SIMULATION OF JUNE 2013 FLOOD AT TEHRI DAM Niraj Agrawal DGM - - PowerPoint PPT Presentation

1

3 rd NATIONAL DAM SAFETY CONFERENCE

18 and 19 February, 2017

SIMULATION OF JUNE 2013 FLOOD AT TEHRI DAM

Niraj Agrawal

DGM (Design) THDC India Limited Rishikesh

2

BROAD CONTENTS

• Background of the Project
• Data Availability
• Methodology
• Results & Discussions
• Conclusions

Type : Rock and Earth fill Height of dam : 260.5 m Base : 1128 m Width at top : 25.5 m Length at the top : 575 m

THDC India Ltd.

4

L-SECTION OF TEHRI HYDRO POWER COMPLEX (UPPER AND LOWER RESERVOIRS)

• FRL. EL. 830.0 M
• MDDL. EL. 740.0 M
• FRL. EL. 612.5 M

MIN.LEVEL EL. 606 M

TEHRI DAM ( STAGE-I) 1000 MW KOTESHWAR DAM 400 MW ~ 22 KM UPPER RESERVOIR LOWER RESERVOIR

TEHRI HPP KOTESHWAR HPP PSP

TEHRI DAM ( STAGE-II) 1000 MW

5

Water Spread : 42 Km2 Gross Storage : 3540 M CM Live Storage : 2615 M CM M ax. Flood Level : EL 835 m Full Reservoir Level : EL 830 m M DDL : EL 740 m

THDC India Ltd.

CHUTE SPILLWAY

Head : 220 M Type : Conventional Stilling Basin Type M ax Discharge : 5500 cumecs Width : 39.5m at top 50m at toe Regulating Gates : Radial 15.5m high/ 10.5m wide Aerators : 3 nos on Glacis Stilling Basin : 140 x 50 m 22m deep pool 55 m high walls D/s River Bed : Protection by Concrete Blocks

THDC India Ltd.

RIGHT BANK SHAFT SPILLWAYS

Head : 220 M Type : Vertical Shafts Nos : 2 (Un-gated) Intakes : Funnel type (34m dia) Discharge : 3900 cumecs Vertical Shaft : 12m dia Junction with : T angential with 6.0m tunnel

• pening

De-aeration System : Through a Separation Chamber and de-aeration shaft opening above M WL Velocity at toe : 45 m/s

10

LEFT BANK SHAFT SPILLWAYS

Head : 220 M Type : Vertical Shafts Nos : 2 (Gated) Intakes : Tunnel type (80m long) Discharge : 3800 cumecs Vertical Shaft : 12m dia Junction with : T angential with 5.5m tunnel

• pening

De-aeration System : Through a Separation Chamber and de-aeration shaft opening in a De- aeration tunnel Velocity at toe : 45 m/s

THDC India Ltd.

THDC India Ltd.

INTERM EDIATE LEVEL OUTLET

• BOTTOM M OST OUTLET IN THE

PROJECT AT EL 700M .

• BEING USED FOR INITIAL FILLING OF

RESERVOIR

• TO BE USED FOR IRRIGATION

RELEASES

• TO BE USED FOR EM ERGENCY

EVACUATION

• T

ANGENTIALL Y JOINING T-3 SHAFT SPILLWAY

• LINING COATED WITH POL

YUREA

THDC India Ltd.

Designed For PM F Of 15540 Cumecs

FLOODS ON 16th ,17th & 18th JUNE 2013

• Peak Discharge at Haridwar

due to River Alakhnanda = 15000 Cumecs

• Water level at Haridwar at 4pm

= 295.90 m (1.90 m above danger level)

• Actual Peak Discharge of River Bhagirathi at Tehri at 6am

= 7500 Cumecs (262500 Cusecs)

• Releases from Tehri Dam

= 500 Cumecs (17500 Cusecs)

• Prevented Flooding of Rishikesh & Haridwar by about 2.5-3.0 m

2110 2280 2700 3150 3700 3999 4179 4259 4419 4799 5474 6549 7799 8549 8549 8049 7299 5949 5774 6124 6299 6799 6299 5599 4649 4179 3949 3750 3590 3420 2231 2523 3307 4608 6536 6537 7464 6947 7405 6889 8161 9236 11083 15284 15728 13923 12357 9376 9201 9061 9823 10736 10049 8224 7087 6242 5637 5437 5089 4169 2000 4000 6000 8000 10000 12000 14000 16000 18000 10:00 12:00 14:00 16:00 18:00 20:00 22:00 00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 16.6.13 17.6.13 18.6.13 D i s c h a r g e i n C u m e c s Date and Time

Actual observed discharge of Ganga at Rishikesh and anticipated discharges after superimosing Tehri actual inflows with 10 hrs time lag

Actual inflow at Rishikesh (Cumecs) Anticipated inflow at Rishikesh if actual inflows at Tehri superimposed with 10 hrs time lag (Cumecs)

2241 3446 4843 9633 9112 9166 9023 8503 9939 11063 11251 12321 13008 13037 14340 14457 12951 12713 12321 12208 11722 11483 11167 10284 9651 8939 8222 8004 79197705 2363 3689 5451 11091 11949 11704 12307 11190 12925 13153 13938 15008 16292 19772 21519 20331 18009 16140 15748 15145 15246 15420 14917 12909 12088 11001 9909 9692 9419 8455 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 22000 24000 12:00 14:00 16:00 18:00 20:00 22:00 00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 16.6.13 17.6.13 18.6.13 D i s c h a r g e i n C u m e c s Date and Time

Actual observed discharge of Ganga at Haridwar and anticipated discharges after superimosing Tehri actual inflows with 12 hrs time lag

Actual inflow at Haridwar ( Cusecs) Anticipated inflow at Haridwar if actual inflows at Tehri superimposed with 12 hrs time lag (Cusecs)

350 350 350 350 350 350 350 350 350 350 350 350 355 355 355 355 355 355 355 355 355 355 355 355 425 425 425 425 425 425 471 593 958 1808 3186 2888 3634 3037 3336 2440 3037 3037 3639 7090 7535 6229 5413 3782 3782 3292 3879 4292 4104 2980 2862 2487 2112 2112 1925 1175 1000 2000 3000 4000 5000 6000 7000 8000 00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 00:00 02:00 04:00 06:00 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 00:00 02:00 04:00 06:00 08:00 10:00 16.6.13 17.6.13 18.6.13 D i s c h a r g e i n C u m e c s Date and Time

Actual observed inflow and outflow at Tehri

Outflow from Tehri Reservoir (Cumecs) Inflow Tehri (Cumecs)

20

DATA AVAILABILITY

Rainfall Data: Discharge Data:

• June 2013: Daily peak discharge of River Bhagirathi at

Maneri and River Bhilangana at Ghansali.

• 16th, 17th & 18th June: Hourly discharge at Maneri and

intermittent discharge at Ghansali.

Reservoir Data:

• Reservoir Elevation-Capacity Curve
• 16th, 17th & 18th June: T

wo hourly reservoir level reading

• Outflow from reservoir

21

METHODOLOGY

• 90 m resolution SRTM data was

analyzed in Arc-GIS 9.3 for extracting the geomorphologic features of the catchment.

• Whole catchment has been divided

into three sub- catchment based on the physiographic features and data availability.

Fig 3 : Total Catchment Fig 4 : Bhagirathi upto Dharasu Fig 5 : Bhilangana upto Ghansali Fig 6 : Area between Dharasu and Ghansali

22

• For classification of snow

fed and rain fed area, Land sat data of the catchment for the month of June 2013 has been analyzed in Erdas Imagine 2011.

• Result was matched with

the true color imagery of the area and the previous study report.

Fig 7 : Whole Tehri Catchment Fig 9 : Bhagirathi Catchment upto Dharasu Fig 8: Bhilangana Catchment upto Ghansali

METHODOLOGY

23

HORTON’S RATIOS

• Hortan’s stream order ratio, area ratio and length ratio are calculated as follows:
• (RB) = Nω-1/ Nω,

(RA) = Aω/ Aω-1 & (RL) = Lω/ Lω-1

• S. No.

RB RA RL AΩ LΩ Remarks Bhagirathi 4.99 5.92 2.97 5050.00 152.00 Bhilangana 4.00 4.70 3.00 833.00 61.20 W1 2.75 3.00 2.34 10.73 5.41 W2 3.17 3.73 2.76 108.03 25.21 W3 2.50 3.45 2.98 9.15 5.73 W4 3.50 4.72 3.37 46.15 11.58 W5 2.50 4.11 3.57 11.45 7.89 W6 3.00 4.68 3.76 16.27 9.04 W7 2.74 3.29 2.52 69.68 18.04 W8 3.95 4.53 2.89 259.04 35.86 W9 3.50 4.43 3.57 21.34 10.26 W10 3.83 4.86 3.30 41.81 11.94 W11 3.00 4.28 3.20 13.20 7.35 W12 2.25 3.70 2.35 7.78 4.25 W13 2.50 3.03 2.65 16.99 7.16 W14 2.00 2.65 2.79 12.65 5.90 W15 3.33 4.02 2.49 43.95 10.53 W16 3.00 3.49 1.95 14.24 7.42 W17 5.23 6.39 3.66 472.89 49.27

24

GIUH Model:

qp = 1.31 RL

0.43 (V/ Lω)

tp = 0.44 RL

• 0.38 (RB/ RA)0.55 (Lω/ V)

GcIUH Model:

qp = 0.871/ ᴨi

0.4

tp = 0.585 ᴨi

0.4

ᴨi = LΩ

2.5/ (ir* AΩ* RL* αΩ 1.5)

αΩ = SΩ

0.5/ (n* bΩ 0.667)

GIUH-Nash Model:

The concepts of GIUH and the Nash IUH models are used to derive the GIUH based Nash model. The complete shape of the GIUH can be obtained by linking qp and tp of the GIUH with the scale k and shape parameter n

• f Nash IUH model.

25

Distribution of Daily Rainfall to Hourly Rainfall:

• Rainfall records of SSRG of Chandrabhaga

experimental watershed near the T ehri dam was taken from NIH Roorkee.

• Similar distribution pattern was followed to

distribute the daily rainfall records within the catchment to hourly data.

Abstraction & Infiltration Losses:

• 10 mm initial abstraction loss.
• 2.5mm/ hr infiltration loss.

Fig 10 : Mass curve of hourly & daily rainfall

EFFECTIVE RAINFALL HYETOGRAPH:

26

Average Rainfall Over the Basin:

For the Bhilangana sub-catchment and the region between Dharasu and Ghansali, the rainfall record of T ehri station is taken as the average rainfall. For Bhagirathi catchment, Thiessen polygon method is used to get the average rainfall from the records of Dunda, Uttarkashi and Gangotri station.

Fig 11 : Location of raingauge stations

EFFECTIVE RAINFALL HYETOGRAPH:

27

SIMULATED FLOOD HYDROGRAPHS

• Convolution of excess rainfall hyetograph with the unit hydrograph.
• Reduction of Bhagirathi (Dharasu ) & Bhilangana (Ghansali) DSRO by the

percentage of rainfed area.

• DSRO hydrograph from stream no. 1 to 10 were added and multiplied by area
• ratio. (A10 = 594 km 2, Area between Dharasu & dam axis = 757 km 2)
• Similarly DSRO hydrograph of stream no. 11 to 17 were added and multiplied by

area ratio. (A7 = 582 km 2, Area between Ghansali and dam axis = 647 km 2)

• The resultant DSRO hydrograph were added to the hydrograph of River Bhagirathi

and River Bhilangana respectively.

• Flood hydrograph at T

ehri dam axis was obtained by adding the DSRO hydrograph

• f Bhagirathi and Bhilangana with the average measured base flow for the month
• f June 2013.

28

SIMULATED FLOOD HYDROGRAPHS

Observed Flood Hydrographs:

• Incremental volume retained; from observed reservoir elevation data and elevation capacity curve.
• Discharge through power tunnel was added to get the gross inflow discharge into the reservoir.

Fig 12 : Elevation- capacity curve

29

SIMULATED FLOOD HYDROGRAPHS

Fig 12 : Observed and Simulated Flood Hydrograph at Tehri dam axis and Dharasu

Model Efficiency

30

SIMULATED FLOOD HYDROGRAPHS

• The gross and net volume retained in the Tehri reservoir during 16th ,17th and 18th June 2013

was 748.9 MCM and 651.3 MCM respectively.

• Increase in reservoir level 27.8 m (749.0 m to 766.8 m).
• The gross volume form different sub-catchments are as below:

31

DISCUSSION

• The shape of GIUH and GcIUH model are triangular whereas the GIUH-Nash model gave

a complete shape of IUH.

• The simulated flood hydrographs from each model shows smooth transition whereas

the observed flood hydrographs has number of spikes.

• The time to peak of each flood hydrographs are few hours earlier than the actual
• bserved time to peak as the attenuating action of reservoir is ignored.
• The Nash-Sutcliff efficiency of GIUH, GcIUH and GIUH-Nash model are 75.8%, 79.1 %

and 80.5% respectively. The efficiency of GIUH-Nash model is higher as the model gives the complete IUH shape.

• Volume per unit area from the sub-catchment between Dharasu and Ghansali is higher

as compared to rest catchment.

32

CONCLUSIONS

• The GIUH and GcIUH approach are able to reproduce the characteristics of the

direct surface runoff hydrographs.

• The GIUH-Nash model gives the complete shape of DSRO hydrograph and the

model efficiency is also higher.

• The simulated volume at Tehri dam nearly equals the actual observed volume

during the flood event.

• These models could be efficiently used for predictions in an ungauged basins.

INDIAN INSTITUTE OF TECHNOLOGY ROORKEE