Home // ADVCOMP 2019, The Thirteenth International Conference on Advanced Engineering Computing and Applications in Sciences // View article

Irrigation Reservoir Modeling in Catchments without Measurements of Stream Flow

Authors:
Milan Cisty
Veronika Soldanova
Barbora Povazanova

Keywords: stream flow; ungaged catchment; hydrologic modeling; LASSO; machine learning

Abstract:
This paper considers the determination of time series of river flows in catchments without direct monitoring of this variable. This paper proposes a method for the acquisition of monthly data, which is useful for various purposes. Different parameters of various water management structures can be determined based on information from such data series, such as irrigation reservoir volumes or water demand for irrigation. While identifying unknown stream flows required for such a calculation, authors suppose that historical climatic data for the given area and flows in nearby river catchments are available. This article includes a description of the method of selecting river catchments such that their measured flows can be used in the calculation of an unknown flow of a different stream. This study compares hydrological modeling, linear regression with regularization, and machine learning methods (support vector machines, random forest). Statistical indicators evaluate the calculated flows with the result that the most suitable approach is the support vector machines method using a linear kernel and LASSO regularisation.

Pages: 12 to 18

Copyright: Copyright (c) IARIA, 2019

Publication date: September 22, 2019

Published in: conference

ISSN: 2308-4499

ISBN: 978-1-61208-737-5

Location: Porto, Portugal

Dates: from September 22, 2019 to September 26, 2019