Economic and environmental implications of alternative landscape

Ecological Economics 38 (2001) 119– 139
www.elsevier.com/locate/ecolecon
ANALYSIS
Economic and environmental implications of alternative
landscape designs in the Walnut Creek Watershed of Iowa
Colette Coiner a, JunJie Wu a,*, Stephen Polasky b
a
Department of Agricultural and Resource Economics, Oregon State Uni6ersity, Cor6allis, OR 97331, USA
b
Department of Applied Economics, Uni6ersity of Minnesota, Minneapolis, MN 55455, USA
Received 19 July 2000; received in revised form 18 December 2000; accepted 19 December 2000
Abstract
This paper evaluates the economic and environmental impacts of three alternative landscape scenarios created by
a team of landscape architects, following input from an interdisciplinary team of researchers. In the first scenario, the
main objective was to increase production and profitability of commercial agriculture with environmental objectives
given secondary weight. In the second scenario, water quality improvements were the main objective with secondary
objectives being financial health of the agricultural sector and maintenance and restoration of biodiversity. In the
third scenario, maintenance and restoration of native biodiversity was the main objective with secondary weight given
to the financial health of the agricultural sector and water quality. We evaluate the degree to which the economic and
environmental objectives can be achieved together or involve tradeoffs. We found that some changes in land use or
agricultural practices result in environmental improvements on certain dimensions in addition to making economic
sense. But most changes in land use or agricultural practice do not bring uniform environmental improvement. There
may be difficult tradeoffs between different components of environmental quality in addition to tradeoffs between
economic and environmental objectives. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Biodiversity; Environmental quality; Farm profit; Landscape design; Water quality
1. Introduction
Land use and land management decisions in
agricultural systems not only produce marketed
agricultural goods but also change the level of
* Corresponding author. Tel.: + 1-541-7373060; fax: + 1541-7372563.
E-mail address: [email protected] (J. Wu).
nonmarketed ecosystem services as well. Choosing
among land use/land management options to produce desired combinations of agricultural goods
and ecosystem services requires integrating input
from a number of different disciplines including
agronomy, hydrology, ecology and economics. To
study land use/land management issues in the
context of Midwest agricultural watersheds, an
interdisciplinary team of researchers was assem-
0921-8009/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 1 - 8 0 0 9 ( 0 1 ) 0 0 1 4 7 - 1
120
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
bled to investigate water quality, biodiversity and
returns to farmers for sample watersheds. This
paper reports on results generated by this interdisciplinary investigation for environmental and economic returns for the Walnut Creek Watershed
southwest of Ames, Iowa, USA Walnut Creek is a
fairly typical watershed in the Des Moines Lobe
region of Iowa, which is characterized by flat
terrain and highly productive soils.
The task for the interdisciplinary team was to
investigate what might be the economic and environmental consequences of alternative futures for
the watershed. Three landscape scenarios were
created by a team of landscape architects to capture essential features of three quite different future alternatives for the watershed, following
input from the interdisciplinary team of researchers (for details, see Nassauer et al., 1999).
In each of the three scenarios priority was given
to different objectives. In the first scenario, the
main objective was to increase production and
profitability of commercial agriculture with environmental objectives given secondary weight. In
the second scenario, water quality improvements
were the main objective with secondary objectives
being financial health of the agricultural sector
and maintenance and restoration of biodiversity.
In the third scenario, maintenance and restoration
of native biodiversity was the main objective with
secondary weight given to the financial health of
the agricultural sector and water quality. The
three scenarios are described in greater detail in
Section 2.
In this paper, we examine the economic and
environmental implications of the three alternative scenarios compared to a status quo baseline
(current farm practices and land use) in the Walnut Creek Watershed. The environmental effects
we analyze are nitrate– N (NO3 −N) runoff and
leaching, wind and water erosion. The economic
effect we analyze is total return to land (total
revenue minus total cost except land cost). To
estimate return to land under various alternative
land use/crop management practices, we first estimate yields for various crops under the alternatives. Combining estimated yields with output
prices and production costs then determines total
return to land. Both yield and environmental
effects are derived using a computer simulation
model known as the Interactive Environmental
Policy Integrated Climate (i – EPIC; formerly the
Erosion
Productivity
Impact
Calculator)
(Williams et al., 1988; Sharpley and Williams,
1990). We describe the simulation model and data
in Section 3.
The end result of the analysis shows how economic and environmental objectives may be affected by choosing alternative scenarios, which
represent different possible futures for the watershed. The research also sheds light on the degree
to which economic and environmental objectives
are complementary or involve tradeoffs. The results of the analysis are described in Section 4.
Most prior related research has focused on the
impact of specific agricultural practices on water
quality (Hallberg, 1989; Gren, 1993; Kronvang et
al., 1995; Gren et al., 1997; Byström, 1998). In
particular, the impact of tillage practice on farm
income and water quality has received a lot of
attention in the literature. Conventional tillage
systems using a disc to turn over soil have advantages for weed control, disruption of pest life
cycles, and breaking up of soil layers that may
impede water filtration and plant growth. Disadvantages of conventional tillage include higher
soil erosion rates, loss of soil porosity, and soil
compaction. The National Research Council
(1989) estimated that no-till systems could reduce
soil erosion by as much as 94%. In addition,
Martin et al. (1991) found that overall no-till is
cheaper then conventional tillage. Wiersink et al.
(1992) found no-till had a higher variable cost,
but conventional tillage had a higher fixed cost.
The increased variable costs are associated with
increased herbicide use for weed control in no-till
(Martin et al., 1991).
With the move to conservation tillage practices
such as no-till, there has been concern over the
fate of nutrients in the soil and their possible
effects on water quality. The primary concern has
been focused on NO3 − N runoff and leaching
since NO3 − N is the most commonly detected
pollutant in groundwater. For example, the US
Geological Survey (USGS) found that NO3 − N
concentrations in 21% of samples collected beneath agricultural land exceeded the 10 mg/l max-
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
imum contamination level set by the US Environmental Protection Agency (Mueller et al.,
1995). Although there is a consensus that conservation tillage reduces soil and chemical
runoff, the impact of conservation tillage on
NO3 −N leaching is ambiguous. Most studies
have found that conservation tillage increases
NO3 −N leaching (Tyler and Thomas, 1979;
Thomas et al., 1981; Wu and Babcock, 1999). A
higher leaching rate may be due to the increase
in soil porosity created by microbial activity (Jacobs and Timmons, 1974; McMahon and
Thomas, 1976; Blevins et al., 1977; Thomas et
al., 1981; Blevins et al., 1983; Kitur et al., 1984).
But some observed less leaching with conservation tillage (Kanwar et al., 1985), while others
(Kitur et al., 1984) found no difference. Gilliam
and Hoyt (1987) concluded that conservation
tillage might increase or decrease NO3 −N
leaching, depending upon soil and weather conditions.
There has also been research done on the economic and environmental impacts of strip intercropping. Strip intercropping is the method of
planting two different crops in narrow strips
across the contour of the land. Many producers
experience an increase in yield around the edge
of the strips, known as the edge effect. West
and Griffith (1992) found that corn yield increases of 1255– 2511 kg/ha with strip intercropping, but soybean yield dropped by as much as
397 kg/ha. This decrease in yield was caused by
decreased sun from corn.
These studies are useful in discerning the economic and environmental impacts of particular
practices at the field-level. Reflecting the increased awareness of the scope of nonpointsource water pollution, several national
inventories have been conducted to determine
the status and trend in NO3 −N concentrations
in groundwater or surface water (Smith et al.,
1987; Mueller et al., 1995). Others (Nielsen and
Lee, 1987; Kellogg et al., 1992; Wu et al., 1997;
Wu and Babcock, 1999) have attempted to identify the spatial patterns of high-risk areas by
conducting national or regional assessments of
water contamination potential from agricultural
chemical use.
121
Rather than looking at individual practices,
we analyzed the total integrated effect of
changes in both land uses and management
practices. We analyzed the economic and environmental impacts of alternative landscape scenarios, which were created by integrating
knowledge from the interdisciplinary team to illustrate alternative futures for the watershed.
We found examples where economic and certain
environmental outcomes were complementary in
the sense that certain management practices improved both economic and environmental performance. For example, adopting conservation
tillage increased return to land and reduced soil
erosion and NO3 − N runoff. In other cases,
there were tradeoffs between improving different
economic or environmental measures. Even
holding economic benefit constant, there were
tradeoffs between different environmental objectives.
2. Study area and alternative landscape scenarios
The Walnut Creek Watershed is located in
Boone and Story counties southwest of Ames,
Iowa. Rainfall for this area averages : 821 mm/
yr. The soil texture for this watershed ranges
from silt loams to clay loams. Silt loams are
some of the best soils in which to grow crops
because they are easily worked. But it is this
factor that also makes the land highly erodible
and susceptible to chemical leaching. The topography of the watershed tends to be fairly flat
with slopes ranging from 0% to 9%.
Within the Walnut Creek Watershed several
types of farming methods are currently used.
These methods range from conventional tillage
methods to conservation tillage to strip intercropping. In our baseline analysis, we assumed
that only convention tillage occurs currently.
Commercial agriculture dominates the watershed
with 84% of current landcover in crops (see the
first two columns of Table 1). In the current
landscape, only three kinds of landcover contributed to agricultural production: corn and
soybeans in a 2- year rotation; oats and alfalfa
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
122
Table 1
Land use under alternative landscape designs for the Walnut Creek Watershed of Iowaa
Production enterprise
Current
ha
Corn/Soybean
Grain oat/Alfalfa
Grass hay
C/S/C/S/O/A
Alfalfa
C/S/O strip intercropping
C/S/N strip intercropping
Organic C/S/C/S/O/A strip intercropping
All other land
Total land hectares
a
4174
90
44
830
5138
Production
%
81
2
1
16
100
ha
%
4501
88
1
0
636
5138
12
100
Water quality
Biodiversity
ha
ha
%
%
1719
826
3276
400
636
5138
16
64
8
12
100
33
2
500
1429
143
1345
5138
10
28
3
26
100
Note: C – Corn, S – Soybean, O – Oats, A – Alfalfa, N – Native prairie grass.
in a 4-year rotation, in which oats are planted as
a companion crop and harvested for grain in the
first year and alfalfa is harvested three times per
year in years 2, 3, and 4; and grass hay on a
4-year rotation harvested once a year. The current
distribution of agricultural land within the watershed is shown in Fig. 1(a).
2.1. Scenario one: production priority
In the first scenario, the primary objective is to
expand agricultural production and profitability.
This scenario is considered by many producers
and policy makers to be the most likely future
scenario if the current trends continue (Nassauer
et al., 1999).
In this scenario, corn and soybean production
is expanded at the expense of other land use, as
shown by the third and forth columns of Table 1.
The increase in corn and soybean land comes
mainly from the reduction of grass hay and oat
and alfalfa cropland. In addition to this shift of
crop type, there is an increase in total cropland by
nearly 500 acres, converted from non-cropland.
The distribution of landcover in this scenario is
shown in Fig. 1(b). In this scenario, we assume
that conservation tillage will be widely adopted.
Conservation tillage incorporates several factors,
such as crop residue management and alternative
tillage methods in a Best Management Practice
(BMP) used to reduce soil erosion.
2.2. Scenario two: water quality priority
Many agricultural watersheds in the Midwest
are facing water quality problems due to soil
erosion and large amounts of fertilizer and chemical use. Land use/land management decisions in
this scenario are targeted to reduce soil erosion,
reduce sediment delivery to streams, prevent the
movement of excess nutrients to streams, reduce
the energy and flashiness of storm events downstream, and improve aquatic habitat.
In this alternative, the design team set forth a
corn and soybean rotation of 6 years that incorporates conservation tillage. The first 4 years are a
corn and soybean rotation. In the fifth year, however, oats and alfalfa are planted. The oats are
harvested as ‘‘haylage’’ in year five. Haylage is
much the same as silage except it is produced
from grass hay instead of from corn or sorghum.
In the sixth year the alfalfa is harvested three
times, then chemically burned allowing corn to be
planted the following year. The oat/alfalfa portions of this rotation were intended to improve
the soil organic carbon content, which aids in the
maintenance of soil tilth, fertility, and storage of
carbon. Robinson et al., (1996) showed that over
the longterm continuous corn and corn/soybean
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
Fig. 1. The current agricultural landscape and three designed alternatives for the Walnut Creek Watershed of Iowa.
123
124
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
rotations resulted in significantly lower soil organic carbon content in soils than rotation of
corn/oats/alfalfa. The total land in agricultural
production increased by almost 500 acres in this
scenario relative to the baseline. Although less
land is in corn and soybeans, more land is in
alfalfa and grass hay to help reduce soil and
nutrient losses into both surface and groundwater. To further reduce runoff to streams and improve aquatic habitat, buffer strips were
established along stream corridors using alfalfa
and grass hay. Landcover patterns for this scenario are shown in Fig. 1(c).
2.3. Scenario three: biodi6ersity and water quality
priority
In this scenario, the primary objective is to
maintain and restore indigenous species to the
watershed. Some strategically located farmland
is taken out of production to create biodiversity
reserves. It is assumed that an ecosystem core
reserve of at least 640 acres will be established
in the watershed. In addition, ecologically sound
farming practices (e.g., perennial strip intercropping) are targeted to landscapes to connect and
buffer reserves and riparian corridors. The landcover pattern for this scenario is shown in Fig.
1(d).
In this scenario, different types of farming
methods are used including no-till, contour
planting, and strip intercropping. The last two
columns of Table 1 show the agricultural enterprises and their associated acreage for this scenario. Three strip intercropping systems are
utilized in this scenario. The first system consists
of corn and soybeans intercropped with a grain
oat crop. With careful planning, it is possible to
create rotations of corn, soybeans, and oats
such that the same crop is never planted in the
same place 2 years in a row. The second strip
intercropping system includes the use of native
prairie grass. In this planting scheme the native
prairie grasses remain stationary year after year
with only the corn and soybeans are being rotated. Prairie grass seed requires several years to
mature to full production. The third strip inter-
cropping system includes organic farming, in
which no commercial fertilizer and chemicals are
used. In order to sell a crop as organic it must
be certified by an organic certification agency.
This certification assures the buyer that the crop
was grown following a certain set of guidelines.
3. Research methods and data
For each of the three alternative scenarios
and the current baseline, a GIS database was
created that defined landcover and land management practices for each parcel in the watershed.
The landcover for the current landscape was
produced by a previous research project, the
Midwest Agrichemical Surface-subsurface Transport and Effects Research project (MASTER).
Under the MASTER project, aerial photographs
of the landcover in 1992 were digitized into a
GIS database, then ground-truthed in 1993 and
1994 (Freemark and Smith, 1995). The landcover for the three alternative scenarios were
also digitized into a GIS database (Nassauer et
al., 1999).
Soil and crop yield information for the watershed came from the Iowa Soil Properties And
Interpretation Database (ISPAID).1 This database provided soil type, number of layers with
specific information on each layer (e.g. horizon
depth, bulk density, particle percentages, pH,
and organic matter content) as well as depth of
water table. Todd Campbell of Iowa State University supplied additional information on soils
occurring in the Walnut Creek Watershed. The
soils information was combined with the landcover information for each alternative to create
a combined landcover-soils GIS database for the
watershed.
1
ISPAID was created by the Iowa Cooperative Soil Survey
(ICSS). The ICSS is a partnership between the Iowa Cooperative Extension Service and the Iowa Department of Agriculture and Land Stewardship, Division of Soil Conservation,
and the USDA Natural Resources Conservation Service. The
general purpose of the ICSS is to coordinate the collection,
compilation, interpretation, publication, dissemination and use
of soil surveys in Iowa.
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
3.1. Interacti6e en6ironmental policy integrated climate simulation model
The effect of alternative management practices
on crop yields and various environmental measures is evaluated through the use of the EPIC
program. EPIC was developed through the cooperation of the USDA-Agricultural Research Service, the Natural Resource Conservation Service,
and the Economic Research Service. EPIC is a
field-level simulation model developed to estimate
management impacts on agricultural production
and soil and water resources (Williams et al.,
1988). Its major modeling components include
weather simulation, plant growth, nutrient cycling, hydrology, erosion and sedimentation, pesticide fate, soil temperature, and management of
the plant environment by means of tillage, fertilization, irrigation, and conservation practices.
EPIC separates the soil profile into 10 layers.
Percolation through these layers depends on field
capacity, porosity, and saturated conductivity.
All pathways of the nitrogen cycle are modeled,
including denitrification, mineralization, immobilization, nitrification, volatilization, and fixation,
as well as inputs from rainfall. Williams et al.
(1988) provides a thorough description of nutrient cycling and all model components. In this
study, we used i – EPIC developed by Todd
Campbell of Iowa State University. i – EPIC is a
window version of EPIC, which provides an easier-to-use interface for interactively changing
EPIC input sets and tabulating EPIC output. It
can also control and automate large numbers of
EPIC runs, generating input, repeatedly invoking
EPIC, and cataloging the results (Campbell,
2000).
i – EPIC uses information on weather, soils,
crops, and production practices to simulate crop
yields and a number of environmental indicators
including NO3 −N leaching, NO3 −N runoff,
wind erosion, and water erosion. Weather information for the Walnut Creek Watershed was supplied in i – EPIC. The data for soils were
described above. Each scenario defined crop and
basic production practices. Detailed production
practice information such as the types and dates
125
of tillage operation and the amount of pesticides
and fertilizers application for each crop was
derived from the 1998 Iowa State Crop Enterprise Budgets developed by Iowa State University
Extension Service and other sources. A 30-year
i – EPIC simulation run was conducted for each
field in the Walnut Creek Watershed. The 30-year
simulation runs allowed us to capture the effect
of crop management practices on soil quality
including soil organic percentage. The resulting
changes in soil quality are reflected in changes in
simulated crop yields.
Each run provides daily estimates of NO3 − N
leaching, NO3 − N runoff, wind erosion, and water erosion. NO3 − N leaching is measured as the
quantity of NO3 − N leaving the root zone, and
NO3 − N runoff is measured as the quantity of
NO3 − N leaving the field via surface runoff. Eutrophication, which stimulates toxins-producing
algae, is the primary concern with NO3 − N
runoff, whereas groundwater contamination is
the primary concern with NO3 − N leaching
(Brady and Weil, 1999). The 30-year averages of
simulated NO3 − N runoff, NO3 − N leaching,
water erosion, and wind erosion, were used in the
analysis.
EPIC has been validated and calibrated for a
wide variety of conditions, particularly for those
prevalent in this study region (Jones et al., 1985;
Williams et al., 1988). EPIC has also been shown
to be a reasonable predictor of NO3 − N runoff
and leaching losses on several sites (Jones et al.,
1985; Mapp et al., 1994). In Iowa, Chung et al.
(1999) compared the simulated NO3 − N leaching
levels and tile drain flows with field measurements of these variables at Iowa State University
research farm located near Nashua, Iowa, during
1990–1992. The field experiments were conducted
to evaluate the effects of tillage and crop rotation
systems on groundwater quality (Weed and Kanwar, 1996; Kanwar et al., 1997). The site has 36,
0.4-ha experimental plots with different tillage
and crop rotation systems. Each tillage and rotation combination was replicated three times. In
Chung et al. (1999), EPIC was tested with data
collected from plots managed with different com-
126
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
Fig. 2. The actual corn yields versus the predicted yields.
binations of three tillage systems (moldboard,
chisel plow, and no-till) and two crop rotations
(continuous corn and corn/soybean rotation). The
measured data used for the comparisons with
EPIC output were based on the average levels of
NO3 −N leaching and tile drain flows across the
replicated combinations of tillage systems and
crop rotations. Chung et al. (1999) found that
predicted monthly subsurface drainage flows and
NO3 −N losses agreed well with observed values
and were statistically acceptable for nearly all of
the simulated management systems, however,
there were large variations between the EPIC
daily predictions and observed values. These results suggest that EPIC was able to replicate the
long-term average of water and NO3 −N leaching
losses for the cropping systems evaluated in this
study. However, by using a field-based model, the
analysis is limited to the soil-plant system. No
attempt is made to model the soil and NO3 −N
movements after they leave the field in surface
runoff or leach below the root zone. In essence,
the estimated NO3 − N loadings represent a
‘‘worst-case’’ scenario since the transport process
tends to reduce the amount of NO3 − N actually
reaching surface or groundwater sources. Other
members of the project use Soil and Water Assessment Tool (SWAT) to model the transport
process.
As in previous studies that use EPIC, we calibrated i – EPIC to better reflect crop yields for soil
types in the Walnut Creek Watershed. The calibration consisted of adjusting input parameters
until the average of predicted yields across soils
closely equaled the reported yields in the ISPAID
database. Fig. 2 demonstrates the accuracy of the
calibration for corn yields. The calibration was
less accurate for soybean yields (Fig. 3). Because
the two crops were used in a 2-year rotation, the
parameters were adjusted to best suit the combined crops. ISPAID did not report yields for
alfalfa and grass hay, so a 10-year average of
alfalfa and grass hay yields in Boone County, as
reported by the National Agricultural Statistics
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
127
Fig. 3. The actual soybean yields versus the predicted yields.
Service, was used. Calibration of the alfalfa crop
was not as successful as that of the corn and
soybeans. i – EPIC typically predicted yields of
about 1.2 t over the of the Boone County average.
Thus, the predicted alfalfa yields were adjusted by
the percentage of over prediction across alfalfa
fields. For grass hay the model predicted yields
that closely equaled the county average.
3.2. Methods for estimating returns to land
We combined crop yield information from i –
EPIC with information on crop prices and production costs to calculate return to land for each
crop enterprise. Return to land is defined as revenue minus all costs except land cost.
In order to calculate the production costs for all
cropping systems and tillage operations, crop enterprise budgets were used. The budgets covered
all of the production costs for corn, soybeans,
grass hay, and alfalfa. These budgets are based on
averages for all of Iowa producers’ costs. Individual producers may have higher or lower production costs. Included with these budgets were
supplemental information on farming operation
for both conventional and conservation tillage.2
The National Agricultural Statistics Service
(NASS) collected data on prices received by farmers for various crops by county. For this study,
Boone County prices from 1987 to 1997 were
collected. These prices were indexed to 1998 dollars using the Indexes of Prices Received by
Farmers For all Farm Products. We used the
average of the 10-year indexed prices to calculate
total revenues.
2
Paul Mitchell of Iowa State University supplied the dates
and types of tillage operations that occurred for each crop
enterprise. Mike Duffy, Alan Vontalge, and John Lawrence
from Iowa State University Extension Service supplied 1998
Iowa State Crop Enterprise Budgets.
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
128
Table 2
The economic and environmental impacts of alternative cropping systemsa
Cropping system
Return to land
($/ha)
N runoff (kg/ha) N leaching
(kg/ha)
Water erosion
(t/ha)
Wind erosion
(t/ha)
Current landscape
Corn/Soybean
Grain oats/Alfalfa
Grass hay
Average
Total
363
235
232
361
1 549 665
5.7
1.9
0.0
5.5
23 835
10.4
4.9
0.0
10.2
44 131
8.5
0.9
0.4
8.1
35 269
8.3
1.8
0.2
8.1
34 656
Production scenario
Corn/Soybean
Grass hay
Average
Total
385
242
385
1 730 988
4.0
0.0
4.0
17 922
12.3
0.0
12.3
55 353
2.2
0.0
2.2
10 485
2.2
0.0
2.2
10 563
Water quality scenario
C/S/C/S/O/A
Alfalfa
Grass hay
Average
Total
255
346
245
262
1 176 091
2.6
0.2
0.0
1.9
7 469
13.1
9.1
0.0
10.4
39 528
2.5
0.0
0.5
2.0
7 473
1.6
0.0
0.0
1.1
4 622
385
242
297
460
358
3.8
0.0
1.9
1.9
1.1
12.8
0.0
20.4
18.2
9.3
1.6
0.2
1.1
1.1
3.4
2.5
0.0
1.3
1.3
1.8
400
1 522 612
2.7
10 344
15.7
59 688
1.3
5 347
1.8
6 932
Biodi6ersity scenario
Corn/Soybean
Grass hay
C/S/O strip intercropping
C/S/N strip intercropping
Organic C/S/C/S/O/A strip
intercropping
Average
Total
a
Note: C – Corn, S – Soybean, O – Oats, A – Alfalfa, N – Native prairie grass, t/ha – tonnes/hectare.
There is not much information at present on
native prairie grass production for seed, either in
terms of yields or prices. We estimated returns for
native prairie grass seed production based on
yield and price information from Carl Kurtz, a
private producer in Iowa. The estimates were
highly speculative because both future demand
and supply of native prairie grass seeds are
uncertain. For this reason, sensitivity analysis was
conducted to examine how changes in the prices
of prairie grass seeds affect our results.
4. Results
In this section, results are presented on the
economic and environmental impacts of alterna-
tive landscape scenarios. The first part of the
section shows total return to land from agricultural commodity production under each scenario.
The second part of the section shows how various
environmental measures score under each scenario. The final part of the section evaluates the
complementarities and tradeoffs between environmental and economic objectives.
4.1. Agricultural profits
The first column of Table 2 shows the total
returns to land for the current landscape as well
as for the three alternative scenarios. Included
within the table is a breakdown of average returns
to land per acre for each enterprise. The total
return to land under the current landscape is
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
$1 549 665 ($361/ha). Corn/soybean is the most
profitable of the three cropping systems, followed
by grain oats/alfalfa. The spatial distribution of
return to land across the watershed is shown in
Fig. 4(a).
Under the production scenario, almost all the
current cropland was placed into corn and soybean production. In addition, 194 ha of non-cropland were converted to cropland. With the
increase in total crop area and adoption of no-till,
total return to land was increased to $1 730 988.
This represents an 11.7% increase over the total
return to land under the current landscape. The
returns to land for both corn/soybean and grass
hay production are higher in the production scenario than in the current landscape. The increase
can be attributed to conservation tillage, which
saves fuel, labor and machinery costs. In addition,
the least profitable soil types are not cropped,
under the assumption that precision farming is
being used to identify soils with low corn suitability rating and ‘‘skipping’’ that portion of land.
The spatial distribution of return to land for the
production scenario is shown in Fig. 4(b).
In the water quality scenario, total agricultural
land was increased by 194 ha compared to the
current landscape due to a large increase in grass
hay area. However, the total return to land was
reduced by $373 574 (a 24% reduction) compared
with the baseline. In this scenario, the corn/soybean rotation is extended to a 6-year rotation. In
the final 2 years of the rotation companion crops
of oats for haylage and alfalfa are planted. For
most fields, the oat/alfalfa phase yields low return
to land. After only one full growing season, the
alfalfa is not producing at its full potential. Even
if the alfalfa crop had been allowed to reach full
maturity in an 8-year rotation like C/S/C/S/O/A/
A/A, this enterprise would still not be as
profitable as corn/soybean rotation because the
return to land during the oat/alfalfa phase is only
about $150/ha, which is far lower than for a
corn/soybean rotation. The spatial distribution of
return to land for the water quality scenario is
shown in Fig. 4(c).
The biodiversity scenario generates total returns
to land of $1 522 612. This total is only slightly
below the returns in the current landscape despite
129
the fact that over 400 ha have been taken out of
production agriculture. This scenario generates
the highest average return to land among the four
scenarios ($400/ha). The main reason why this
scenario generates high per acre return is the high
return in the corn/soybean/native prairie grass
strip intercropping, which depends on the assumption of a high price for native grass seeds. As
noted previously, it is difficult to forecast the price
for native prairie grass seed. If the price of native
grass seeds is assumed to be the half of the
current level, the return to land for the corn/soybean/native strip intercropping enterprise is reduced from $460 to $299 per hectare. With this
alternative price assumption, the average return to
land in the biodiversity scenario is reduced to
$341, which is lower than in both the current
landscape and the production scenario. This result
suggests that the extent to which economic and
environmental objectives can be achieved together
heavily depends on assumptions about the productivity and profitability of environmentally benign farming practices.
In the biodiversity scenario, the organic enterprise uses the same 6-year crop rotation (C/S/C/S/
O/A) as in the water quality scenario. The return
to land for this rotation is higher in this scenario
than in the water quality scenario for two reasons.
First, in organic farming no artificial chemicals
and fertilizer were applied, which reduced production costs. Second, in organic farming, the crops
are being grown in a strip intercropping pattern,
which tends to increase crop yields, other things
being equal. One of the challenges of incorporating organic crops in the analysis is that there is
not much data on prices, production costs, or
yields as there is for conventionally grown crops.
Because of the lack of data on prices for organically grown crops, we used prices for conventionally grown crops instead. This assumption will
tend to understate the returns for organics, which
currently enjoy a price premium.3 However, as
more producers switch to organic production,
prices for organic crops may begin to fall. The
3
A 50% increase in organic crop prices would result in an
86% increase in the average return to land for this enterprise
(from $358 to $665 per hectare).
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
Fig. 4. Spatial distribution of returns to land ($/ha).
130
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
spatial distribution of the return to land for the
biodiversity scenario is shown in Fig. 4(c).
4.2. En6ironmental measures
The environmental measures we tracked in this
study were NO3 − N runoff, NO3 −N leaching,
soil water erosion, and soil wind erosion. Table 2
gives a detailed breakdown of these measures per
crop enterprise for each scenario.
In the current landscape, the corn/soybean enterprise generates the largest environmental degradation as measured by each of the four measures
of any crop enterprise, while grass hay causes the
least. The average of NO3 −N runoff per hectare
of cropland is 5.5 kg. NO3 −N leaching is significantly higher, with an average of 10.2 kg/ha. The
grain oat/alfalfa enterprise contributes significantly less NO3 −N pollution to the watershed
with 1.9 kg of NO3 −N runoff and 4.9 kg of
NO3 −N leaching per hectare. Grass hay contributes no NO3 −N pollution to the watershed.
Fig. 5(a) and Fig. 6(a) show the spatial distribution of the NO3 − N runoff and leaching, respectively. The most soil erosion is also caused by
corn and soybean production. Both water and
wind erosion rates for this enterprise are above 8
t/ha. Water and wind erosion rates are below 2
t/ha with an oat/alfalfa rotation. The erosion
rates on grass hay fields are negligible. Fig. 7(a)
and Fig. 8(a) show the spatial distribution of
water and wind erosion rates across the watershed
in the current landscape.
In the production scenario, corn/soybean production is expanded at the expense of other land
use. However, with the move to conservation
tillage in this scenario, all measures of environmental degradation were lower than in the current
landscape except NO3 −N leaching. The adoption
of conservation tillage reduces soil erosion significantly in the production scenario compared to the
current landscape, even though more land is
devoted to corn and soybean production. In total
water erosion is reduced from 35 269 to 10 485 t,
while wind erosion reduces from 34 656 to 10 563
t. NO3 − N runoff also declines, falling from
23 835 to 17 922 kg. The total NO3 −N leaching,
however, increases from 44 131 to 55 353 kg. This
131
result is consistent with some of the previous
studies on the effect of conservation tillage on
NO3 − N water pollution (Tyler and Thomas,
1979; Thomas et al., 1981; Wu and Babcock,
1999). Fig. 5(b) and Fig. 6(b) show the spatial
distribution of the NO3 − N runoff and leaching
in the production scenario, respectively. Fig. 7(b)
and Fig. 8(b) show the spatial distribution of
water and wind erosion rates across the watershed
in the current landscape.
The water quality scenario generally scored the
best on the environmental measures of any of the
alternatives. This scenario provided the lowest
total NO3 − N runoff and leaching as well as
wind erosion. For water erosion, only the biodiversity scenario generated lower erosion. These
results are largely the result of switching land
from corn and soybean production to grass hay
and alfalfa. Surprisingly, average NO3 − N leaching is slightly higher for each crop enterprise in
the water quality scenario. A large portion of the
NO3 − N loss comes from the 6-year crop rotation, in which the average NO3 − N leaching rate
is 13.1 kg/ha. Alfalfa also has a fairly high rate of
NO3 − N leaching at 9.1 kg/ha. Fig. 5(c), Fig.
6(c), Fig. 7(c) and Fig. 8(c) show the spatial
distribution of NO3 − N runoff, NO3 − N leaching, water erosion, and wind erosion rates in the
water quality scenario. It is interesting to note
that little NO3 − N runoff and water and wind
erosion come from alfalfa and grass hay fields
next to the stream.
The biodiversity scenario incorporated the use
of no-till as well as strip intercropping in the
agricultural operations. The average NO3 − N
runoff for this alternative is 2.7 kg/ha, less than
that of both the current landscape and the production scenario, though higher than for the water quality scenario. The overall average NO3 − N
leaching is 15.7 kg/ha, which is the highest of the
four landscapes. The two nonorganic strip-intercropping systems generate more NO3 − N leaching than traditional corn/soybean rotations. The
average NO3 − N leaching rate for the corn–soybean–oats strip intercropping is 18.2 pounds per
acre, and the average NO3 − N leaching rate for
the corn/soybean/native prairie grass seed strip
intercropping is 18.2 kg/ha. The high NO3 − N
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
Fig. 5. Spatial distribution of NO3 runoff (kg/ha).
132
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
Fig. 6. Spatial distribution of NO3 leaching (kg/ha).
133
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
Fig. 7. Spatial distribution of soil water erosion (t/ha).
134
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
Fig. 8. Spatial distribution of wind erosion (t/ha).
135
136
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
leaching rates can be attributed to a combination
of several factors, including no-till that results in
greater soil porosity and the strip intercropping
system. The organic farming system generates a
NO3 −N leaching rate of 9.3 kg/ha, which is
: 3.5 kg less than that of the traditional 2-year
corn and soybean rotation. In the organic farming
system, no artificial fertilizer is applied.
The average water and wind erosion rates from
the biodiversity scenario are 1.3 and 1.8 t/ha,
respectively. The highest average water erosion
rate occurs in the organic farming operation,
probably as a result of the tillage operations that
occur in the organic farming. Total water erosion
was only 5 347 t, which is less than in any of other
scenarios. The corn/soybean rotation is the largest
contributor to wind erosion in this scenario. Total
wind erosion to the watershed is 6 932 t, which is
much lower than that in the current and production scenarios, but higher than the water quality
scenario. The spatial distributions of the environmental measures are shown in Fig. 5(d) Fig. 6(d)
Fig. 7(d) Fig. 8(d).
changes in land use or agricultural practices result
in environmental improvements on certain dimensions in addition to making economic sense. But
most changes in land use or agricultural practice
do not bring uniform environmental improvement. There may be difficult tradeoffs between
different components of environmental quality.
For example, the switch from conventional to
conservation tillage improves soil erosion and
NO3 − N runoff measures but worsens NO3 − N
leaching. Comparing the biodiversity scenario
with the water quality scenario, we find less water
erosion but higher wind erosion.
In each alternative there are some negative
consequences along with the gains. Changes in
agricultural practices should be targeted to specific environmental goals, and the tradeoff accompanying each choice should be carefully studied.
With the method presented in this study, it is
possible to explore these tradeoffs and to better
understand the impacts our choices have on economic and environmental goals.
5. Conclusions
4.3. Economic and en6ironmental objecti6es:
complementary or tradeoff?
The ranking of which landscapes are most preferred is not strictly an economic or environmental question. The two must be viewed jointly in
making an informed policy decision. Fig. 9 shows
total soil and NO3 − N loss from all fields versus
the total return to land for the whole watershed.
It thus provides indications as to the degree to
which economic and environmental objectives can
be achieved together or involve tradeoffs. For
example, moving from the current landscape to
the production scenario improves return to land
and all environmental measures except NO3 −N
leaching. On the other hand, comparing the water
quality scenario with other scenarios shows that
improved environmental quality may come at
some economic cost in terms of lower return to
land.
It is important to note that not all tradeoffs are
between economics and the environment. Environmental quality has many components. Some
This paper evaluates the economic and environmental impacts of alternative landscape designs
for the Walnut Creek Watershed of Iowa. Return
to land and the four measures of externalities are
estimated by combining site-specific information
on soil, climate, cropping systems, and production
practices through a computer simulation model
known as the i – EPIC. This method also allows
for comparisons between different cropping systems and production practices.
Our results reveal that the production scenario
produced the highest return to land among the
four landscapes evaluated in this study. This is
due to an increase in total harvested acres and a
reduction in production costs associated with the
adoption of no-till. The biodiversity scenario results in the second largest return to land. The
result, however, is sensitive to the assumption
about the price of native prairie grass seed. The
water quality scenario produces the lowest return
to land of the four landscapes, but it generally
scores the best on the environmental measures.
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
Fig. 9. Total externalities versus total return to land.
137
138
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
The water quality scenario provides the lowest
total NO3 −N runoff and leaching as well as
wind erosion. For water erosion, only the biodiversity scenario generates lower erosion. These
results are largely the result of switching land
from corn and soybean production to grass hay
and alfalfa, which holds the soil and slows the
movement of water across the surface. The biodiversity scenario has the highest total and per acre
rate of NO3 − N leaching. The current landscape
results in the largest amount of soil erosion and
NO3 −N runoff of the four scenarios. These results suggest that there are tradeoffs between alternative environmental objectives as well as
between economic and environmental objectives.
This study is only one part of a much larger
multidisciplinary project that looks at many other
social and environmental aspects of the watershed. Other project members are studying the
effects of the alternative landscapes on biodiversity, other water quality measures, and social
acceptability. These other concerns led to particular design of the alternatives scenarios. For example, the 6-year rotation involving corn, soybeans,
oats and alfalfa in the water quality scenario,
which had low returns to land without significant
improvement in environmental indicators considered in this study, was chosen based on soil
organic carbon improvements, which was not
measured in this study. For biodiversity, the loss
of habitat for forest species and grassland species,
which leads to further declines in native biodiversity, was also not analyzed in this study. The loss
of habitat and the decline in native biodiversity
are evaluated in Santelmann et al. (2001). With
additional information, this study could help develop more economically and environmentally
sound policies, and help guide choices of alternative cropping systems and agricultural practices to
address environmental problems in agricultural
watersheds.
Acknowledgements
This study was a part of a project funded by
US EPA under grant cR825335-01. We thank
the project PI Mary Santelmann and Kelly Vache
for help with the GIS data, as well as other
members of the project. We thank seminar participants at the Heartland Environmental Resource
Economics Conference in Ames, Iowa, in September, 1999, and three referees for helpful
comments.
References
Blevins, R.L., Thomas, G.W., Cornelius, P.L., 1977. Influence
of no-tillage and nitrogen fertilization on certain soil properties after five years of continuous corn. Agron. J. 69,
383 – 386.
Blevins, R.L., Smith, M.S., Frye, W.W., Cornelius, P.L., 1983.
Changes in soil properties after 10 years non tilled and
conventionally tilled corn. Soil Tillage Res. 3, 135 – 146.
Brady, N.C., Weil, R.R., 1999. The Nature and Properties of
Soils, 12th ed. Simon and Schuster, Upper Saddle River,
NJ.
Byström, O., 1998. The nitrogen abatement cost in wetlands.
Ecol. Econ. 24, 321 – 331.
Campbell, T., 2000. I – EPIC: Users’ Guide. Center for Agricultural and Rural Development, Iowa State University.
Chung, S.W., Gassman, P.W., Kramer, L.A., Williams, J.R.,
Gu, R., 1999. Validation of EPIC for two watersheds in
Southwest Iowa. J. Environ. Qual. 28, 971 – 979.
Freemark, K., Smith, J., 1995. A landscape retrospective for
Walnut Creek, Story County. Technical Report to the
United States Environmental Protection Agency.
Gilliam, J.W., Hoyt, G.D., 1987. Effect of conservation tillage
on fate and transport of nitrogen. In: Logan, T.J., Davidson, J.M., Baker, J.L., Overcash, M.R. (Eds.), Effects of
Conservation Tillage on Groundwater Quality: Nitrates
and Pesticides. Lewis Publishers, Chelsea, MI, pp. 217 –
240.
Gren, I.M., 1993. Alternative nitrogen reduction policies in the
Mälar region, Sweden. Ecolog. Econ. 7, 159 – 172.
Gren, I.M., Elofsson, K., Jannke, P., 1997. Cost-effective
nutrient reductions to the Baltic Sea. Environ. Res. Econ.
10, 341 – 362.
Hallberg, G.R., 1989. Nitrate in ground water in the United
States. In: Follet, R.F. (Ed.), Nitrogen Management and
Ground-Water Protection. Elsevier, New York, pp. 35 – 74.
Jacobs, J.J., Timmons, J.F., 1974. An economic analysis of
agricultural land use practices to control water quality.
Am. J. Agric. Econ. 56, 791 – 798.
Jones, C.A., Williams, J.R., Sharpley, A.N., Cole, C.V., 1985.
Testing the nutrient components of EPIC. In: Decoursey,
D.G. (Ed.), Symposium Proceedings of the Natural Resources Modeling, Pingree Park, CO.
Kanwar, R.S., Baker, J.L., Laflen, J.M., 1985. Nitrate movement through the soil profile in relation to tillage system
and fertilization application method. Trans. ASAE 28,
1802 – 1807.
C. Coiner et al. / Ecological Economics 38 (2001) 119–139
Kanwar, R.S., Colvin, T.S., Karlen, D.L., 1997. Subsurface
drain water quality under ridge and three other tillage
systems. J. Prod. Agric. 10, 227 –234.
Kellogg, R.L., Maizel, M.S., Goss, D.W., 1992. Agricultural
chemical use and ground water quality: where are the
potential problem areas? USDA Staff Rep., Soil Conservation Service/Economic Research Service, Washington, DC.
Kitur, B.K., Smith, M.S., Blevins, R.L., Frye, W.W., 1984.
Fate of N-Depleted ammonium nitrate applied to notillage and conventional tillage corn. Agron. J. 76, 240 –
242.
Kronvang, B., Grant, R., Larsen, S.E., Svendsen, L.M., Andersen, H.E., 1995. Non-point-source nutrient losses to the
aquatic environment in Denmark, impact of agriculture.
Mar. Freshwater Res. 46, 167 –177.
Mapp, H.P., Bernardo, D.J., Sabbagh, G.J., Geleta, S.,
Watkins, K.B., 1994. Economic and environmental impacts of limiting nitrogen use to protect water quality: a
stochastic regional analysis. Am. J. Agric. Econ. 76, 889 –
903.
Martin, M.A., Schreiber, M.M., Rosscup Riepe, J., Bahr, J.R.,
1991. The economics of alternative tillage systems, crop
rotations and herbicide use on three representative EastCentral Corn Belt farms. Weed Sci. 39, 299 –307.
McMahon, M.A., Thomas, G.W., 1976. Anion leaching in two
Kentucky soils under conventional tillage and killed sod
mulch. Agron. J. 71, 1009 –1015.
Mueller, D.K., Hamilton, P.A., Helsel, D.R., Hitt, K.J.,
Ruddy, B.C., 1995. Nutrients in ground water of the
United States – an analysis of data through 1992. WaterResources Investigations Rep. No. 95-4031, US Geological
Survey, Denver, CO.
Nassauer, J.I., Corry, R.C., Cruse, R.M., 1999. Alternative
future landscape scenarios for corn belt agricultural watersheds. J. Soi1 Water Conserv. (in review). http://
www.nass.usda.gov:100/ipedb/.
National Research Council, 1989. Alternative Agriculture. National Academy of Press, Washington, DC.
Nielsen, E.G., Lee, L.K., 1987. The magnitude and costs of
groundwater contamination from agricultural chemicals: a
national perspective. Agric. Econ. Rep. No. 576, USDA,
Economic Research Service, Washington, DC.
Robinson, C.A., Cruse, R.M., Ghaffarzadeh, M., 1996. Crop-
.
139
ping system and nitrogen effects on mollison organic carbon. Soil Sci. Soc. Am. J. 60, 264 – 269.
Santelmann, M., Freemark, K., White, D., Nassauer, J.,
Clark, M., Danielson, B., Eilers, J., Cruse, R., Galatowitsch, S., Polasky, S., Vache, K., Wu, J., 2001. Applying
ecological principles to land-use decision making in agricultural watersheds. In: Dale, V., Haeuber, R. (Eds.),
Applying Ecological Principles to Land Management.
Springer, New York.
Sharpley, A.N., Williams, J.R., 1990. EPIC – Erosion Productivity Impact Calculator: Vol. I. Model Documentation.
USDA Technical Bulletin No. 1768, Washington, DC.
Smith, R.A., Alexander, R.B., Wolman, M.G., 1987. Water
quality trends in the nation’s rivers. Science 235, 1607 –
1615.
Thomas, G.W., Wells, K.L., Murdock, L., 1981. Fertilization
and Liming. In: Phillips, R.E., Thomas, G.W., Blevins,
R.L. (Eds.), No Tillage Research: Research Reports and
Reviews. College of Agriculture and Agricultural Experiment Station, University of Kentucky, Lexington, KY.
Tyler, D.D., Thomas, G.W., 1979. Lysimeter measurement of
nitrate and chloride losses from conventional and notillage corn. J. Environ. Qual. 6, 63 – 66.
Weed, D.A.J., Kanwar, R.S., 1996. Nitrate and water present
in and flowing from root-zone soil. J. Environ. Qual. 25,
709 – 719.
West, T.D., Griffith, D.R., 1992. Effect of strip-intercropping
corn and soybean on yield and profit. Purdue University,
West Lafayette, IN, CES-257.
Wiersink, A., Walker, M., Swanton, C., Shaw, J.E., 1992.
Costs of conventional and conservation tillage systems. J.
Soil Water Conserv. 47, 328 – 333.
Williams, J.R., Jones, C.A., Dyke, P.T., 1988. EPIC, the
erosion productivity index calculator. Model documentation, vol. 1. US Department of Agriculture, Agricultural
Research Service, Temple, TX.
Wu, J., Babcock, B.A., 1999. Metamodeling potential nitrate
water pollution in the Central United States. J. Environ.
Qual. 28, 1916 – 1928.
Wu, J., Bernardo, D.J., Mapp, H.P., Geleta, S., Teague, M.L.,
Watkins, K.B., Sabbagh, G.J., Elliot, R.L., Stone, J.F.,
1997. An evaluation of nitrogen runoff and leaching potential in the High Plains. J. Soil Water Conserv. 52, 73 – 80.