NONLINEAR MODELING AND PREDICTION OF A RIVER FLOW SYSTEM<sup>1</sup>

Tabrizi, M.H.N.; Said, Salim; Badr, A. W.; Mashor, Y; Billings, S.A.

doi:10.1111/j.1752-1688.1998.tb05435.x

Cited by 5 publications

(4 citation statements)

References 11 publications

(8 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Direct approaches to modeling rainfall and flow include ARX, NARMAX (Tabrizi et al, ) and functional coefficient modeling (Wong et al, ). It has been recognized that the shape of h is an important model choice and some authors have implemented polynomial constraints (Tabrizi et al, ) on h or used local polynomial smoothers (Wong et al, ). Models of the form where the impact of one time‐dependent variable, x ( t ), on another, y ( t ), is spread over time, can be called a DLM.…”

Section: Modeling With Dlmsmentioning

confidence: 99%

“…This is known as the instantaneous unit hydrograph (Nash, 1957), describing the impact over time that a unit of rainfall has on flow. Jakeman, Littlewood, and Whitehead (1990) suggested filtering rainfall data to first estimate "effective runoff" before proceeding to estimate h. Direct approaches to modeling rainfall and flow include ARX, NAR-MAX (Tabrizi et al, 1998) and functional coefficient modeling (Wong et al, 2007). It has been recognized that the shape of h is an important model choice and some authors have implemented polynomial constraints (Tabrizi et al, 1998) on h or used local polynomial smoothers (Wong et al, 2007).…”

Section: Modeling With Dlms 21 the Dlmmentioning

confidence: 99%

See 1 more Smart Citation

Distributed Lag Models for Hydrological Data

et al. 2013

View full text Add to dashboard Cite

Summary: The distributed lag model (DLM), used most prominently in air pollution studies, finds application wherever the effect of a covariate is delayed and distributed through time. We explore the use of modified formulations of DLMs to provide flexible varying-coefficient models with smoothness constraints, applicable in any setting in which lagged covariates are regressed on a time-dependent response. The models are applied to simulated flow and rainfall data and to flow data from a Scottish mountain river, with particular emphasis on approximating the relationship between environmental covariates and flow regimes in order to detect the influence of unobserved processes. It was found that under certain rainfall conditions some of the variability in the influence of rainfall on flow arises through a complex interaction between antecedent ground wetness and the time-delay in rainfall. The models are able to identify subtle changes in rainfall response, particularly in the location of peak influence in the lag structure and offer a computationally attractive approach for fitting DLMs.

show abstract

Section: Modeling With Dlmsmentioning

confidence: 99%

Section: Modeling With Dlms 21 the Dlmmentioning

confidence: 99%

Distributed Lag Models for Hydrological Data

et al. 2013

View full text Add to dashboard Cite

show abstract

“…In the past, studies (Liang et al 1994;Young and Beven 1994;Tabrizi et al 1998) have employed different function optimization techniques such as multiple input transfer functions (Ahsan and O'Connor 1994) to automate parameter estimation for hydrological models. These traditional optimization methods use deterministic transition rules to find the optimum solution for an optimization problem.…”

Section: Introductionmentioning

confidence: 99%

Selecting Model Parameter Sets from a Trade-off Surface Generated from the Non-Dominated Sorting Genetic Algorithm-II

Dumedah

Berg

Wineberg

et al. 2010

Water Resour Manage

View full text Add to dashboard Cite

There is increasing trend in the use of multi-objective genetic algorithms (GAs) to estimate parameter sets in the calibration of hydrological models. Multiobjective GAs facilitate the evaluation of several model evaluation objectives, and the examination of massive combinations of parameter sets. Typically, the outcome is a set of several equally-accurate parameter sets which make-up a trade-off surface between the objective functions, usually referred to as Pareto set. The Pareto set is a set of incomparable parameter sets as each solution has unique parameter values in parameter space with competing accuracy in the objective function space. As would be required for decision making purposes, a single parameter set is usually chosen to represent the model calibration procedure. An automated framework for choosing a single solution from such a trade-off surface has not been thoroughly investigated in the model calibration literature. As a result, this study has outlined an automated framework using the distribution of solutions in objective space and parameter space to select solutions with unique properties from an incomparable set of solutions. Our Pareto set was generated 4470 G. Dumedah et al.southern Ontario. Using cluster analysis to evaluate the distribution of solutions in both objective space and parameter space, we developed four auto-selection methods for choosing parameter sets from the trade-off surface to support decision making. Our method generates solutions with unique properties including a representative pathway in parameter space, a basin of attraction (or the center of mass) in objective space, a proximity to the origin in objective space, and a balanced compromise between objective space and parameter space (denoted BCOP). The BCOP method is appealing as it is an equally-weighted compromise for the distribution of solutions in objective space and parameter space. That is, the BCOP solution emphasizes stability in model parameter values and in objective function values-in a way that similarity in parameter space implies similarity in objective space.

show abstract

“…The forecasting, backcasting and flow gaps in-filling abilities of well formulated streamflow models and the insights to the processes generating and controlling the flows that such models can provide make them useful tools in many areas of catchment scale water resources management. Such areas include flood forecasting and control, drought management, assessment of water supply potentials and waste load carrying capacities of streams, design of engineering hydraulic structures such as on-the-river reservoirs and the assessment of the impacts of anthropogenic effects on both the spatial and temporal distribution of streamflow as well as the water quality of the streams (Tabrizi et al, 1998;Beven, 2000).…”

Section: Introductionmentioning

confidence: 99%

A hybrid metric‐conceptual (HMC) model for monthly riverflow prediction in the semi‐arid Volta Basin of West Africa

Amisigo

Giesen

Andah

2007

International Journal of River Basin Management

View full text Add to dashboard Cite

Data-based mechanistic modelling techniques have been applied to catchment monthly runoff, potential evaporation and rainfall time series to model monthly catchment runoff at selected gauging sites in the Volta Basin of West Africa. The aim of the study was to obtain a modelling framework that not only accounts for the rainfall-runoff non-linearity in the basin and provides acceptable predictions of the monthly catchment runoff for the sub-basins studied, but also reveals the necessary insights for a plausible interpretation of the rainfall-runoff mechanism in the basin.The rainfall-runoff process was considered in two stages -a nonlinear transformation from rainfall to effective rainfall and then a linear transformation from effective rainfall to runoff. First a linear time varying, state dependent parameter (LTV-SDP) transfer function model was applied to the monthly rainfall-runoff calibration series of each sub-basin to determine the form of the rainfall-effective rainfall non-linear transformation. The observed series (runoff or rainfall) that was significantly related to effective rainfall was then identified. Next, the functional form of this relationship was established and used to fit linear time invariant (LTI) transfer function models relating monthly runoff to monthly effective rainfall and potential evaporation.The best estimate of the monthly effective rainfall from the nonlinear modelling was obtained from the product of monthly rainfall and a fractional power of monthly runoff -the runoff acting as surrogate for catchment wetness. Since this form of effective rainfall cannot be used for simulation in validation mode (as the runoff series is unknown then), a non-linear rainfall filter used in the Identification of unit Hydrographs And Component flows from Rainfall, Evaporation and Streamflow data (IHACRES) application was used to model effective rainfall in this mode. The study showed that monthly catchment runoff in the basin can be decomposed into two parallel flows, an instantaneous (within a month) and a slower (with one month delay) flow components. These results suggest that baseflow contribution to streamflow is insignificant in the Volta Basin.

show abstract

NONLINEAR MODELING AND PREDICTION OF A RIVER FLOW SYSTEM¹

Cited by 5 publications

References 11 publications

Distributed Lag Models for Hydrological Data

Distributed Lag Models for Hydrological Data

Selecting Model Parameter Sets from a Trade-off Surface Generated from the Non-Dominated Sorting Genetic Algorithm-II

A hybrid metric‐conceptual (HMC) model for monthly riverflow prediction in the semi‐arid Volta Basin of West Africa

Contact Info

Product

Resources

About

NONLINEAR MODELING AND PREDICTION OF A RIVER FLOW SYSTEM1

Cited by 5 publications

References 11 publications

Distributed Lag Models for Hydrological Data

Distributed Lag Models for Hydrological Data

Selecting Model Parameter Sets from a Trade-off Surface Generated from the Non-Dominated Sorting Genetic Algorithm-II

A hybrid metric‐conceptual (HMC) model for monthly riverflow prediction in the semi‐arid Volta Basin of West Africa

Contact Info

Product

Resources

About

NONLINEAR MODELING AND PREDICTION OF A RIVER FLOW SYSTEM¹