A simple methodology for using pre-computed hydrological models in flood early warning systems

Presenter Information

Herman Dolder

Location

Eccles Conference Center

Event Website

http://water.usu.edu

Start Date

4-1-2014 6:00 PM

End Date

4-1-2014 6:20 PM

Description

In this paper I present a methodology to produce nearly instantaneous model results given a combination of input parameter values. This is achieved by selecting from a set of pre-computed models the one that better matches the input parameters provided. In this paper I analyze, first, how to define the combination of input parameters to more evenly populate the parameters’ space, and second, how to determine which is the model that corresponds to the parameters specified. For the first part I will use a Latin Hypercube approach, detailing certain particularities related to the probability distribution to be used, and for the second I will explain how the selection can be done using a simple homogenized Euclidian distance calculation. I devised this methodology in the context of the CI-WATER project, with the intention to use it in the development of flood early warning systems.

This document is currently not available here.

Share

COinS
 
Apr 1st, 6:00 PM Apr 1st, 6:20 PM

A simple methodology for using pre-computed hydrological models in flood early warning systems

Eccles Conference Center

In this paper I present a methodology to produce nearly instantaneous model results given a combination of input parameter values. This is achieved by selecting from a set of pre-computed models the one that better matches the input parameters provided. In this paper I analyze, first, how to define the combination of input parameters to more evenly populate the parameters’ space, and second, how to determine which is the model that corresponds to the parameters specified. For the first part I will use a Latin Hypercube approach, detailing certain particularities related to the probability distribution to be used, and for the second I will explain how the selection can be done using a simple homogenized Euclidian distance calculation. I devised this methodology in the context of the CI-WATER project, with the intention to use it in the development of flood early warning systems.

https://digitalcommons.usu.edu/runoff/2014/2014Posters/39