Inverse Problem for Looped River Networks – Lower oder River Case Study

Abstract

Identifi cation of coeffi cients determining fl ow resistance, in particular Manning's roughness coeffi cients, is one of the possible inverse problems of mathematical modeling of fl ow distribution in looped river networks. The paper presents the solution of this problem for the lower Oder River network consisting of 78 branches connected by 62 nodes. Using results of six sets of fl ow measurements at particular network branches it was demonstrated that the application of iterative algorithm for roughness coeffi cients identifi cation on the basis of the sensitivity-equation method leads to the explicit solution for all network branches, independent from initial values of identifi ed coeffi cients.

Improved method for identifying Manning’s roughness coefficients in plain looped river network area

“…However, the use of spatially varying n values could significantly increase the number of possible roughness sets and thereby incur high calibration workloads (Ayvaz, 2013). In addition, any adjustment of n values may lead to unpredictable water level or flow behaviors in a complicated river network (Kurnatowski, 2013;Ong et al, 2017).…”

“…Plain looped river networks differ from a dendritic river network in four aspects: substantial rivers, stronger river structure connectivity, considerable pump stations and sluice structures, and abundant gauging stations for monitoring water levels. Any localized alteration in a plain looped river network may lead to unpredictable changes in water level or flow in other remote regions (Kurnatowski, 2013). A large number of rivers increase the dimension of the n values to be calibrated.…”