Sediment is a universal issue that is generated in the river catchment and affects the river flow, reservoir capacity, hydropower generation and dam structure. This paper aims to present the result of experimentation in sediment load estimation using various machine learning algorithms as a powerful AI approach. The data was collected from eight locations in upstream area of Ringlet reservoir catchment. The input variables are discharge and suspended solid. It was found that there is strong correlation between sediment and suspended solid with correlation coefficient of R = 0.9 . The developed ML model successfully estimated the sediment load with competitive results from ANN, Decision Tree, AdaBoost and SVM. The best result was produced by SVM (v-SVM version) where very low RMSE was generated for both training and testing dataset despite its more complicated hyperparameters setup. The results also show a promising application of machine learning for future prediction in hydro-informatic systems.
W artykule omówiono główne problemy z podejściem do zagospodarowania den dolin rzecznych, w tym dolnej Wisły. Wzrost dynamiki odpływu wód, warunkowanego zmianami globalnymi klimatu, a także coraz większą ingerencją człowieka (rolnictwo, urbanizacja, przemysł), skutkuje tym, że mamy do czynienia ze wzrostem ilości i wysokości wezbrań z jednoczesnym potęgowaniem się susz. Z kolei podstawową zasadą prawidłowo funkcjonującego systemu rzecznego jest to, by umożliwić swobodny odpływ nadmiaru wód, zwłaszcza podczas wezbrań. I tak korytem następuje odpływ wód najczęściej trwających, tj. od niżówek poprzez stany (pochodne przepływy) średnie do tzw. stanów pełnokorytowych, zaś wody wezbraniowe są odprowadzane, specjalnie wykształconym przez rzekę, poziomem (równina) zalewowym. Tymczasem powierzchnia międzywala jest traktowana jako korytarz ekologiczny, obszar specjalnej ochrony ptaków (OSO) – Dyrektywa Ptasia, oraz specjalne obszary ochrony siedlisk (SOO) – Dyrektywa Siedliskowa. Utrzymanie roślinności na międzywalu powoduje zagrożenie w momencie występowanie zwiększonych przepływów wody, przyczyniając się do wezbrań o charakterze powodzi. W ocenie autorów złą praktyką jest m.in. zachowanie drzewostanu i zakrzewienia na powierzchni równiny zalewowej międzywala, które dzięki „zaszłościom” historycznym i gospodarczym (nie prowadzono wyrębu i wycinki) dziś uniemożliwiają prawidłowe funkcjonowanie tej makroformy korytowej (odcinki zatorogenne, utrudniające spływ wód wezbraniowych).
EN
The article discusses main problems related to the approach to development of river valley bottoms, including that of lower Vistula. Increased water outflow dynamics, determined by the global climate change, as well as growing human influence (agriculture, urbanization, industry) lead to escalation of the number and height of floods with simultaneous accumulation of droughts. On the other hand, the basic principle of a correctly functioning river system is to enable free outflow of the excess of waters, especially during floods. The most frequent water stages, i.e. from low water, thorough average water (derived flows), to the so-called bankfull stages, flow downstream the riverbed, whereas high water stages flow into the flood plain shaped by the river. The surface of the inter-embankment zone is treated as an ecological corridor, the area of special protection of birds (SPA) – the Birds Directive, and the area of special protection of habitats (SAC) – the Habitats Directive. Conservation of vegetation in the inter-embankment zone is dangerous at the moment of increased water flows, contributing to the occurrence of high water stages with flood characteristics. The authors consider it to be a wrong practice to preserve wood stands and bushes in the inter-embankment flood plain which, due to historical and economic reasons (no felling nor logging were carried out), prevent today the correct functioning of this riverbed macro-form (sections generating jams, hindering the outflow of flood waters).
W artykule przedstawiono koncepcję krótkoterminowej prognozy hydrologicznej w oparciu o modele typu trójwarstwowe sztuczne sieci neuronowe (SSN). Przedstawiono również możliwość łączenia modeli typu SSN z modelami regresji liniowej. Zwrócono uwagę na wady i zalety prezentowanego podejścia i możliwości jego praktycznego zastosowania w częściowo kontrolowanym systemie rzecznym. Zaprezentowaną koncepcję poparto przykładami symulacji prognozy stanów wysokich w przykładowej zlewni różnicowej rzeki Raby.
EN
In the paper the approach of short term hydrologic forecasting was presented using artificial neural networks (SSN). The possibilities of the fusion of SSN models with models of linear regression were also presented. Advantages and disadvantages of the presented method were underlined in the perspective of its practical implementation within partly monitored river system. The forecasting approach was supported by results of numerical simulation of high water states in the differential catchment of the Raba river.
4
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The cosmopolitan freshwater oligochaete Aulodrilus pigueti Kowalewski, 1914 (Naididae: Tubificinae) was recorded for the first time in Poland. The species was found at a few groyne field sites and in the main flow of the middle and lower course of the Oder River. Forty two other species of oligochaetes were found in the analyzed material, including rarely found in Poland Aulodrilus japonicus and Bothrioneurum vejdovskyanum, as well as alien species: Branchiura sowerbyi and Quistadrilus multisetosus.
5
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Longitudinal organisation of macroinvertebrate fauna along river is one of the most important problem commonly used to explain the functioning of flowing water ecosystems. The river system can be treated as a mosaic of landscapes patches and riverine macroinvertebrates’ community structure is a function of longitudinal changes in its key abiotic patterns. The aims of the study was to analyse the taxonomic structure of molluscs in the river-lake system, to compare river and lake malacofauna and to determine the factors responsible for its diversity in lakes and river stretches. The study based on comprehensive analysis of malakofauna sampled at 10 sites in the Krutynia River and in all 19 lakes it flows through were performed in the years 2008–1011. River Krutynia is one of the most important rivers in Masurian Lakeland (north-eastern Poland) with length of 100 km and mean annual discharge of 10.6 m3 h-1. It forms a characteristic river-lake system typical for the lakeland landscape in Central Europe. The density and taxonomic composition of molluscs were found as strongly dissimilar in a local scale – between closely located lakes and between particular parts of flowing waters, divided by the lakes. Multivariate methods were used to demonstrate a clear dissimilarity of lake and river malakofauna and to show that the mean content of phosphorus, nitrogen and organic matter in bottom sediments were correlated with each other in lakes but not in the river. The most important species differentiating river sites into larger groups with respect to the similarity was Theodoxus fluviatilis while Stagnicola corvus and Anisus vortex were such species differentiating lakes. The numbers and percentages of Dreissena polymorpha and Unionidae were negatively correlated with nutrients in river sediments. There was a strong positive relationship between nutrients’ concentrations in sediments and the percentages of Viviparus, Obtained results of multiple regression indicate a strong effect of nutrient and organic matter concentrations in the sediments and the distance from the site to the lake on the domination structure of molluscs.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.