidation. The shale‐normalized rare earth element patterns of Sutera stromatolites show features similar to those in present‐day microbial mats with enrichment in light rare earth elements, and M‐type tetrad effects (enrichment around Pr coupled to a decline around Nd and a peak around Sm and Eu). Taken together, the petrography and geochemistry of the Sutera stromatolites provide diverse and compelling evidence for microbial influence on carbonate precipitation.  相似文献