1

F&C Lab

Environment and Computer Laboratory HS 2011 Water Resources Management Flood Routing Methods and Models Exercise Hydraulic Modelling of Open Channel Flows

Institut fr Umweltingenieurwissenschaften

IfU

F&C Lab

Outline
Organisational Issues Flood Routing Methods and Models Introduction of the Exercise Getting to know HEC-GeoRAS Getting to know HEC-RAS

Institut fr Umweltingenieurwissenschaften

IfU

F&C Lab

Organisational Issues
Supervisor: Verena Maleska, Christina Schornberg, Ellen Cerwinka Contact: [email protected] Time table
Date Friday, 11. Nov Topic Open Channel Flow hydrologic flood routing Introduction Exercise 1 HEC-RAS Paper assignments Individual work on Exercise Office hours: 12.45-16.30, HIL D 21.3 Room C 29

Friday, 18. Nov Friday, 25. Nov Friday, 2. Dec Friday, 9. Dec Friday, 16. Dec

C 29 C 29 C 29 C 29 C 29

Institut fr Umweltingenieurwissenschaften

IfU

F&C Lab

Organisational Issues
1 Exercise + Paper Assignments

- Building up the hydraulic model with HEC-GeoRAS in ArcGIS - Hydraulic Modelling of Open Channel Flows using HEC-RAS - Transferring the results of HEC-RAS into ArcGIS
Exercises carried out in groups of 2 students Hand in one report for the exercise and the paper assignments

- Report around 15 pages

Grading of WRM1

- HEC-RAS: 80% - Paper Assignments: 20%

Deadline for submission: 19th of December 2011

Institut fr Umweltingenieurwissenschaften

IfU

F&C Lab

Organisational Issues
Literature
Lecture notes Hydraulik 1 Prof. Kinzelbach Bollrich: Technische Hydromechanik Band 1 Manual HEC-RAS www.sciencedirect.com for paper

Institut fr Umweltingenieurwissenschaften

IfU

F&C Lab

Flood Routing methods and models

Lecture Outline
1 What is Flood Routing? motivation examples 2 Flood Routing hydraulic flood routing - approximations to the St. Venant equations

Institut fr Umweltingenieurwissenschaften

IfU

F&C Lab

Flood Routing methods and models

1) What is flood routing?
Flood (Flow) routing is a mathematical procedure for predicting the changing in magnitude, speed and shape of a flood wave as a function of time at one or more points along a river (Handbook of Hydrology, Fread, 1993)

Institut fr Umweltingenieurwissenschaften

IfU

F&C Lab

runoff production runoff routing

Institut fr Umweltingenieurwissenschaften

IfU

F&C Lab

1) floods predict flood propagation protection warning 2) design water conveyance systems protective measures hydrosystem operation 3) water dynamics ungauged rivers peak flow estimation river-aquifer interaction

Institut fr Umweltingenieurwissenschaften

IfU

10

F&C Lab

Risk analysis

R = 1 [1 P ( X xt )]n
R = hydrologic risk of failure is given by the probability that the annual flood event x larger than xt will occur at least once in n years e.g. a weir with a life period of n=50 years, P(HQT=100)=1/T=1/100

1 R = 1 [1 P( X x100 )] = 1 1 0.4 100

50
Institut fr Umweltingenieurwissenschaften

50

IfU

11

F&C Lab

2) Flood Routing
Flow (Flood) Routing Analysis

Institut fr Umweltingenieurwissenschaften

IfU

12

F&C Lab

Institut fr Umweltingenieurwissenschaften

IfU

13

F&C Lab

Steady Uniform Flow

dv/dt=0 dv/dx=0

Institut fr Umweltingenieurwissenschaften

IfU

14

F&C Lab

Gradually Varied Flow

dv/dt=0 dv/dx=0

Institut fr Umweltingenieurwissenschaften

IfU

15

F&C Lab

Institut fr Umweltingenieurwissenschaften

IfU

16

F&C Lab

Unsteady Flow: Physically Based = Hydraulic Routing

Institut fr Umweltingenieurwissenschaften

IfU

17

F&C Lab

St. Venant Equations

Institut fr Umweltingenieurwissenschaften

IfU

18

F&C Lab

Institut fr Umweltingenieurwissenschaften

IfU

19

F&C Lab

Institut fr Umweltingenieurwissenschaften

IfU

20

F&C Lab

Institut fr Umweltingenieurwissenschaften

IfU

21

F&C Lab

Institut fr Umweltingenieurwissenschaften

IfU

22

F&C Lab

Exercise: Hydraulic Modelling of Open Channel Flows

Objectives
Get to know

- a numerical hydraulic model for flood routing - the general methodology, possibilities and limitations of numerical
flood routing

- the role of different parameters and their influence on the

simulations and not

- to reproduce a flood event as accurate as possible no tuning

exercise

Institut fr Umweltingenieurwissenschaften

IfU

23

F&C Lab

Tasks
1 Steady State Simulations (d/dt = 0) 1.1 Rating curve (Q-H function) 1.2 Influence of roughness coefficient 2 Unsteady Flow Simulations 2.1 Influence of initial conditions (=> warm up time) 2.2 Investigation of the temporal discretization 2.3 Rating curve for unsteady flow 2.4 Numerical damping (influence of weighting factor ) 2.5 Influence of downstream boundary condition Answer questions on papers dealing with roughness in open channel flow

Institut fr Umweltingenieurwissenschaften

IfU

24

F&C Lab

River Reach & Flood Events

River Reach: Thur length of about 3 km Flood Events: Each group chooses and signs in for a flood event Flood hydrograph: hourly mean flow for the Thur in m3/s with a duration of 30 to 50 hours Too flashy to be realistic - but suitable for this exercise To be used for unsteady flow simulations

Bern

Thun

Institut fr Umweltingenieurwissenschaften

IfU

25

F&C Lab

Available Data
On the webpage: Tasks Flood events Orthophoto Elevation data Paper and questions Lecture notes, etc.

Bern

Thun

Institut fr Umweltingenieurwissenschaften

IfU

26

F&C Lab

Building up the Model for HEC-RAS Using ARGIS-Extension HEC-GeoRAS

Institut fr Umweltingenieurwissenschaften

IfU

27

F&C Lab

Flood routing model: HEC-RAS

HEC-RAS -> Hydrologic Engineering Centers River Analysis System developed by the US Department of Defense, Army Corps of Engineers in order to manage the rivers, harbors, and other public works 1D - hydraulic flow model a) Steady & Unsteady Flow b) Sediment Transport c) Water Temperature & Water Quality
http://www.hec.usace.army.mil/software/hec ras/ Institut fr Umweltingenieurwissenschaften

IfU

28

F&C Lab

Flood routing model: HEC-RAS

HEC-RAS -> Hydrologic Engineering Centers River Analysis System

http://www.hec.usace.army.mil/software/hec ras/ Institut fr Umweltingenieurwissenschaften

IfU

29

F&C Lab

Flood routing model: HEC-RAS

Steady Flow based on the solution of the one-dimensional energy equation Energy losses evaluated by friction and contraction / expansion momentum equation may be used in situations where water surface profile is rapidly varied (hydraulics of bridges ) Unsteady Flow full, dynamic Saint-Venant equations using an implicit, finite difference method

Institut fr Umweltingenieurwissenschaften

IfU

30

F&C Lab

Solutions of the Saint Venant Equations

Partial Differential Equation Analytically for 1-Dimensional Case Closed Integration e.g. Laplace/Fourier Transformation Numerically for Common Case use partial difference approximation Solutions: finite differences finite elements finite volume Nonlinear equation system: iteration!

Institut fr Umweltingenieurwissenschaften

IfU

31

F&C Lab

HEC-RAS: Implicit scheme to solve discretized equations

Explicit Schemes:

explicit scheme: Xt+1 = F(Xt) explicit schemes can be solved for Xt+1 straightforward computation stability issues for large t (CFL-criterion) e.g Forward step

dX X t +1 X t = F ( X t ) X t +1 = F ( X t ) t + X t t dt

Implicit Schemes:

- implicit scheme: Xt+1 = F(Xt and Xt+1) - generally, implicit schemes cant be solved
explicitly for Xt+1 (Xt+1 = .) (depending on F) solvable e.g. with matrix inversion - high computational demand - less strict stability criteria larger time steps possible - e.g. Trapezoidal rule
dX X t +1 X t 1 = F ( X t +1 ) + F ( X t ) t dt 2

-> Exc 2.2 & 2.4

Institut fr Umweltingenieurwissenschaften

IfU

32

F&C Lab

Courant Friedrich Levy (CFL) - Stability Criterion

To choose t for a given x Stability Criterion for a large class of explicit advection schemes:

x u t

u t 1 x

x t u

u: highest velocity in the system ( flood wave speed) t: computational time step x: spatial discretization ( distance between cross sections)

Time step t shouldn't be longer than the time a water particle needs to get from one computational node to the next (x) Or: The distance which a water particle moves in one time step (u*t) shouldnt be larger than x u t
x x

If t too long or u too high: Interpolation errors and instability Within limits of stability: trade off between computation time and accuracy
-> Exc 2.2: temporal discretization
Institut fr Umweltingenieurwissenschaften

IfU

33

F&C Lab

HEC-RAS: Preissmann-Cunge scheme

variables:

weighting factor

temporal derivative:

spatial derivative

how to weight present (t=n+1) variable / spatial derivative compared to past (t=n) variable / spatial derivative? numerically stable for 0.5 1 (in practice: 0.6 1)
-> Exc 2.4: weighting factor
Institut fr Umweltingenieurwissenschaften

IfU

34

F&C Lab

Deriving Inundation Maps Using HEC-GeoRAS

Institut fr Umweltingenieurwissenschaften

IfU

35

F&C Lab

And now
Form groups and choose a flood event Get to know HEC-RAS start working on the paper assignments Bathurst: What are the main elements governing flow resistance in natural channels? Is the river roughness constant in space and time? Wohl: How can you estimate the roughness coefficient? Which parameters influence the roughness in a channel? How do they influence the roughness? What is the influence of a varying roughness coefficient on the river flow? Can you make a general statement? Next week: working on the exercise Deadline 19th December 2011
IfU

Institut fr Umweltingenieurwissenschaften