Basin Impacts of Irrigation
Water Conservation
University of California
Department of Environmental Sciences
Riverside
Frank A. Ward (NM State University)
February 25, 2011
1
Background
• Climate Change: more floods/droughts
• Continued Population Growth (esp poor countries)
• Growing values reduced supplies of ecological assets
• Growing values of treated urban water
• Search for ways to conserve water in irrigated agriculture
• Special search for ag water conservation, esp if it
protects the farm economy (food security)
– technology (drip, sprinkler, water saving crops)
– policy (subsidies, regulations, pricing)
– Projects (infrastructure, leveling, … )
2
Road Map
• Pose questions
– What is water conservation in agriculture?
– What policies could promote it?
– Can river basin policy models help discover?
– Findings about effects of water conservation
incentives in the Rio Grande Basin?
– Lessons learned?
• About water conservation
• Generally
• Possibly for California
3
Basin Scale Choices
Watershed runoff
Compact Obligation
Reservoir
Fish and wildlife
Irrigated crops
Hydropower
Groundwater
Flooding
Urban water supply
Treaty obligation
1
4
Rio
Grande
Basin
5
Journey down the Rio Grande
Snow melt: 1 a-f Rio Grande Silvery Minnow
CBP pumped water Elephant Butte, Caballo
SLV Irrigation EBID Irrigation
Sangre De Cristo Headwaters El Paso urban (sw +gw)
Heron, El Vado, Abiquiu , Cochiti West TX Irrigation
Albuquerque urban (sw + gw) Mexico Ag
MRGCD Irrigation Mexico Urban
6
High Valued Uses of Water in
RGB, Albuquerque, El Paso
7
High Valued Use: Rio Grande
Silvery Minnow
8
High Uses of Water in RGB, Irrigation
9
Approach
10
Policy Debates
Basin Models Can Inform
• Water Pricing and Cost Recovery
• Timing, sizing, sequencing of new storage
• Population growth, increased food demands, ‘more crop
per drop.’
• Water rights adjudication
• Meeting growing demands for environment
• How to develop/allocate water for food security
• Cheapest way to reduce water use (conservation)
11
Basin Models:
The Dark Side
• Too academic, too theoretical, too little use to inform
real policy debates
• Nobody understands them
• Models are hungry for data that aren’t there.
• Expensive and slow to build
• Who wants to work with a bunch of academics with
uncertain use of results?
12
Basin Model (Optimization)
• Maximize
– Objective
• Economic
• Environmental
• Social Justice
• Hydrologic
• Subject to
– Constraints
• Hydrologic
• Agronomic
• Institutional
13
• Economic
GAMS Basin Model Structure
SETS
H: time, reservoirs, diversion locations, headwater flow locations, aquifers,
U: cities, income levels …; A: irrigated areas, crops…; E: assets, services
DATA
prices, costs, population, compact delivery requirements,
elasticities, acres available, headwater flows…
(DEPENDENT) VARIABLES
diversions, use, return flows, acres in production,
pumping, prices, reservoir levels, NPV…
EQUATIONS
objective functions and constraints
SOLVER
14
Policy Assessment Framework
Data Policy Process Outcomes
Headwater Baseline: no Crop prodn
supplies new policy Crop ET
Min Flows Alt 1: Urban water
Sharing rules Constrain diversions, use,
Maximize
Outflows aquifers to Return flows,
NPV for the
return to start Flows by gauge
basin
Crop prices
Crop costs Alt 2: Urban, farm,
Water price Renew environmental
Treat cost aquifers to benefits
Elasticities historical
Land supply levels NPV
15
Connections
• Connections: River basin models
– Hydrologic: stocks, flows, over time, space
– Economic: optimizes total benefits from use
– Agronomic: acreage, water use, crops
– Demographic: urban income, population, demand
– Institutional: rules that limit use or require delivery
• Use connections to gain insights for policies that
best adapt to climate: resilient conservation
institutions
– For basin as a whole
– For targeted users (farm, city, environment)
16
Aquifer mass balance
Seepage to Pumping from
Aquifer Aquifer
Stream to
Aquifer
groundwater inflow 390 440 Aquifer to
80
Stream
Return flows 70
220
Inflow – Outflow = Change in Storage
17
Reservoir mass balance
Precipitation on Reservoir
Evaporation
Upstream 390 440
inflow 80
70
220
Reservoir Release
Inflow – Outflow = Change in Storage
18
Water Balance
19
Crop Water Use Data, RG Basin, NM
Yield Yield
A ET DP tons/ac A ET DP tons/ac
Crop Tech ac-ft/ac/yr Tech ac-ft/ac/yr
Alfalfa f 5.0 2.2 2.9 8.0 d 2.7 2.7 0.0 10.0
Cotton f 2.8 1.2 1.6 0.4 d 1.5 1.5 0.0 0.5
Lettuce f 2.5 1.1 1.4 12.5 d 1.4 1.4 0.0 15.6
Onions f 4.0 2.3 1.7 16.9 d 2.9 2.9 0.0 21.1
Sorghum f 2.0 0.9 1.1 2.0 d 1.1 1.1 0.0 2.5
Wheat f 2.5 1.1 1.4 4.6 d 1.4 1.4 0.0 5.8
Green Chile f 4.6 2.0 2.6 11.0 d 2.5 2.5 0.0 13.8
Red Chile f 5.0 2.2 2.9 1.7 d 2.7 2.7 0.0 2.2
Pecans f 6.0 2.6 3.4 0.6 d 3.2 3.2 0.0 0.7
20
NM Pecans: Water Balance
Flood Drip
6’
3.2’ 3.2’
2.6’
0
3.4’ 0
21
Under the Hood
22
Objective
NBuut NBeet
Max NPV (1 r )t
t (1 r )
t
u u e t e
NBAuckt
NPV Ag
t (1 r )
t
u c k u
NBAuckt [ Pct Yield uckt Cost uckt ] Luckt
NBut (e.g , urban), NBet (e.g., wetlands)
23
Constraints
• Irrigable land, Headwater supplies
• Sustain key ecological assets
• Hydrologic balance
• Reservoir starting levels (sw, gw)
• Reservoir sustainability constraints (sw, gw)
• Institutional
– Endangered Species Act
– Rio Grande Compact (CO-NM; NM-TX)
– US Mexico Treaty of 1906
– Rio Grande Project water sharing history (NM/TX)
24
Gauged Flows: Hydro Balance
X vt B
h
hv X ht Bvv X vt Bdv X dt
v d
Brv X rt B Lv X Lt
r L
• E.g.: Lobatos gauge (CO-NM border):
X(Lobatos_v,1) = X(RG_h,1) - X(SLV_d,1) + X(SLV_r,1)
25
Ag water use
X ut B
c k
uck Luckt
u irrigated region
c crop
k irrigation tech ( flood , drip , pivot ,...)
26
Reservoir Stocks
Z rt Z rt 1 X Lt
27
Institutions: e.g. Rio Grande
Compact
X vt Lobatos
B0 h B1h X ht RG _ h
X vt SA
C0 h C1h X vt Otowi
28
Potential Institutional Constraints
• U.S. Mexico Groundwater Sharing Treaty
• U.S. Mexico Water Quality Treaty
• Limiting domestic well development
• Adjudicate MRG water rights
29
Results
• Ag Water Use and Savings
– Status Quo
– Sustain Natural Capital
– Renew Natural Capital
30
Water Use by Technology and Policy
LRGB (AF/yr, ET)
Alternative 1: Alternative 2:
Base Sustaining Renewing Natural
Tech Units Natural Capital Capital
use use change use change
absolute 146,266 94,917 -51,349 94,375 -51,891
Flood
pct 100 65 -35 65 -35
absolute 52,604 4,402 -48,202 1 -52,602
Drip
pct 100 8 -92 0 -100
absolute 198,869 99,318 -99,551 94,376 -104,493
Total 31
pct 100 50 -50 47 -53
Lessons Learned:
Water Conservation
• Farmers seek income, not conservation.
Conservation must be profitable for irrigators to do it.
– Subsidizing water conserving irrigation technology will
reduce water applied per unit land for a given crop
– But if a water right is for total water applied to a farm
• Acreage may increase to maintain total water applied
• Crop mix may change to maintain total water applied
– Reduced water applied doesn’t mean reduced water
depleted by the crop.
– Requiring sustainable reservoirs and aquifers in NM
reduces the use of drip irrigation.
32
Lessons Learned:
Research Challenges
• Water conservation is hard to define, measure,
forecast, evaluate, alter.
• Counterfactual: How much less water would have
been (will be) used if X irrigation technology would
have been (is) subsidized.
• River basin models are fun to build and write about, if
you start small and grow them
• Check that your model re-produces what you publish.
• Mathematically document model, data, assumptions.
• Calculate sensitivities: Yi
a j 33
Lessons learned for California:
“California Water Myths”
• California is running out of water.
• ________ is responsible for California’s water problems.
• We can build our way out of California’s water problems.
• We can conserve our way out of California’s water problems.
– Effectiveness of conservation is often overstated.
– Principle: Look for cheapest ways to reduce use.
– Practice: Requires defining use, comparing B, C of saving.
• Water markets can solve California’s water problems
• Healthy aquatic ecosystems conflict with a healthy economy.
• More water will lead to healthy fish populations.
• California’s water laws impede sustainable management.
34
Top 10 Lies told by Watershed Policy Modelers
1. The model is well-documented with all limits
2. The model is user-friendly
3. The model fits the data
4. Results make sense
5. The model does that
6. We did a sensitivity analysis
7. Anyone can run this model
8. This model links to other models
9. The model will be in the public domain
10. The new version fixes all previous problems
35