18
Water Security, Risk and Society Oxford, April 2012 Small probability events that could cause water supply collapse Jerson Kelman CEO of Light Group

Small probability events that could cause water supply collapse – how to deal with them

Embed Size (px)

DESCRIPTION

Professor Jerson Kelman, CEO of Light S.A., Federal University of Rio de Janeiro, Brazil Small probability events that could cause water supply collapse – how to deal with them PLENARY: A risk perspective on water security Introduces risk as a unifying framework to address these challenges at multiple scales

Citation preview

Page 1: Small probability events that could cause water supply collapse – how to deal with them

Water Security, Risk and Society Oxford, April 2012

Small probability events

that could cause water supply collapse  

Jerson Kelman

CEO of Light Group

Page 2: Small probability events that could cause water supply collapse – how to deal with them

Hydrological uncertainty rbhrghrbgrfrrfrr

Região Sudeste / Centro-Oeste

0

20

40

60

80

100

120

Nov/99 Dez/99 Jan/00 Fev/00 Mar/00 Abr/00 Mai/00 Jun/00 Jul/00 Ago/00 Set/00 Out/00 Nov/00 Dez/00

% EAR MÁX

Page 3: Small probability events that could cause water supply collapse – how to deal with them

0

10

20

30

40

50

60

70

80

90

100

jan/97

jul/97

jan/98

jul/98

jan/99

jul/99

jan/00

jul/00

jan/01

jul/01

jan/02

Storage in the equivalent reservoir

Beginning 2000 Beginning

2001

Page 4: Small probability events that could cause water supply collapse – how to deal with them

Water Storage in the Sobradinho Reservoir Simulação do comportamento do reservatório de FURNAS

(Vazão firme = 670 m³/s)

0

5000

10000

15000

20000

25000

Jan-3

1

Jan-3

4

Jan-3

7

Jan-4

0

Jan-4

3

Jan-4

6

Jan-4

9

Jan-5

2

Jan-5

5

Jan-5

8

Jan-6

1

Jan-6

4

Jan-6

7

Jan-7

0

Jan-7

3

Jan-7

6

Jan-7

9

Jan-8

2

Jan-8

5

Jan-8

8

Jan-9

1

Jan-9

4

Jan-9

7

Jan-0

0

Mês/ano

Volu

me

(hm³)

Armazenamento final (hm3) Volume Morto (hm3)

Nov/56 = 5733 hm³

Volume morto = 5733 hm³

Fonte dos Dados: ONS

Fourth worst drought in record

Page 5: Small probability events that could cause water supply collapse – how to deal with them

Lessons from case 1 •  Decisions under uncertainty are politically evaluated by their consequences, not by their wisdom

•  The layman expects that the water (or electric energy) supply system is able to endure the worst drought in record

•  It is sometimes preferable to make easy to understand but less than optimal decisions

Page 6: Small probability events that could cause water supply collapse – how to deal with them

Sobradinho Reservoir

Sao Francisco River Basin

Brazil

Drought prone area

Page 7: Small probability events that could cause water supply collapse – how to deal with them

During the 1887-89 drought, about one million people emigrated from the drought prone area Countless died

Page 8: Small probability events that could cause water supply collapse – how to deal with them

HYDROLOGICAL VICIOUS

CYCLE

No firm water supply

Little investment

in high-value crops or industry

People are

Poor

It  is  necessary  an  initial  stock  of  investments  on  water  infrastructure  before  reaching  the  “in8lexion  point…  and  then  real  progress  starts    (David  Grey  and  Claudia  Sadoff,  “Sink  or  Swim?  Water  security  for  growth  and  development”)

Page 9: Small probability events that could cause water supply collapse – how to deal with them

INTERBASIN WATER TRANSFER

Page 10: Small probability events that could cause water supply collapse – how to deal with them
Page 11: Small probability events that could cause water supply collapse – how to deal with them

Qmean São Francisco River = 2600 m3 /s

26 m3 /s ≤ Qdiversion ≤ 127 m3 /s 1% of Qmean ≤ Qdiversion ≤ 5% of Qmean

600 Km of channels

Page 12: Small probability events that could cause water supply collapse – how to deal with them

Water  Storage  in  the  Sobradinho  Reservoir    Simulação do comportamento do reservatório de FURNAS

(Vazão firme = 670 m³/s)

0

5000

10000

15000

20000

25000

Jan-3

1

Jan-3

4

Jan-3

7

Jan-4

0

Jan-4

3

Jan-4

6

Jan-4

9

Jan-5

2

Jan-5

5

Jan-5

8

Jan-6

1

Jan-6

4

Jan-6

7

Jan-7

0

Jan-7

3

Jan-7

6

Jan-7

9

Jan-8

2

Jan-8

5

Jan-8

8

Jan-9

1

Jan-9

4

Jan-9

7

Jan-0

0

Mês/ano

Volu

me

(hm³)

Armazenamento final (hm3) Volume Morto (hm3)

Nov/56 = 5733 hm³

Volume morto = 5733 hm³

Fonte dos Dados: ONS

Page 13: Small probability events that could cause water supply collapse – how to deal with them

Opportunity cost of water in the Sobradinho Reservoir

90% of capacity

Sobradinho Reservoir

Page 14: Small probability events that could cause water supply collapse – how to deal with them

Continuous flow riv

erSmall reservoir in the recipient region

PipelineWater flowfrom theSão FranciscoRiver Basin

Inte

rmitt

ent r

iver

Continuous flow riv

erSmall reservoir in the recipient region

PipelineWater flowfrom theSão FranciscoRiver Basin

Inte

rmitt

ent r

iver

Page 15: Small probability events that could cause water supply collapse – how to deal with them

95% 95%

Water reliability in different site of the recipient region

w without diversion with diversion

Page 16: Small probability events that could cause water supply collapse – how to deal with them

Continuous flow riv

erSmall reservoir in the recipient region

PipelineWater flowfrom theSão FranciscoRiver Basin

Inte

rmitt

ent r

iver

Continuous flow riv

erSmall reservoir in the recipient region

PipelineWater flowfrom theSão FranciscoRiver Basin

Inte

rmitt

ent r

iver

•  No Interbasin Diversion Project •  Build individual water tanks •  Store rain falling on the roofs

Water for survival

•  Interbasin Diversion Project •  Build local reservoirs •  Build pipes and channels

Water for production

X

“I will sustain a hunger strike until Government cancels the Project” (Bishop Dom Cappio)

Page 17: Small probability events that could cause water supply collapse – how to deal with them

Lessons from case 2 •  When the probability P of a catastrophic drought tends to zero and the corresponding cost C tends to infinity…

Expected Cost … 0 × ∞ …undefined

Mathematical lack of definition… … political lack of resolution

•  When ideology opposes hydrology, it is very difficult to discuss water security challenges using risk as a unifying framework

Page 18: Small probability events that could cause water supply collapse – how to deal with them

Thank you!

[email protected]