Analysis of the relative contributions -

(hydrographs) of the sub-catchments during the flood

Contents

Interpolated Rainfall :« simple to complexe » methods

Hydrographs calculation SCS Method

Calibration of MIKE SHE for the VAR catchments Parameters, values, graphs

Contribution analysis during the floodCalibration???Conclusions

Rainfall Interpolation Methods :« simple to complexe »

Homogeneous Rainfall on the sub-catchment

Thiessen Method

Kriging Method

Hypothesis : Spatial distribution of the rainfall are the same on the all catchment=> mean of the six rain gauges station

o Homogeneous Rainfall on the sub-catchment

Rainfall Interpolation Methods :« simple to complexe »

o Thiessen Method

• Thiessen Polygon (ARCGIS)

• Assigning to each station an influence area (%) that represents weighting factor.

• To calculate the interpolated rain :

∑ rainfall for each station x weighting factor ---------------------------------------------------- Total area conerning

Tinée Upper Var Vésubie Down Var Esteron

Carros 0 0 0 64 8

Levens 6 2 22 36 7

Roquesteron 0 7 0 0 46

Puget Théniers 0 38 0 0 39

Guillaumes 48 53 0 0 0

St Martin de Vésubie 47 0 78 0 0

• Estimating rainfall weighted taking into account each station.

Rainfall Interpolation Methods :« simple to complex »

o Kriging Method

Interpolation by kriging for each sub-catchment and for each hour

Hydrographs calculation

SCS Method (Soil Conservation Service) Hypothesis 1: Infiltration capacity tends to zero as time increases.Hypothesis 2: Runoff appear after it dropped some rainfall.

Hypothesis 3: R (t) = [ si Pu (t) > 0 ]

Temps

Lames d'eau cumulées

R(t)

Pu(t)P(t)

So

S

Temps infini

0

tJ(t) dt

Cumulated Water

Finish Time

Time

)t(PS)t(P

u

2u

SCS Parameters:• S: Maximum infiltration capacity, depend on Soil characteristics, cover, condition of initial wetting.

Tinée Upper Var Vésubie Down Var Esteron SCS Value 85 85 85 30 80• Tm: Time of the peak discharge, base on Concentration Time ( Tm

= 3/8 Tc). • Tc: Concentration Time, calculate with PASSINI Method (take in account: Surface, Slope and Longest Flow Path).

• Area (km²) : Surface of each sub-catchment.

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162

0

2

4

6

8

10

12

14

16

18 0

50

100

150

200

250

300

350

400

Conversion Rainfall-Discharge for the Tinee Sub-catchment

Total Rainfall Runoff

Hours

Rain

fall

mm

Disc

harg

e m

3/s

Hydrographs calculation

SCS Result:

Catégo

rie 1

Catégo

rie 2

Catégo

rie 3

Catégo

rie 4

0

2

4

6

Série 1Série 2Série 3

• Almost no big differences appear between the rainfall distribution results from the Thiessen and the homogeneous method

• The discharge value are globally in accord with calculate value in the Napoleon Bridge

• Except for Surfer method. Doesn’t take in account the different landuse, the slope or the topography. With more than we could obtain better result including topography in Surfer.

• Homogeneous discharge is more important than the Thiessen value. Due to Thiessen method take in account spatial reference of the station.

11/0

4/19

94 /

11 0

013

00

15 0

017

00

19 0

021

00

23 0

011

/05/

1994

/ 01

00

03 0

005

00

07 0

009

00

11 0

013

00

15 0

017

00

19 0

021

00

23 0

011

/06/

1994

/ 24

00

04 0

006

00

08 0

010

00

12 0

014

00

16 0

018

00

20 0

022

00

24 0

00

500

1000

1500

2000

2500

3000

3500

4000

4500

5000 Hydrographs for the Var’s Catchment

Time (hours)

Discharge (m3/s)

Kriging

Napoléon B.

Thiessen

Homogen-eous.

Tinee hydrograph

Hydrographs calculation

SCS Result: 

Discharge (m3/s)

Contributions of the sub-catchments during the flood (%)

Tm (hours)

Tinee 1177,8 28 8

Vésubie 869,1 20 5

UpperVar 1204 28 12

DownVAr 336 8 4

Estéron 689 16 8

Calibration of MIKE SHE

First calibration-using only MIKE SHE Using: 300 m grid sizeExperiences : very little peak of runoff

the width of the imagined river bed is 1500 mReasons: big grid size too big width ofriver bed, big hydraulic radius and little water

depthlittle velocity and discharge

Conclusion: we have to use river network for modelling

coupling with MIKE11

overland flow in y-direction [m^3/s]

00:001994-11-05

04:00 08:00 12:00 16:00 20:00 00:0011-06

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

Calibration of MIKE SHE and coupling with MIKE 11

Parameters:IWD - Initial Water depth 0,00-0,005DS – Detention storage 0,00-0,05 Manning number (overland)10,0-40,0Net Rainfall Fraction 0,90-0,95

Calibration of MIKE SHE and coupling with MIKE 11

Parameters of the best calibration: M=24 m1/3/sNRF=0.93IWD = 0.000 mDS= 0.00 mm

Results of calibration: Peak of discharge Qc= 3701 m3/s Qm= 3680m3/s

Wrong time of the peak 2.5 hours differences sensitivity analysis not sensitive M,IWD,NRF little sensitive

DS Conclusions: We can’t calibrate more accurately under these conditions (300

grid size) and It’s not necessery because there are not observed data!

Contributions analysis

The runoff’s peak and timing depends on the following parameters:

Shape of the catchmentLanduse surface roughnessTopographyRainfall, Area

Var sub-catchments:same Landuse more than 90% forest and natural area except Down Var sub-catchment similar topography

Differences: rainfall, area, shape of the sub-catchments

Contributions analysis

Similar runoff characteristic on every sub-catchmentRelative contributions of runoff: Q%=∑Q/Qi A%=∑A/Ai

Esteron:20% c= Q%/A%=128%

Vesubie:8% c=57%Tineé: 32% c=120%Upper Var: 36% c=93% Down Var: 4% c=74%

Calibration ???

Similar runoff characteristic on every sub-catchmentRelative contributions of runoff: Esteron:21%

Vesubie:5%Tineé: 36%Upper Var: 36,5%

Down Var: 1,5%

Calibration ???

Conclusions

The relative contribution of sub-catchments only depends on the distribution of rainfall.

The Tinee, Upper Var, Esteron gave more than 90% of the whole runoff.

CONCLUSIONS OF MIKE PART:If we calculate the relative contributions of the sub-catchments (during the flood), we don’t need to use calibrated modell, because the relative contribution is not sensitive for the calibrated parameters.

TEAM SIX . . .

Thank for your attention