This document discusses water management and poverty in the São Francisco River Basin in Brazil. It summarizes trends in population growth, rural to urban migration, aging of the rural population, changes in market structure and agriculture. Key policy issues around water allocation, irrigation, and poverty reduction are identified. Research being conducted in the basin aims to develop hydro-economic models to understand farmer behavior under different policies and assess trade-offs between agriculture, poverty, and other sectors. The models link hydrologic and economic components to analyze impacts at various spatial scales from plot to basin level.

1. WATER MANAGEMENT ACROSS
SCALES IN THE SÃO FRANCISCO
RIVER BASIN, BRAZIL:
POLICY OPTIONS AND POVERTY
CONSEQUENCES
Marco Maneta
Marcelo Torres
Stephen A. Vosti
&
November 2008 SFRB Team UCD/Embrapa

2. Presentation Overview
• Trends, Current Conditions and Driving Forces
–P
Population, Poverty, Agriculture, Market Structure
l ti P t A i lt M k t St t
• Key Policy Issues
• Fundamental Gaps in Knowledge
– Participant Action
• Research Undertaken in the SFRB to Fill these
Gaps
• Possible Contributions of the SFRB in Phase II
– Participant Action
• Concluding Remarks (Our Personal Stories)
UCD/Embrapa

3. Three Worlds of the WDR
2008
Source: World Development Report 2008 UCD/Embrapa

4. Changing Market
Structure
UCD/Embrapa

5. Total Population 1991 2000 Change % change
Total for SFRB 14,059,006 15,723,771
1991-2000
1,664,765 11.84%
Population
Alagoas
Bahía
668,980
2,712,391
705,851
2,897,898
36,871
185,507
5.51%
6.84%
Change i the
Ch in h
Distrito Federal
Goiás
1,601,094
147,656
2,051,146
173,079
450,052
25,423
28.11%
17.22%
SFRB
Minas Gerais 6,971,994 7,885,366 913,372 13.10%
Pernambuco 1,568,446 1,591,141 22,695 1.45%
Sergipe 229,819 260,180 30,361 13.21%
Rural 1991 2000 Change % change
Population 1991-2000
Total for SFRB 4,445,920 3,691,016 -754,904 -16.98%
Alagoas
g 409,421
, 400,418
, -9,003
, -2.20%
Bahía 1,579,751 1,442,275 -137,476 -8.70%
Distrito Federal 85,205 89,647 4,442 5.21%
Goiás 50,576 37,650 -12.926 -25.56%
Minas Gerais 1,256,533 809,764 -446,769 -35.56%
Pernambuco 842,515 700,970 -141,545 -16.80%
Sergipe 124,703 131,202 6,499 5.21%
UCD/Embrapa

6. Aging of the Rural
SFRB Population
Shifting Demographic Profile
12.00
12 00
10.00
ipios
Proporti of Munici
8.00
6.00
ion
4.00
2.00
0.00
10%
11%
12%
13%
14%
15%
16%
17%
18%
19%
20%
21%
22%
23%
24%
25%
26%
27%
28%
29%
30%
31%
32%
33%
34%
35%
55+ Dependency Ratio -- 1991 55+ Dependency Ratio -- 2000
UCD/Embrapa

7. Soil Types and Distance to Market
UCD/Embrapa

8. Water Availability
UCD/Embrapa

9. Agriculture in the SFRB --
1991
UCD/Embrapa

10. Agriculture in the SFRB --
2004
São Francisco River Basin Harvested Specialty
Harvested Grains Crops,
Crops 2004
2004
Petrolina
Harvested Area (ha) Harvested Area (ha)
at Município Level at Município Level
None None
1 - 2,000 1 - 2,500
Montes Claros 2,001 - 5,000 2,501 - 10,000
5,001 - 10,000 10,001 - 50,000
10,001 - 20,000 50,001 - 200,000
20,001 - 431,441 200,001 - 401,980
Sete Lagoas
g Scale 1:9,000,000
Ribeirao das Neves 0 50 100 200 300
Belo Horizonte
Betim Kilometers
Divinopolis
Lambert Azimuthal Equal Area
Projection, WGS-84.
Map by J A Young, 12 September 2007
UCD/Embrapa

11. Frequency Distribution of Corn Production
60
(in tons/ha)
Changes in Land
Productivity
P d ti it
40
ount
Co
20
0
1 2 3 4
Unweighted Yields in 1991
Frequency Distribution of Corn Production
(in tons/ha)
60
Count
40
20
Scale of Farming is Changing Rapidly
g g g p y
• Vast Majority of Area Expansion is by
0
0 2 4
Unweighted Yields in 2004
6
Large-Scale Enterprises
UCD/Embrapa

12. Water Productivity in Corn
UCD/Embrapa

13. Water Productivity
UCD/Embrapa

14. Poverty in the SFRB
UCD/Embrapa

15. Extreme Poverty in
the SFRB
2003 Rural Extreme Poverty-
Poverty
Poverty Rural Poverty Rural
% of % of
Rural Rural
Absolute Pop Absolute Pop
Total for SFRB 1,012,095 28% 345,677 9%
Alagoas 163,307 41% 70,400 2.6%
Bahía 328,313 23% 139,941 9.7%
Goiás 7,792
7 792 21% 2,846
2 846 7.5%
7 5%
Minas Gerais 178,006 22% 43,214 5.3%
Pernambuco 258,004 37% 53,484 7.6%
Sergipe 48,635 37% 19,603 14.9%
UCD/Embrapa

16. Spatial
Distribution f
Di ib i of
Rural Povertyy
UCD/Embrapa

17. Key Policy Issues
• Agricultural Sector
– How much surface water should be diverted for agriculture, and where*?
– How much groundwater should be pumped?
– What is the optimal level of irrigation efficiency?
– What public policy action (if any) is required to better manage water resources?
• What are the effects of water management policies on the poor?
• Poverty
– How is water productivity or access to water linked to poverty in the SFRB?
– If linked, how much water should be diverted to poor farmers to reduce poverty?
• What additional public policy action will be required to reduce poverty?
• Inter-Sectoral Trade-Offs
– Wh t are th i
What the impacts on agriculture of the di
t i lt f th diversion of water f h d power?
i f t for hydro ?
– How much water should remain in the river system for environmental benefits?
• Inter-Basin Trade-Offs
– What are the agricultural and other costs in the SFRB associated with inter-basin
transfers?
* ‘and where’ applies to all issues UCD/Embrapa

18. Fundamental Gaps in
Knowledge
• Farmer Responses to Policy and Other Changes
p y g
– Water policies (e.g., water prices, regulations, etc.)
– Market conditions (e.g., input and output p
( g, p p prices) )
– Weather conditions (e.g., drought)
• Effects of Farmer Behavior on Water Resources
– Surface water
– Groundwater
UCD/Embrapa

19. Pause for Discussion
• Do These Situations or Trends ‘Ring True
Ring True’
for your Basins?
• Do the Fundamental Gaps in Knowledge
Reflect those in your Basins?
UCD/Embrapa

20. SFRB Team Activities
• Research at Three Spatial Extents – Basin-Wide,
Buriti Vermelho Sub-Catchment, Plot Levels
– Characterization
• Poverty
• Hydrology
• A i lt
Agriculture
– Water use in agriculture
• Water productivity
– Modeling
• Hydro inter-relationships
• Human behavior in agricultural
• Linking models
– Use Models to Assess the Effect of Selected Interventions
and Policy Changes
• T i i and Capacity Strengthening
Training d C it St th i
• Outreach
UCD/Embrapa

21. Key Objectives of Hydro-
Economic Models
• Understand Farmer Behavior and Outcomes
– Cropping patterns, input mix, employment, water use
– Income and poverty
– Surface water and groundwater availability
• Predict the Effects of Proposed Policy and other
Changes on Farmer Behavior/Outcomes
• Inform Policy
• Modeling at Three Spatial Extents
– Plot-Level LUS Models
– Buriti Vermelho Models
– Basin-Wide Models
UCD/Embrapa

22. Basic Components of
Hydro-Economic Models
• Hydrologic Models
– Water flows/stocks, in space/time
• E
Economic Models of Agriculture
i M d l f A i lt
– Farming decisions
•C
Crop mix, production technology, water use
i i
• Linking the Models
UCD/Embrapa

23. Burití Vermelho
Brazil San Francisco
River Basin
UCD/Embrapa

24. A Spatially Distributed Hydrologic
Model for Buriti Vermelho
UCD/Embrapa

25. Hydrologic Model
(MOD HMS)
(MOD-HMS)
Overland Channel
flow flow
∂ hO ∂ ⎛ ∂h ⎞
− ⎜ dk x O ⎟ ∂ hC ∂ ⎛ ∂ hC ⎞
∂t ∂x ⎝ ∂x ⎠ B − ⎜κl ⎟
∂t ∂ ⎝ ∂ ⎠
∂ ⎛ ∂h ⎞
+ ⎜ dk y O ⎟ − dq og − dq oc = 0 − Aq co − Aq cg = 0
∂y ⎜
⎝ ∂y ⎟
⎠
∂ ⎛ ∂h ⎞ ∂ ⎛ ∂h ⎞
⎜ K xx k rw G ⎟ + ⎜ K yy k rw G ⎟
∂x ⎝ ∂x ⎠ ∂y ⎜
⎝ ∂y ⎟
⎠
Flow in
porous media
∂ ⎛ ∂h ⎞
+ ⎜ K zz k rw G ⎟ − W + q go + q gc with
∂z ⎝ ∂z ⎠
variable
Simulation ∂S w ∂h
=φ + SwSs G degrees of
∂t ∂t saturation
Panday & Huyakorn (2005). Adv. Wat. Res.
UCD/Embrapa

26. Changes in Reservoir Demand
Reservoir #2
Reservoir #5
UCD/Embrapa

27. Groundwater & Surface
Water Use (Farmer 4)
( )
Applied Water
By Source
Baseline
Depth to Water
Drought
Table
UCD/Embrapa

28. Basin-Wide Hydrology Model
Upscale
UCD/Embrapa

29. Monte Carlo simulations using
ARCGIS – Excel and VBA
UCD/Embrapa

30. Hydrologic Modeling
Conclusions
C l i
• Physics-based models capture spatial and
Physics based
temporal impacts of economic activity on the
hydrologic system
• Give insights and enhance understanding of the
biophysical system
• For the larger scale, stochastic techniques can
calculate water availability in terms of frequency
and permits quantification of risk
• Dynamic models have predictive capability and
therefore allow f policy testing
th f ll for li t ti
UCD/Embrapa

31. Core of the Economic Models of
Agriculture: Farmer Objective Function
• Maximize Profits
– Choose product mix and production technology
• Including the amount and sources of water, and how it is
applied
• Subject to an Array of Constraints
j y
– Socioeconomic
• Feed the family
• Access to markets and credit, etc.
– Biophysical
• Soils, weather, etc.
–A
Access t water
to t
• Surface water, groundwater
UCD/Embrapa

32. Core of the Economic Model of
Agriculture: Farmer Objective Function
max ∑ pit qit (x nirrt , ewit (xirrt )) − ∑ w jt xijt − ∑ cewit (pirrt , xirrt ; z)
i i i
Agricultural Production Function Effective Water
•Vector of Non Irrigation Inputs (xnirr):
Vector Non-Irrigation Cost
Crop •Fertilizers, seeds, land, pesticides, Non-Irrigation • Irrigation Input
Prices machinery etc Input Cost Prices – pirr
•Effective Water – ew • Price - wsj • Irrigation Input
•Function of Irrigation Inputs (xirr)
F ti f I i ti I t ( ): •Q
Quantity - xsij
tit Quantities - xirr
Q titi
•Applied water • z – Vector of
•Irrigation Capital factors that may
•Irrigation Labor
g affect irrigation costs
•Irrigation Energy (e.g. distance to
( di t t
river)
UCD/Embrapa

33. Hydrologic & Economic Model Links
y g
• Crop-specific Algorithm to translate
g HYDROLOGIC
• poduction cropping decisions into MODEL
• water use water demand
• irrigation efficiency
Cropping Decisions Hydrologic Consequences
ECONOMIC • Water available for ag
Algorithm to translate
O
MODEL hydrologic consequences • rainfall
into water availability •surface water
UCD/Embrapa

34. Land Use System Analysis
(LUS)
• Space
– Single parcel of land
g p
• Time
– Multi-year duration, specific end date, seasonal time steps
• Economic Model of Agriculture
– S ifi series of cropping activities, specific production and
Specific i f i ti iti ifi d ti d
water use technologies
• Hydrology Model
– Farmer’s assessments of water availability
y
• All Data Collected at Farm Level
Field #1
Field #1
Year 1 Field #1
Year 2 Field #1
Year 3
Year 4 Field #1
Year 10 Field #1
Year 15
UCD/Embrapa

35. LUS Results for Alternative
Production Systems in Petrolina
Labor Employ
LUS Economic Performance Requirements Water for Irrigation ment
Establish
Excess ment
Total Establish Water
Returns Cost -- Opera- Operatio
NPV per Returns Establish Family ment Productivi
NPV to Plot tional Water Use nal
hectare to Land ment Labor Cost -- ty (NPV/
Family
y Costs Phase
Used Property 1000m³)
Labor (per
hectare)
$R/ Person- person-
person- $R/ha Person- days/ ha/ $R/ha/ 1000M3/ $R/ days/ha/
$R $R/ha day /year days /ha year $R $R/ ha year ha/year 1000m³ year
Goats and Sheep -12 0 0 0 1.5 6.3 0 0 6 4 0.00 0
Melon -Onion 43,963 21,981 11 1,099 28 102 50 25 2,466 21 53.26 229
Manga -- flood
irrigation 3,087 772 1 39 35 45 553 138 1,177 12 3.12 93
Mango -- micro
sprinkler 11,057 2,764 4 138 44 32 4,212 1,053 973 10 14 69
Table grapes
with seeds 778,074 129,679 31.14 6,484 151 208 96,600 16,100 3,157 18 368 524
Table grapes
seedless 1,369,349 228,225
1 369 349 228 225 54.81
54 81 11,411
11 411 151 208 96,600
96 600 16,100
16 100 3,157
3 157 18 648 438
UCD/Embrapa

36. Effects of Uncertainty
Effects f G t M t lit
Eff t of Goat Mortality
Uncertainty on NPV per Year
Effect of Uncertainty in Mango Prices
on NPV per year
UCD/Embrapa

37. Policy E
P li Experiments Using LUS
i t Ui
UCD/Embrapa

38. Modeling the Buriti
Vermelho Sub Catchment
Sub-Catchment
Benefits of Co-Location
of Research Sites
Brazil San Francisco
River Basin
UCD/Embrapa

39. Water Stored and Withdrawn from Reservoir #2
Water
Availability
A il bili
and Use in BV
% Groundwater Used by Farmer X
Depth of Groundwater in Well Field
UCD/Embrapa

40. Economic Effects of Drought
corn dried beans limes vegetables orchards wheat
Change with respect to
30
baseline (%)
20
Changes
e
10 Farmer 1
0 Farmer 2
Farmer 3
in Applied
-10
Farmer 4 Water
20
-20
-30
(a)
-40
Change with respect to
Changes
t
40
20
in Land
aseline (%)
Farmer 1
0 Farmer 2
-20 Farmer 3 Allocation
ba
-40 Farmer 4
-60
-80 (a)
-100 UCD/Embrapa

41. Change with respect to
eline (%)
o
Economic Effects of Drought
20
Farmer 1
0
base
w
Farmer 2
F
-20 Farmer 3 Changes
-40 Farmer 4 in Profits
-60
-80 (a)
-100
Change with respect to
.
Changes
40
in Hired
%)
baseline (%
Farmer 1
20
0
Farmer 2
Labor
Farmer 3
-20 Farmer 4 Use
-40
40
-60
-80 (a)
-100 UCD/Embrapa

42. Basin-Wide Setting
• Variable Weather Conditions
– Wet year and drought
– Rainfall and evapotranspiration
• Water Policy Setting
– Application of the ANA guidelines
• Price Shock
– Large increase in sugarcane prices
• Use Hydro-Econ Models to Predict:
– Cropping p
pp g patterns, water use, employment, income
, , p y ,
– Water availability in river system
UCD/Embrapa

43. Precipitation
P i it ti
in the SFRB
and Focus of
the Basin-
Wide Policy
Experiment
UCD/Embrapa

44. Upstream Water Demand
Upstream Water Demand for Boqueirão
(sample município)
Blue = baseline
Green = Sugarcane Price Increase
Total Demand of all Simulated
Upstream Responses to
Sugarcane Price Increases (m3s‐1)
January y 39.5
February 33.4
March 40.1
April 22.3
Mayy 27.1
June 37.8
July 54.4
August 89.5
September 99.4
October 92.5
November 74.6
December 43.1
UCD/Embrapa

45. Water Available at Sobradinho Dam ‐‐ Before Price Shock
Water
Available for
Agriculture
Water Available at Sobradinho Dam ‐‐ After Price Shock
“Available” for Ag =
River Flow Entering
Sobradinho Dam Minus
2000 m3s-1 for
Environmental Flows
(following Braga and Lotufo
2008)
UCD/Embrapa

46. Upstream Cultivated Areas
(by scenario, irrigation)
500,000
400,000 Agricultural
300,000
200,000 Land Use
L dU
100,000
0
Baseline Sugar Price -- Sugar Price -- Wet
Drought Year
Downstream Cultivated Areas
Rainfed Irrigated Total Cultivated Area (by scenario, irrigation)
900,000
800,000
700,000
600,000
500,000
500 000
400,000
300,000
200,000
100,000
0
Baseline Sugar Price -- Sugar Price --
Drought Wet Year
Rainfed Irrigated Total Cultivated Area
UCD/Embrapa

47. Area in Sugarcane
Upstream Sugarcane Areas
(by scenario, irrigation)
30,000
30 000
25,000
20,000
15,000
10,000
5,000
5 000
0
Baseline Sugar Price -- Sugar Price --
Drought Wet Year
Downstream Sugarcane Areas
(by scenario, irrigation)
Total Sugarcane Total Irrigated Sugarcane
50,000
40,000
30,000
20,000
10,000
0
Baseline Sugar Price -- Sugar Price --
Drought Wet Year
Total Sugarcane Total Irrigated Sugarcane
UCD/Embrapa

48. Upstream Agricultural Employment
(by scenario, irrigation) R l
Rural
6,000
5,000
,
4,000
Employment
p y
3,000
2,000
1,000
0
Baseline Sugar Price -- Sugar Price --
Drought Wet Year
Downstream Agricultural Employment
Total Rural Employment
p y Total Irrigated Ag Employment
g g p y ( y
(by scenario, irrigation)
, g )
50,000
40,000
30,000
20,000
10,000
0
Baseline Sugar Price -- Sugar Price --
Drought Wet Year
Total Rural Employment Total Irrigated Ag Employment
UCD/Embrapa

49. Upstream Sugarcane and Total Ag Profits
(by scenario, irrigation)
120,000,000
100,000,000
80,000,000
60,000,000
40,000,000
40 000 000
Agricultural
20,000,000
0
Baseline Sugar Price -- Sugar Price --
Profits
Drought Wet Year
Total Ag Profits Irrigated Ag Profits
Total Sugarcane Profits Irrigated Sugarcane Profits
Downstream Sugarcane and Total Ag Profits
(by scenario irrigation)
scenario,
300,000,000
250,000,000
200,000,000
150,000,000
100,000,000
50,000,000
0
Baseline Sugar Price --
S gar Sugar Price -- Wet
S gar
Drought Year
Total Ag Profits Irrigated Ag Profits
Total Sugarcane Profits Irrigated Sugarcane Profits
UCD/Embrapa

50. Behavioral Modeling
• Value Added
– Insight into farming and farm household decisions
– These decisions can affect water use
– Insights into water-poverty links
• Practicality
– Array of tools available
• Static models (LUS)
• Equilibrium models (PMP)
• Agent-based models
• Others
– Linking hydro and behavioral models can be
challenging
• Depending on circumstances, it can be worth the effort
UCD/Embrapa

51. Knowledge Pathways and
Impact Pathways
Min. of Ag. Agents of Change
Embrapa CPWF
Embrapa
ANA Research Centers BFPPolicy
Central
Policy
y g
Water Mgmt. Policy
y
World Bank
W ld B k ‘Placeholders’ ‘Placeholders’ ‘Placeholders’
‘Pl h ld ’
Policy
‘Placeholders’
IEB -- NGOs
Policy
‘Placeholders’
Placeholders
Embrapa HQ
Policy Core Team IPEA
‘Placeholders’ •Research Policy
•Training
g ‘Placeholders’
•Outreach
Research Collaborators
Brazilian Embrapa Research NGOs Farmers International
Universities Centers • Cooperatives Research
• Brasilia • Savannah Community y
• Petrolina •Coastal zone
• Minas Gerais •Corn and sorghum
• Ceara •Semi-arid
UCD/Embrapa

52. Research Outputs
• Written Output
– Journal papers conference papers working
papers, papers,
papers, posters, etc.
– Policy Briefs (Portuguese and English)
• Methodologies
– Linked, hydro-economic models
– LUS models
• Human Capital
– Embrapa, UC Davis, U. of Brasilia
• Data Sets
– A i lt
Agriculture, water resources, poverty
t t
UCD/Embrapa

53. Next Steps for the SFRB
Research Team
• Refine and Use the Models to Address Pressing
Water-Ag-Poverty Issues in Brazil
• Deliver these Messages to Decision Makers
• Contribute to CPWF Research/Training Efforts
• Convey Models and Data to Collaborators
• Publish our Findings
g
– Journal papers
– Book on our multi scale effort in the SFRB
multi-scale
UCD/Embrapa

54. SFRB Potential Contributions
g
to Phase II Basin Challenges
• Benefit Sharing Mechanisms
– Sharing water versus sharing the benefits of water
• Adaptive Management
d p ve ge e
– Objectives? – rural poverty alleviation, managing environmental flows,
etc.
– What are we reacting to? – weather, climate change, market conditions
• Improved Livelihoods
– Which stakeholders, by how much?
– Uncertainty and risk
• Th I t
The Integrated M
t d Management of P d ti S t
t f Production Systems B d on
Based
Groundwater
– Surface water/groundwater interactions
• Improved Planning and Management of Hydroelectric Facilities
– Long-term management with variable rainfall
– Effects of agricultural change
• Developing and Maintaining Sustainable Small Reservoirs
– Volume, placement and management
UCD/Embrapa

55. Pause for Discussion
• How Have We Done?
• What Have We Missed?
• What Would YOU Like to See the SFRB
O i S S
Team Contribute to the CPWF?
UCD/Embrapa

56. Concluding Remarks
• Our Stories
– Steve Vosti and Marcelo Torres
– Marco Maneta
• Your Views
UCD/Embrapa

57. Muito Obrigado!
UCD/Embrapa