К ПРОБЛЕМЕ ОЦЕНКИ ТРАНСПОРТА НАНОСОВ

Abstract

The paper discusses the problem of high errors in the design ratios proposed for assessing sediment transport in natural watercourses. The question is why numerous empirical design ratios obtained and successfully used for some watercourses can give an error of 1000% when applied to other rivers. Calculation formulas worked out on hydraulic trays and channels are found to be inappropriate for specific rivers and natural channels. The problem is caused by the complexity of the natural watercourses geometry and the heterogeneity of the bottom sediment composition. Two approaches that are currently used in assessing the transporting capacity of river flows are studied in the paper. They were laid down in river hydraulics as early as the second half of the 18th century in the works by A. Chesi and P. Dubois. The conditions and possibilities of using both methods are considered and theoretically proven in the paper. The approach developed by P. Dubois is distinguished by a detailed study of the influence of individual factors based on numerous experimental models, the construction of fairly rigorous physical models on this basis, the complication of design ratios, the inclusion of new additional parameters. A. Chesi offered the construction of a maximally simplified initial physical model, with the calibration parameters established for specific conditions on the basis of field observations or qualitative estimates, and therefore being less accurate but more stable. It is shown that despite the scientific attractiveness of the approach to constructing and using more complex calculation models containing new additional parameters, the inclusion of additional parameters entails the inclusion of additional errors associated with the estimation of these parameters. The effectiveness of both approaches is proved; the application of one or the other approach is determined by the conditions and nature of the tasks to be solved, as well as the volume and accuracy of the initial data.