UCD/Embrapa

Assessing the Effects of Alternative

Policies and Conditions in the São

Francisco River Basin, Brazil:

A Multi-Scale Approach

Marco Maneta

Stephen A. Vosti

&

SFRB Team

�ovember 2008

Center for Natural

Resources Policy

Analysis -- CNRPA

UCD/Embrapa

• Objectives of Modeling Exercises

• Basic Components of Predictive Models

– Hydrologic Models

– Economic Models of Agriculture

• Examples of Model Uses

– Plot Level

– Sub-catchment Spatial Extent

– Basin Spatial Extent

• Conclusions and Policy Implications

Presentation Overview

UCD/Embrapa

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

Basic Components of

Predictive Models

• Hydrology Models

– Plot level

– Sub-catchment spatial extent

– Basin spatial extent

• Economic Models of Agriculture

– Plot level

– Sub-catchment spatial extent

– Basin spatial extent

UCD/Embrapa

Core of the Economic Model of

Agriculture: Farmer Objective Function

Crop

Prices

Agricultural Production Function•Vector of 4on-Irrigation Inputs (x

nirr):

•Fertilizers, seeds, land, pesticides,

machinery etc

•Effective Water – ew

•Function of Irrigation Inputs (xirr

):

•Applied water

•Irrigation Capital

•Irrigation Labor

•Irrigation Energy

4on-Irrigation

Input Cost• Price - wsj

• Quantity - xsij

Effective Water

Cost• Irrigation Input

Prices – pirr

• Irrigation InputQuantities - x

irr

• z – Vector of factors that may

affect irrigation costs(e.g. distance to

river)

)z;,())(,(maxirrtirrt

i i i

ewijtjtirrtitnirrtitit itcxwewqp xpxx∑ ∑ ∑−−

UCD/Embrapa

Hydrologic & Economic Model Links

HYDROLOGIC

MODEL

• Crop-specific

• poduction

• water use

• irrigation efficiency

ECO4OMIC

MODEL•Water available for ag

• rainfall

•surface water

Algorithm to translate

cropping decisions into

water demand

Algorithm to translate

hydrologic consequences

into water availability

Cropping Decisions Hydrologic Consequences

UCD/Embrapa

Land Use System (LUS) Analysis• Space

– Single parcel of land• Time

– Multi-year duration, specific end date, seasonal time steps• Economic Model of Agriculture

– Specific series of cropping activities, specific production and water use technologies

• Hydrology Model– Farmer’s assessments of water availability

• All Data Collected at Farm Level

Field #1

Year 1Field #1

Year 2 Field #1

Year 3 Field #1

Year 4 Field #1

Year 10 Field #1

Year 15

UCD/Embrapa

LUS Results for Alternative

Production Systems in Petrolina

LUS Economic Performance

Labor

Requirements Water for Irrigation

Employ

ment

4PV4PV per

hectare

Excess

Returns

to

Family

Labor

Returns

to Land

Establish

ment

Total

Family

Labor

Used

Establish

ment

Cost --

Property

Establish

ment

Cost --

Plot

(per

hectare)

Opera-

tional

Costs

Water Use

Water

Productivi

ty (4PV/

1000m³)

Operatio

nal

Phase

$R $R/ha

$R/

person-

day

$R/ha

/year

Person-

days /ha

Person-

days/ ha/

year $R $R/ ha

$R/ha/

year

1000M3/

ha/year

$R/

1000m³

person-

days/ha/

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 54.81 11,411 151 208 96,600 16,100 3,157 18 648 438

UCD/Embrapa

Policy Experiments Using LUS

UCD/Embrapa

Effects of Uncertainty

Effects of Goat Mortality

Uncertainty on NPV per Year

Effect of Uncertainty in Mango Prices

on 4PV per year

UCD/Embrapa

A Spatially Distributed Hydrologic

Model for Buriti Vermelho

UCD/Embrapa

Modeling the Buriti Vermelho

Sub-Catchment

San Francisco River Basin

Brazil

UCD/Embrapa

Water

Availability

and Use

UCD/Embrapa

Economic Effects of DroughtChanges in Applied Water

Changes in Hired Labor UseChanges in Profits

Changes in Land Allocation

UCD/Embrapa

• Variable Weather Conditions

– Wet year and drought

– Rainfall and evapotranspiration

• Water Policy Setting

– Application of the A4A guidelines

• Price Shock

– Large increase in sugarcane prices

• Use Hydro-Econ Models to Predict:

– Cropping patterns, water use, employment, income

– Water availability in river system

Setting the Policy

Experiment Stage

UCD/Embrapa

A Basin-Wide Hydrology Model

Barreiras

Paracatu

Rio Paranaiba

Petrolina

UCD/Embrapa

Water

Available for

Agriculture

Water Available at the Entrance to Sobradinho Dam

Wet-Year Water

Availability (m3s-1)

Drought-Year Water

Availability (m3s-1)

January 5477.3 2991.8

February 5471.1 2955.0

March 5718.0 2364.9

April 3130.6 1578.3

May 1724.2 681.8

June 1573.5 274.0

July 1391.7 66.9

August 919.1 10.0

September 380.7 10.0

October 621.2 10.0

4ovember 1740.4 627.7

December 3863.4 2153.5

“Available” for Ag =

River Flow Entering

Sobradinho Dam Minus

2000 m3s-1 for

Environmental Flows (following Braga and Lotufo

2008)

Water Available at the Entrance to Sobradinho Dam

UCD/Embrapa

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 39.5

February 33.4

March 40.1

April 22.3

May 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

Downstream Water

Availability after

Price Shock

Water Available at the Entrance to Sobradinho Dam

Available Water Downstream after

Sugarcane Price Increase (m3s

-1)

Wet Year Drought

January 5442 2973

February 5388 2927

March 5723 2154

April 3175 1585

May 1743 650

June 1483 222

July 1366 10

August 827 10

September 296 10

October 543 10

November 1718 574

December 3794 2016

UCD/Embrapa

Agricultural

Land Use

UCD/Embrapa

Area in Sugarcane

UCD/Embrapa

Rural

Employment

UCD/Embrapa

Agricultural

Profits

UCD/Embrapa

• Application of A4A Guidelines Will Affect Agriculture– Effects will depend on product mix, irrigation technology, location and

upstream effects, weather conditions, and product prices

• Hydro-Econ Model Can Help Predict:– The location and extent of effects on (say) profits

– Provide estimates of willingness to pay for more water• Hence, help develop water markets

• Effects of Sugarcane Price Increase on Ag– Shift in product mix

– Increased irrigated area

– Profits increase

– Upstream farmers not affected by drought; not so for downstream farmers

• Effects of Sugar Price Increase on Poverty – Bad news: little employment growth, small-scale sugarcane not likely to

participate in boom

– Good news: increased water use in sugarcane does not ‘crowd out’ crops with higher labor demand patterns

Conclusions and Policy

Implications

UCD/Embrapa

Muito Obrigado!