Abstract: Today’s cities face the challenge of climate change adaptation worldwide. In this context, prevention of damage caused by flash floods plays an important role. This requires a cooperative pluvial flood risk management approach, which includes planning, technical, and administrative measures and involves preliminary flood risk analyses. This article outlines the main components of this risk management approach, which has proven its effectiveness in Europe. The recommendations formulated for this purpose are applicable or adaptable to regions with other constraints, such as China, for example.

Abstract: Machine learning today becomes more and more effective instrument to solve many particular problems, where there are difficulties to apply well known and described math model. In other words – it is a great tool to describe non-linear phenomena. We tried to use this technique to improve existing process of stratigraphy, and reduce costs on site by applying computer leaded predictions on the basis of existing on-field collected data. Article describes usage of machine learning algorithms for stratigraphy boundaries classification based on geophysics logging data for uranium deposit in Kazakhstan. Correct marking of stratigraphy from geophysics logging data is complex non-linear task. To solve this task we applied several algorithms of machine learning: random forest, logistic regression, gradient boosting, k nearest neighbour and XGBoost.

Abstract: Concentrations of uranium (U) \textgreater30 µg/L in groundwater are relatively uncommon in drinking water in the United States but can be of concern in those areas where complex interactions of aquifer materials and anthropogenic alterations of the natural flow regime mobilize U. High concentrations (\textgreater30 µg/L) of U in the southeastern San Joaquin Valley, California, USA, have been detected in 24 percent of 257 domestic, irrigation, and public-supply wells sampled across an approximately 110,000 km2 area. In this study we evaluated mechanisms for mobilization of U in the San Joaquin Valley proposed in previous studies, confirming mobilization by HCO3 and refuting mobilization by NO3 and we refined our understanding of the geologic sources of U to the scale of individual alluvial fans. The location of high concentrations depends on the interactions of geological U sources from fluvial fans that originate in the Sierra Nevada to the east and seepage of irrigation water that contains high concentrations of HCO3 that leaches U from the sediments. In addition, interactions with PO4 from fertilized irrigated fields may sequester U in the aquifer. Principal component analysis of the data demonstrates that HCO3 and ions associated with high total dissolved solids in the aquifer and the percentage of agriculture near the well sampled are associated with high U concentrations. Nitrate concentrations do not appear to control release of U to the aquifer. Age dating of the groundwater and generally increasing U concentrations of the past 25 years in resampled wells where irrigation is prevalent suggests that high U concentrations are associated with younger water, indicating that irrigation of fields over the past 100 years has significantly contributed to increasing concentrations and mobilizing U. In some places, the groundwater is supersaturated with uranyl-containing minerals, as would be expected in roll front deposits. In general, the interaction of natural geological sources high in U, the anthropogenically driven addition of HCO3 and possibly phosphate fertilizer, control the location and concentration of U in each individual fluvial fan, but the addition of nitrate in fertilizer does not appear control the location of high U. These geochemical interactions are complex but can be used to determine controls on anomalously high U in alluvial aquifers.