A Whole Farm Analysis of the Influence of Auto-Steer Navigation on Net Returns, Risk, and Production Practices
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Journal of Agricultural and Applied Economics, 43,1(February 2011):57–75
Ó 2011 Southern Agricultural Economics Association
A Whole Farm Analysis of the Influence of
Auto-Steer Navigation on Net Returns,
Risk, and Production Practices
Jordan M. Shockley, Carl R. Dillon, and Timothy S. Stombaugh
A whole farm economic analysis was conducted to provide a detailed assessment into the
economic, risk, and production implications due to the adoption of auto-steer navigation. It
was determined that auto-steer navigation was profitable for a grain farmer in Kentucky with
net returns increasing up to 0.90% ($3.35/acre). Additionally, the technology could be used
in reducing production risk. Adoption of the technology also alters production practices for
optimal use.
Key Words: economics, farm management, mean-variance, precision agriculture, simulation
JEL Classifications: C61, C63, D81, Q12, Q16
Automated steering (auto-steer) is a navigation correspondingly different levels of accuracy.
aid that utilizes the global position system (GPS) The potential benefits of these systems include
to guide agricultural equipment. Auto-steer has reduction of overlaps and skips, reduced inward
been commercially available for many years. drift of implements, the lengthening of operator’s
There are many combinations of auto-steer workday, accurate placement of inputs, and re-
systems and GPS receivers available with duced machinery costs resulting from an increase
in machinery field capacity. The increase in ma-
chinery field capacity not only could reduce
Jordan M. Shockley is post doctoral scholar, Department
of Agricultural Economics, University of Kentucky, direct costs, but permit more land area to be
Lexington, KY. Carl R. Dillon is professor, Department planted closer to the optimal date. These ad-
of Agricultural Economics, University of Kentucky, vantages provide incentive for producers to
Lexington, KY. Timothy S. Stombaugh is associate
extension professor, Department of Biosystems and
evaluate the potential of this technology in
Agricultural Engineering, University of Kentucky, their farm operation.
Lexington, KY. Many Kentucky farmers have adopted some
The authors would like to thank the editor and the form of GPS-enabled navigation technology. The
three anonymous reviewers for their helpful com-
ments. This research was partially funded by a U.S.
trend for most Kentucky farmers is to first adopt
Department of Agriculture-Cooperative State Re- a bolt-on auto-steer system equipped with a sub-
search, Education, and Extension Service Grant titled meter receiver on the self-propelled sprayer. For
‘‘Precision Agriculture: Development and Assessment the utmost accuracy, a Kentucky farmer upgrades
of Integrated Practices for Kentucky Producers’’ and
was supported by the University of Kentucky Agricul- to an integral valve system with a Real Time Ki-
tural Experiment Station. It is published by permission nematic (RTK) GPS receiver on the tractor. Few
of the Director as station manuscript number 10-04- Kentucky farmers have utilized auto-steer systems
110. Any opinions, findings, conclusions, or recom-
on their harvesters, but when they do, it is common
mendations expressed in this publication are those of
the authors and do not necessarily reflect the views of to use a bolt-on system. Despite rapid adoption,
the U.S. Department of Agriculture. the quantitative benefits of auto-steer have been58 Journal of Agricultural and Applied Economics, February 2011
scrutinized by farmers. A thorough investigation risk was incorporated. However, the lack of
by researchers into this technology is long over- yield data, and the interactive effects of pro-
due. Issues regarding profitability, interactive ef- duction practices, necessarily led to overly re-
fects of production practices, and the often ignored strictive assumptions and results. Others have
issue of risk are all essential to evaluate. developed theoretical models that suggested
The majority of existing studies conducted variable rate technology could be utilized in
on navigational technologies focused on field managing production risk (Lowenberg-DeBoer,
performance or general overviews of the tech- 1999). The investigation into auto-steer as a risk
nology, which ultimately emphasized engineer- management tool was meager, therefore it be-
ing concepts. Field performance studies focused came an objective of this study.
on issues regarding accuracy, topography, speed, The objectives of this study are to: (1) de-
and evaluation methods (Ehsani, Sullivan, and termine profitability of auto-steer under various
Walker, 2002; Gan-Mor, Clark, and Upchurch, scenarios; (2) determine if auto-steer can be
2007; Stombaugh et al., 2007; Stombaugh and utilized as a tool for risk management; (3) de-
Shearer, 2001). Research involving the overall termine optimal production practices under var-
status of navigational technologies in North ious scenarios with and without auto-steer; (4)
America and Europe had also been reported determine the break-even acreage level, payback
(Adamchuk, Stombaugh, and Price, 2008; Keicher period, and return on investment for the adoption
and Seufert, 2000; Reid et al., 2000). Other re- of auto-steer; and (5) determine the impact of
search previously conducted had focused on the input price on the profitability of auto-steer. A
economics of auto-steer. whole farm economic model is used to provide
Economic studies regarding auto-steer often a detailed assessment of auto-steer options for
utilized simple techniques which failed to en- a hypothetical grain farm in Kentucky. Due to
compass all benefits and costs of the technology. the adoption trend for auto-steer by Kentucky
A limited number of whole farm economic farmers, investigations are undertaken consider-
studies of auto-steer had been conducted (Griffin, ing three scenarios: (1) the addition of a bolt-on
Lambert, and Lowenberg-DeBoer, 2005, 2008). auto-steer system with a sub-meter receiver on
These economic studies had not included a farm a self-propelled sprayer, (2) the addition of an
manager’s ability to exploit the technology by integral valve auto-steer system with an RTK
altering production practices to increase profit- GPS receiver on a tractor, and (3) the addition of
ability or reduce risk. While some of these both auto-steer systems to the farm enterprise.
studies were helpful, the economic potential of Scenario three investigates the situation in which
the technology may be understated to the extent a farmer is utilizing sub-meter auto-steer on the
that substitution of inputs and alteration of pro- sprayer and an RTK auto-steer on the tractor.
duction practices were not addressed in these Hence, the benefits and costs of both systems are
models. Widespread interest, coupled with the incorporated into the model. All five of the
scarcity of studies, motivates the incorporation above objectives are investigated for each of the
of this technology into a more complete whole above scenarios as well as incorporating four
farm planning model. By including alternative farmer risk aversion attitudes: neutral, low, me-
production practices, economic optimization can dium, and high risk aversion for objectives one
be achieved. In turn, this allows investigation through four. The four risk aversion levels rep-
into the full potential of auto-steer on the farm. resent a desire to maximize net returns that are
Few researchers conducted in-depth risk 50%, 65%, 75%, and 90% likely to be achieved
analyses of precision agriculture technologies, for neutral, low, medium, and high risk aversion
beyond the present focus of this study. Dillon levels, respectively.
et al. (2005) conducted educational workshops
to inform farmers of the risk management po- Analytical Procedure
tential of precision agriculture. Oriade and
Popp (2000) conducted a whole farm planning The experimental framework for this study in-
model of precision agriculture technology where cludes the production environment, the economicShockley, Dillon, and Stombaugh: Influence of Auto-Steer Navigation 59
Table 1. Summary of Corn Production Practices Utilized within this Studya
Planting Date March 25, April 1, April 8, April 15, April 22,
April 29, May 6, May 13, May 20
Maturity group (growing degree days) 2,600–2,650, 2,650–2,700, 2,700–2,750
Plant population (plants/acre) 24,000; 28,000; 32,000
Row spacing 300
Plant depth 1.50
Nitrogen rate (actual lbs/acre) 100, 150, 175, 200, 225
Summary of Soybean Production Practices Utilized within this Study
Planting date April 22, April 29, May 6, May 13, May 20,
May 27, June 3, June 10, June 17
Maturity group MG2, MG3, MG4
Plant population (plants/acre) 111,000; 139,000; 167,000
Row spacing 150, 300
Plant depth 1.250
a
Both corn and soybean production practices chosen for this investigation were consistent with the University of Kentucky
Cooperative Extension Service grain crop recommendations.
optimization whole farm model, and the specific Department of Agricultural (USDA) Natural
conditions and resource base of the hypothetical Resources Conservation Service (NRCS) (2008).
farm that represents the study focus. These are After identifying all soil series located in
each discussed in turn to establish the analytical Henderson County, information on those soil
framework of the study. series was gathered using the NRCS Official Soil
Series Description from their website. Four rep-
resentative soils (deep silty loam, deep silty clay,
The Production Environment shallow silty loam, and shallow silty clay) were
utilized in the biophysical simulation models.
Production data estimates were determined using Finally, numerous production practices were
Decision Support System for Agrotechnology defined to complete the minimum requirements
Transfer (DSSAT v4), a biophysical simulation to operate DSSAT. Production practices were
model (Jones et al., 2003). DSSAT has provided identified for both corn and full season soybeans
underlying production data for almost 15 years in in accordance with the University of Kentucky
studies covering a multitude of geographic Cooperative Extension Service Bulletins (2008).
locations and experimental requirements as evi- Varying production practices utilized in this
denced by relevant refereed publications in study included planting date, crop variety, plant
numerous journals. When coupled with the vali- density, row spacing, and fertilizer practices
dation specific to the study at hand, as discussed (Table 1).
later, DSSAT was determined to be an appropri- A comprehensive validation was performed
ate model for this study. on the response of yield estimates to varying
The minimum input required to develop production practices and compared with pertinent
yield estimates in DSSAT includes site weather literature. For instance, the response of corn yield
data for the duration of the growing season, site to nitrogen rates exhibited a quadratic response
soil data, and definition of production prac- which was consistent with previous studies (e.g.,
tices. Site weather data were obtained from the Schmidt et al., 2002; Cerrato and Blackmer,
University of Kentucky Agricultural Weather 1990). Also, comparisons of simulated to actual
Center (2008). Daily climatology data were historical yield trends were made for Henderson
collected for 30 years in Henderson County, County, Kentucky. Regression analyses were
Kentucky. Soil data were obtained from a Na- conducted, in which t-tests confirmed that the
tional Cooperative Soil Survey of Henderson simulated yields for both corn and soybeans
County, Kentucky from the United States were not statistically different from the actual60 Journal of Agricultural and Applied Economics, February 2011
historical yields, with a significance of 99%.1 discounts the expected net returns by the var-
Discussions with specialists were also con- iance of net returns.
ducted to confirm that the simulated yield re- The economic model included decision vari-
sults were reasonable. Overall, yield estimates ables, constraints, and other data and coefficients.
were believed to be representative of produc- The decision variables for the model were the
tion in Henderson County, Kentucky. These land area in corn and soybean production. These
yield data were a key element of the economic were identified by alternative production possi-
model. bilities and soil types (Table 1).2 Based on the
decision variables, expected average yields and
The Economic Model net returns were calculated. For the model to
determine these decision variables, constraints
The economic framework of a commercial were required within the model.
Kentucky corn and soybean producer under no- Constraints included land available, labor,
till conditions was embodied in a resource allo- crop rotation, and ratio of soil type. The land
cation model employed within a mean-variance constraint guaranteed that the combined pro-
(E-V) quadratic programming formulation. duction of corn and soybeans did not exceed
The model incorporated risk, as measured by the available land assumed for this study. In
the variance of net returns across years, which addition, agricultural tasks performed in the
was consistent with formulations developed production of both corn and soybeans were re-
by Freund (1956). Specifically, the model was quired. These tasks included: planting, spraying,
modified from Dillon’s (1999) risk management fertilizing, and harvesting which were con-
model to include additional production practices strained by the estimated suitable field hours per
such as nitrogen rate and row spacing. The model week available for performing each operation.
was also modified by allowing multiple weeks The rotation constraint required 50% of the land
for harvesting. In addition, four land types were to produce corn and 50% to produce soybeans.
incorporated within the model. Finally, the in- This represented a 2-year crop rotation typical of
clusion of various auto-steer scenarios distin- a Kentucky grain producer. Furthermore, con-
guished this model from Dillon’s (1999) model. straints were required to ensure that production
The objective of this model was to maximize practices were uniformly distributed across all
net returns above selected costs less the Pratt risk soil types. This implied that variable rate by soil
aversion function coefficient multiplied by the types could not occur.
variance of net returns (referred to hereafter as Besides the constraints, additional in-
expected net returns). The mathematical repre- formation required within the model included
sentation of the model can be found in the Ap- establishing the coefficients, data, and further
pendix. The selected costs incorporated in the assumptions of the model. The coefficients
model included input variable costs (fertilizer, necessary for this investigation included labor
herbicide, seed, hauling, and custom application hours and the prices for corn and soybeans.
of lime, phosphorous, and potassium), operating Labor hours for producing corn or soybeans
costs (labor, fuel, repairs and maintenance, and were based on the field capacities of the oper-
interest on operating capital), and the ownership ating machines. To determine the total labor
cost of auto-steer. The Pratt risk aversion co- required for each operation, a 10% increase in
efficient measured a hypothetical producer’s field capacities was employed. This reflected
aversion to risk and was in accordance with additional labor required for performing other
the method developed by McCarl and Bessler tasks such as travelling from field to field,
(1989). Intuitively, the model represented the refilling the seed bins and sprayer tanks, and
typical risk-return tradeoff in which the model
2 The economic model had the ability to choose
various production alternatives across the allotted
R2 for corn and soybean regression analyses
1 The acres for all scenarios including the base case with
were 0.22 and 0.46, respectively. no auto-steer technologies.Shockley, Dillon, and Stombaugh: Influence of Auto-Steer Navigation 61
unloading the grain bin. Furthermore, the pri- was assumed. As a result, the annualized costs
ces of corn and soybeans were also necessary of sub-meter and RTK auto-steer were $980
for this analysis. Prices for corn and soybeans and $4,900, respectively.4 For the addition of
were determined from the World Agricultural both auto-steer systems, the costs were added
Outlook Board (2008). Prices used were the together for a total investment of $42,000,
2009 median estimates less Kentucky’s basis, with an annualized cost of $5,880 (Griffin,
which resulted in $9.75/bu and $4.25/bu for Lambert, and Lowenberg-DeBoer, 2005, 2008;
soybeans and corn, respectively. Stombaugh, 2009; Stombaugh, McLaren, and
Supplementary data crucial for this inves- Koostra, 2005).
tigation included the proper land area, suitable Finally, there were two assumptions within
field hours, and the cost of auto-steer. The land the model in need of clarification. First, the
area chosen for this study was reflective of amount of area initially overlapped by the sprayer
Henderson County, Kentucky. Henderson County and the amount of inward drift by the implements
ranks second in the state in both corn and soy- attached to the tractor without using auto-steer
bean production (National Agricultural Statistics was assumed. Secondly, there was an increase in
Service Kentucky Field Office, 2008). According the operators work day due to the adoption of
to the Kentucky Farm Business Management auto-steer. Unfortunately, scientific research per-
Program, a 2,600 acre farm corresponded to the taining to these factors was lacking since each
upper one third of all farms in management was operator dependent. Therefore, these factors
returns as represented by net farm income in the were evaluated over a range to provide general
Ohio Valley region of Kentucky, where Hender- economic insight into the profitability of auto-
son County is located (Pierce, 2008). Therefore, steer under two technological scenarios. The
the acreage level assumed for this investigation technical coefficients of the model were varied to
was deemed an appropriate size. Suitable field reflect various overlap and inward drift scenarios.
days were calculated based on probabilities of it The overlap scenarios for the sprayer were varied
not raining 0.15 inches or more per day over from 5 ft to 10 ft and the inward drift of imple-
a period of a month.3 This was determined from ments on the tractor from 0.5 ft to 3 ft. By doing
the 30-year historical climatological dataset pre- so, both field capacity and cost (inputs, labor, and
viously mentioned. The probabilities were mul- fuel) for the relative machinery operations were
tiplied by the days worked in a week and hours affected. The operator’s work day was also
worked in a day to determine expected suitable evaluated for various increases in hours worked
field hours per week. Moreover, the annualized per day which reflected the operator’s ability to
ownership cost of both auto-steer systems in- work longer hours with less fatigue. It was de-
cluded depreciation and the opportunity cost of termined that varying the hours worked per day
capital invested. Depreciation of the auto-steer had no impact on profitability unless the farmer
technologies were calculated using the straight- worked below the original hours assumed for the
line method with an assumed 10-year useful life base case of 13 hours per day. On the other hand,
and no salvage value. The opportunity cost of the results from varying the overlapped area and
capital investment was calculated using an inward drift of the implements were provided in
8% interest rate. A total investment of $7,000 the results section. To address the specific ob-
for auto-steer with a sub-meter receiver and jectives of this study, inquiries into the appro-
$35,000 for auto-steer with an RTK receiver priate overlap of the sprayer, inward drift of
3 An example for determining suitable field days is 4 The annualized cost of an auto-steer technology
given for clarification. Over the 30 year timeframe, the was calculated using the following equation for the
median days in January that it rained 0.15 inches or straight-line depreciation method plus the opportunity
more was five. Therefore, the probability of it NOT cost of capital represented by the average value times
raining 0.15 inches per day in January was (1-(days the interest rate. [((Total Investment – Salvage Value)/
rained/days in the month)). This probability was used to (Useful Life)) 1 ((Total Investment 1 Salvage Value)
estimate the number of suitable field days for the model. Interest Rate)/2].62 Journal of Agricultural and Applied Economics, February 2011
implements, and workday scenario were required by one foot on average per acre. Due to operator
to represent a Kentucky farmer, and discussed in error and inward drift of the planter more rows
the following section. per acre are planted than considered optimal.
Since more rows are planted than optimal,
The Hypothetical Model Farm a farmer could incur a larger seed cost. There-
fore, all operations following planting would be
A base machinery complement was determined drift inward by one foot on average per pass
for a hypothetical grain farm in Henderson since the implements would follow the enterprise
County which practices no-till farming. The row (Stombaugh, 2009). By adopting auto-steer
machinery set included one 250-hp 4WD tractor navigation the above overlaps and inward drifts
with the following implements: a split row no-till could potentially be reduced.
planter (16 rows), a 42-ft anhydrous applicator, When adopting bolt-on auto-steer with a sub-
a grain cart with 500 bushel capacity, and a 20-ft meter receiver on the self-propelled sprayer,
stalk shredder for corn. A 300-hp harvester was a reduction in overlap from 8 ft to 3 ft for pre-
also utilized with an 8 row header for corn and planting operations was utilized (Stombaugh,
a 25-ft flex header for soybeans. A self propelled McLaren, and Koostra, 2005). There was no re-
sprayer with an 80-ft boom that applied herbicides duction in overlaps for post-planting operations
on corn for pre-plant burn down and post planting due to the base accuracy of the planter (one ft
weed control (glyphosate and atrazine), herbicides overlap) since post planting operations would
on soybeans for pre-plant burn down and post follow the enterprise row. When adopting an
planting weed control (glyphosate and 2, 4-D, B), integral valve auto-steer system with an RTK
and insecticide on soybeans (acephate) completed base station, reductions in inward drifts from 1 ft
the equipment set. All equipment specifications to 1 inch were utilized for implements operated
(e.g., speed, width, and efficiency) were from the by the tractor. By utilizing RTK, the total number
Mississippi State Budget Generator (MSBG), of rows planted is reduced; therefore there is the
which complies with the American Society of possibility for seed cost savings (Stombaugh,
Agricultural and Biological Engineering Stan- McLaren, and Koostra, 2005).
dards (Laughlin and Spurlock, 2007). However, The potential benefits of auto-steer not only
both the planter and applicator were added into included a reduction in overlap and inward drift
MSBG with the appropriate data, which were but also an increase in field speed and length
compiled from the Illinois Farm Business Man- of operator’s work day. For the sprayer, it was
agement (Schnitkey and Lattz, 2008) machinery assumed field speeds increased 20% for pre-
operation specifications. For the base case, no planting applications and 10% for post-planting
machines were equipped with any GPS-enabled applications. An increase in speeds were as-
navigation technologies. sumed because of the ability to drive faster during
It was recognized for the base machinery set headland turns and the ability to quickly de-
that, due to operator error and/or fatigue and lack termine which row to enter to continue operating.
of navigational technologies, varying overlaps Speed increases of 5% for planting and 10% for
occurred. The degree of overlap depended on the both fertilizer application and stalk shredding
timing of application (i.e., pre-plant or post- were also assumed (Stombaugh, 2009). With the
plant). For this study, the focus was on the overlap above benefits of both auto-steer systems quan-
of the self-propelled sprayer and the inward drift tified, a percent multiplier was computed and
of the implements attached to the tractor since implemented to calculate the new field capacities
those machines would be most impacted by auto- for the appropriate machines and the reduction in
steer. First, an overlap of 10% of the equipment the impacted input costs (Table 2). Only the re-
width was assumed for the pre-planting opera- duction in overlap and inward drift was consid-
tions of the self propelled sprayer, hence eight ft ered for calculating the multiplier for reduced
of overlap (Griffin, Lambert, and Lowenberg- input costs.
DeBoer, 2008; Palmer, 1989). Next, it was as- In addition, suitable field days were al-
sumed that the planter passes would drift inward tered to represent the adoption of auto-steerShockley, Dillon, and Stombaugh: Influence of Auto-Steer Navigation 63
Table 2. Base Field Capacities, as well as New Field Capacities when Auto-Steer is Adopted on the
Self-Propelled Sprayer and Tractor
Old Field New Field Multiplicative
Implement Capacity (hr/ac) Capacity (hr/ac)a Factorb
Sprayer: Pre-Plant 0.0132 0.0103 0.7792
Sprayer: Post-Plant 0.0132 0.0120 0.9090
Herbicide Cost 0.9306
Planter 0.0491 0.0457 0.9305
Seed Cost 0.9765
Anhydrous Applicator 0.0491 0.0437 0.8892
Nitrogen Cost 0.9776
Stalk Shred 0.0825 0.0715 0.8672
a
New field capacities are calculated according to the changes in width and speed due to the adoption of auto-steer. These field
capacities directly impacted the total labor (hrs) used in each scenario.
b
The multiplicative factor represents the percent change between the new and old field capacities. For example, sub-meter auto-steer
decreased the hours per acre for pre-planting application of chemicals by approximately 22%. The multiplicative factor related to
input cost represents the reduction in cost due solely to the reduction in overlap or inward drift that occurred when adopting auto-steer.
by increasing the operator’s workday from 13 The addition of sub-meter auto-steer increased
hours to 15 hours. This was attributed to the expected net returns under all four risk sce-
ability of the operator to work further into the narios compared with the base without navi-
night with less fatigue (Griffin, Lambert, and gational technology (Table 3). Across all risk
Lowenberg-DeBoer, 2008).5 To determine the aversion levels, the average increase in expec-
influence of auto-steer on net returns, risk, and ted net returns was 0.58% ($2.14/acre).6 Also,
production practices, both old and new field the minimum and maximum net returns were
capacities, as well as suitable field days, were both higher compared with the base scenario
utilized in the economic model. for all risk aversion levels.
The break-even acreage level, payback pe-
Results and Discussion riod, and the return on investment for the
adoption of sub-meter auto-steer were also de-
Three auto-steer scenarios were investigated and termined. To spread out the fixed cost associated
then compared with the base scenario without with sub-meter auto-steer, a land area of
auto-steer navigation: (1) the addition of a bolt- 394 acres was required under the risk neutral
on auto-steer system with a sub-meter receiver scenario.7 According to the 2007 Census of
on a self-propelled sprayer, (2) the addition of an Agriculture, approximately 13% of the grain
integral valve auto-steer system with an RTK farms in Kentucky exceed the break-even acre-
GPS receiver on a tractor, and (3) the addition of age level and would be candidates for sub-meter
both auto-steer systems to the operation. auto-steer for the conditions analyzed (U.S.
Sub-Meter Auto-Steer Results
6 For the risk neutral scenario, if the operator over-
The economic, risk, and production impacts of lapped by only 5 ft, net returns increased by 0.17% with
a return on investment of 29.33%. On the other hand, if
sub-meter auto-steer were first investigated. the operator overlapped by 10 ft, the net returns increased
by 0.79% with a return on investment of 119.04%. Note:
Base overlap for the self-propelled sprayer was 8 ft.
5 Scientific research for determining increased work 7 Break-even acreage level could not be determined
hours is non-existent since it is the farmer’s preference based solely on a calculation. Both the base scenario and
on how many hours are worked each day, but it is known the specific auto-steer scenarios acreage level were varied
that farmers have the ability to work longer hours due to within the model such that both their expected net returns
auto-steer if they wish. Therefore, alterations in suitable converged. Once the net returns for each model con-
field days were modeled after the cited study. verged, the break-even acreage level was determined.64 Journal of Agricultural and Applied Economics, February 2011
Table 3. Economics of Auto-Steer Navigation under Various Risk Aversion Scenarios
Risk Aversion Levels
Basea Neutral Low Medium High
Expected Net Returns $1,020,336 $996,251 $961,761 $908,897
Percent Optimal 100.00% 100.00% 100.00% 100.00%
C.V. (%) 17.81 14.78 13.00 10.90
Minimum Net Returns $658,928 $678,852 $684,355 $706,205
Maximum Net Returns $1,364,489 $1,313,763 $1,235,161 $1,123,673
Sub-Meterb
Expected Net Returns $1,025,835 $1,001,797 $967,527 $914,387
Percent Optimal 100.54% 100.56% 100.60% 100.60%
C.V. (%) 17.72 14.70 12.94 10.84
Minimum Net Returns $664,350 $684,310 $687,922 $711,695
Maximum Net Returns $1,370,054 $1,319,393 $1,241,078 $1,129,163
RTKc
Expected Net Returns $1,023,618 $1,000,645 $964,242 $911,256
Percent Optimal 100.32% 100.44% 100.26% 100.26%
C.V. (%) 17.76 14.84 12.97 10.88
Minimum Net Returns $662,133 $681,971 $688,020 $708,385
Maximum Net Returns $1,367,838 $1,321,461 $1,237,622 $1,126,714
Bothd
Expected Net Returns $1,029,108 $1,006,135 $970,101 $916,747
Percent Optimal 100.86% 100.99% 100.87% 100.86%
C.V. (%) 17.66 14.76 12.90 10.81
Minimum Net Returns $667,623 $687,460 $691,501 $713,874
Maximum Net Returns $1,373,328 $1,326,952 $1,243,633 $1,132,204
a
Base refers to operating without any auto-guidance systems.
b
Sub-Meter refers to the adoption of a bolt-on auto-steer system with a sub-meter receiver on a self propelled sprayer.
c
RTK refers to the adoption of an integral valve auto-steer system with an RTK GPS receiver on a tractor.
d
Both refers to the adoption of both auto-steer systems above and operating together.
Department of Agriculture-National Agricul- In addition to determining the economic
tural Statistics Service, 2010). Also, the payback impact of sub-meter auto-steer, investigating its
period was calculated under the risk neutral potential to become a tool for risk management
scenario, and sub-meter auto-steer was able to was also an objective. For this study, production
pay for itself in 1.08 years for farms with 2,600 risk was measured by the coefficient of variation
acres. Furthermore, sub-meter auto-steer had an (C.V.) of net returns across years. If the adoption
82.56% return on investment.8 of sub-meter auto-steer decreased the C.V. as
well as increased expected net returns when
compared with the base scenario, it could be
8 The following formula was utilized to calculate the
inferred that the technology could be used to
returns on investment: [(Net returns gained from tech-
nology) 1 (Opportunity cost of capital)]/(Total invest- manage production risk. Evidence of reduced
ment in the technology). The opportunity cost of capital risk through sub-meter auto-steer was displayed
was calculated using the following formula: (Interest by more favorable C.V. across risk aversion
rate Total investment/2). The return on invest was
levels when compared with the base scenario.
adjusted such that the opportunity cost of capital was not
accounted for twice. Therefore, the percentages appro- In addition, an increase in expected net returns
priately reflected the return on the capital invested. occurred under all risk scenarios when comparedShockley, Dillon, and Stombaugh: Influence of Auto-Steer Navigation 65
Table 4. Production Results and Acres Planted for Various Risk Aversion Levels under the Base
Scenario with No Auto-Steer Navigational Technologies
Section 1. Corn Management Practices
Risk Aversion Levels
Planting Maturity Nitrogen
Date Group Plant Pop Rate Neutral Low Medium High
March 25 2,600 28,000 150 0 111 359 460
March 25 2,650 24,000 150 0 251 0 0
March 25 2,650 32,000 150 362 0 0 0
March 25 2,700 24,000 100 0 0 0 129
March 25 2,700 24,000 150 0 0 373 0
March 25 2,700 28,000 175 0 501 0 0
March 25 2,700 32,000 175 722 0 0 0
April 1 2,700 24,000 100 0 0 0 231
April 1 2,700 28,000 100 0 0 132 0
April 1 2,700 28,000 150 0 0 83 0
April 1 2,700 32,000 150 0 185 0 0
April 8 2,600 24,000 100 0 0 0 264
April 8 2,700 28,000 150 0 36 119 0
April 15 2,700 28,000 225 216 0 0 0
April 22 2,650 28,000 150 0 216 234 216
Yieldsa 163 156 152 146
Section 2. Soybean Management Practicesb
Risk Aversion Levels
Planting Maturity
Date Group Neutral Low Medium High
April 22 MG3 0 372 553 0
April 22 MG4 1,290 0 0 0
April 29 MG2 0 0 253 913
April 29 MG4 10 672 0 0
May 6 MG4 0 256 0 0
May 13 MG4 0 0 322 0
June 10 MG4 0 0 172 0
June 17 MG4 0 0 0 387
Yields 62 62 60 57
a
Yields for both corn and soybeans are in bu/acre.
b
Optimal plant population and row spacing was the same for all risk scenarios of 111,000 plants per acre and 15 inch row
spacing.
with the base scenario (e.g., 17.729 and risk management. The farm manager’s capability
$1,025,835 under risk neutrality compared with to alter production practices was a large con-
17.81 and $1,020,336). The average decrease in tributor in reducing the C.V.
C.V. due to the adoption of sub-meter auto-steer It was also determined that the optimal
was 0.07%. As a result, it was determined that production practices for the base scenario
sub-meter auto-steer could be used as a tool for (Table 4) were altered when sub-meter auto-steer
was adopted (Table 5). This demonstrated the
importance of a whole farm analysis and the
9 617.72% of mean net returns occur in about 2/3
need to adjust production practices to take full
of the years. C.V. is a relative measure of risk with advantage of the new technology. The major-
a decrease representing a reduction in risk. ity of changes occurred in the production of66 Journal of Agricultural and Applied Economics, February 2011
Table 5. Production Results and Land Area Planted for Various Risk Aversion Levels under Sub-
Meter Auto-Steer
Section 1. Corn Management Practices
Risk Aversion Levels
Planting Maturity Nitrogen
Date Group Plant Pop Rate Neutral Low Medium High
March 25 2,600 28,000 150 0 111 345 460
March 25 2,650 24,000 150 0 251 0 0
March 25 2,650 32,000 150 362 0 0 0
March 25 2,700 24,000 100 0 0 0 129
March 25 2,700 24,000 150 0 0 383 0
March 25 2,700 28,000 175 0 501 0 0
March 25 2,700 32,000 175 722 0 0 0
April 1 2,700 24,000 100 0 0 0 231
April 1 2,700 28,000 100 0 0 127 0
April 1 2,700 28,000 150 0 0 91 0
April 1 2,700 32,000 150 0 185 0 0
April 8 2,600 24,000 100 0 0 0 264
April 8 2,700 28,000 150 0 36 121 0
April 15 2,700 28,000 225 216 0 0 0
April 22 2,650 28,000 150 0 216 233 216
Yieldsa 163 156 152 146
Section 2. Soybean Management Practicesb
Risk Aversion Levels
Planting Maturity
Date Group Neutral Low Medium High
April 22 MG3 0 370 506 0
April 22 MG4 1,300 0 0 0
April 29 MG2 0 0 255 913
April 29 MG3 0 0 0 0
April 29 MG4 0 685 0 0
May 6 MG4 0 245 0 0
May 13 MG4 0 0 405 0
June 17 MG4 0 0 133 387
Yields 62 62 60 57
a
Yields for both corn and soybeans are in bu/acre.
b
Optimal plant population and row spacing was the same for all risk scenarios of 111,000 plants per acre and 15 inch row
spacing.
soybeans. For example, there was a removal of The largest change that occurred due to the
planting on April 29 with maturity group four adoption of sub-meter auto-steer was for medium
under risk neutrality. This was due to the com- risk aversion. Planting the week of June 10 with
petition for suitable field hours during the week maturity group four was removed from the op-
of planting of soybeans on April 22 with spraying timal production set and replaced with planting
post-emergence herbicide on corn planted on the week of June 17 with maturity group four.
March 25. With the ability to spray more ef- This was due to the competition of suitable
fectively with sub-meter auto-steer, more suit- field hours during the week of spraying in-
able field hours were available for planting the secticide on soybeans planted on the week of
highest yielding soybean planting date/maturity May 13 with harvesting the corn planted on
group combination. March 25 with maturity group 2,600. SoybeansShockley, Dillon, and Stombaugh: Influence of Auto-Steer Navigation 67 planted on May 13 exhibited the highest yield compared with the addition of a bolt-on auto- potential of all planting dates in the optimal set. steer system with a sub-meter receiver on a self- Therefore, as sub-meter auto-steer improved the propelled sprayer, expected net returns across all efficiency of the sprayer, more acres were risk levels was lower. However, the minimum planted on the highest yielding planting date and maximum net returns were higher compared (May 13). However, planting more soybeans on with the base scenario for all risk aversion levels. May 13 increased the standard deviation of The break-even acreage level, payback pe- yields, hence net returns, which were penalized riod, and the return on investment for the adop- in the E-V framework. Therefore, substitution tion of RTK auto-steer were also determined. The occurred from planting on June 10, which had the break-even acreage under the risk neutral sce- second largest standard deviation (behind May nario was 1,553 acres. According to the 2007 13), with planting on June 17, which had a lower Census of Agriculture, approximately 1% of the standard deviation with minimal reduction in grain farms in Kentucky exceed the break-even yield (
68 Journal of Agricultural and Applied Economics, February 2011
Table 6. Production Results and Land Area Planted for Various Risk Aversion Levels under RTK
Auto-Steer
Section 1. Corn Management Practices
Risk Aversion Levels
Planting Maturity Nitrogen
Date Group Plant Pop Rate Neutral Low Medium High
March 25 2,600 28,000 150 0 167 374 465
March 25 2,650 24,000 150 0 150 0 0
March 25 2,650 28,000 150 0 45 0 0
March 25 2,650 32,000 150 362 0 0 0
March 25 2,700 24,000 100 0 0 0 128
March 25 2,700 24,000 150 0 0 356 0
March 25 2,700 28,000 175 0 520 0 0
March 25 2,700 32,000 175 722 0 0 0
April 1 2,700 24,000 100 0 0 0 232
April 1 2,700 28,000 100 0 0 121 0
April 1 2,700 28,000 150 0 0 95 0
April 1 2,700 32,000 150 0 177 0 0
April 8 2,600 24,000 100 0 0 0 258
April 8 2,700 28,000 150 0 25 121 0
April 15 2,700 28,000 225 216 0 0 0
April 22 2,650 28,000 150 0 216 234 216
Yieldsa 163 156 152 146
Section 2. Soybean Management Practicesb
Risk Aversion Levels
Planting Maturity
Date Group Neutral Low Medium High
April 22 MG3 0 334 555 0
April 22 MG4 1,300 0 0 0
April 29 MG2 0 0 251 912
April 29 MG4 0 966 0 0
May 6 MG4 0 0 0 0
May 13 MG4 0 0 320 0
June 10 MG4 0 0 174 0
June 17 MG4 0 0 0 388
Yields 62 62 60 57
a
Yields for both corn and soybeans are in bu/acre.
b
Optimal plant population and row spacing was the same for all risk scenarios of 111,000 plants per acre and 15 inch row
spacing.
Production practices under the base scenario of 150 lbs/acre was added to the optimal pro-
(Table 4) were also altered due to the adoption duction set of corn. Additionally, planting soy-
of RTK auto-steer (Table 6). Unlike sub-meter beans on May 6 with maturity group four was
auto-steer, RTK impacted optimal corn pro- removed from the optimal production set of
duction practices as well as soybean production soybeans. Both of these results were attributed to
practices. The largest change occurred under the competition for suitable field hours under
low risk aversion where corn planted on March various production practices. Specifically, there
25 with a maturity of 2,650 growing degree days, was competition during the week of planting
28,000 plants/acre, and nitrogen application rate soybeans on April 29, fertilizing corn planted onShockley, Dillon, and Stombaugh: Influence of Auto-Steer Navigation 69
March 25, and spraying herbicide on corn plan- integral valve auto-steer system with an RTK
ted April 1. RTK auto-steer increased the effi- GPS receiver on a tractor, the average increase in
ciency of tractor operations (both planting and expected net returns across all risk levels was
fertilizing), which allowed more acres of soy- 0.58% ($2.15/acre). Also, the minimum and
beans that could be planted on April 29. There- maximum net returns were higher compared
fore, May 6 was removed from the optimal set. with the base scenario for all risk aversion levels.
Moreover, improvement in fertilizer efficiency The break-even acreage level, payback period,
allowed for more acres to be planted on March and the return on investment for the adoption of
25, therefore March 25 with a maturity of 2,650 both auto-steer systems were also determined.
growing degree days, 28,000 plants/acre, and The break-even acreage under the risk neutral
nitrogen application rate of 150 lbs/acre was scenario was 1,056 acres. According to the 2007
added to the optimal production set of corn. Census of Agriculture, approximately 3% of the
Two interesting results occurred under risk grain farms in Kentucky exceed the break-even
neutrality and medium risk aversion. Similar to acreage level and would be candidates for both
sub-meter auto-steer, planting soybeans during auto-steer systems for the conditions analyzed.
the week of April 29 was also removed from Also, the payback period was calculated under the
the optimal set under risk neutrality. This was risk neutral scenario and it was determined that
attributed to the competition for suitable field the addition of both auto-steer systems would pay
hours during the week of planting soybeans and for themselves in 2.91 years for farms with 2,600
spraying corn. The difference was that RTK acres. Furthermore, the addition of both auto-steer
increased the efficiency of planting whereas systems had a 24.38% return on investment.
sub-meter auto-steer increased the efficiency of These results were superior to operating with only
spraying. However, the change in the production RTK auto-steer but not superior to operating with
of soybeans remained the same. For medium sub-meter auto-steer alone.
risk aversion, planting the week of June 10 with The possibility of utilizing both auto-steer
maturity group four was not removed from the systems for reducing production risk was also an
optimal set like the occurrence with the adoption objective. The addition of both auto-steer sys-
of sub-meter auto-steer. This was because the tems exhibited the greatest ability to reduce risk.
competition for suitable field days was between When compared with other auto-steer scenarios,
spraying soybeans and harvesting corn, neither both coefficient of variations and expected net
of which were influenced by RTK in this study. returns were favored. The average decrease in
In addition, soybean yields were not impacted the C.V. was 0.09%; therefore the addition of
by the utilization of RTK. both auto-steer systems could be used as a risk
management tool. Similar to the first two in-
Sub-Meter and RTK Auto-Steer Results vestigations, the farm manager’s capability to
alter production practices was a large contributor
Similar to the first two investigations, the eco- in reducing the C.V., hence the possibility to
nomic, risk, and production impacts were an- managing production risk.
alyzed for both auto-steer systems operating Alterations in the base production practices of
together. The adoption of both auto-steer systems corn and soybeans (Table 4) due to the adoption
increased the expected net returns under all four of both auto-steer systems were also investigated.
risk scenarios when compared with the base When adding both auto-steer systems, production
scenario (Table 3). Across all risk aversion practices and division of acres were similar as
levels, the average increase in expected net the scenario when adding just RTK auto-steer
returns was 0.90% ($3.35/acre). When compared (Table 7). Only a few differences occurred within
with the addition of a bolt-on auto-steer system the optimal production set. The alterations in
with a sub-meter receiver on a self-propelled production practices when using both systems
sprayer, the average increase in expected net independently were seen when joined together
returns across all risk levels was 0.32% ($1.20/ in this analysis. Notably, there was the removal
acre). When compared with the addition of an of planting soybeans on April 29 under risk70 Journal of Agricultural and Applied Economics, February 2011
Table 7. Production Results and Land Area Planted for Various Risk Aversion Levels under Both
Auto-Steer Scenarios
Section 1. Corn Management Practices
Risk Aversion Levels
Planting Maturity Nitrogen
Date Group Plant Pop Rate Neutral Low Medium High
March 25 2,600 28,000 150 0 167 354 465
March 25 2,650 24,000 150 0 150 0 0
March 25 2,650 28,000 150 0 45 0 0
March 25 2,650 32,000 150 362 0 0 0
March 25 2,700 24,000 100 0 0 0 128
March 25 2,700 24,000 150 0 0 370 0
March 25 2,700 28,000 175 0 520 0 0
March 25 2,700 32,000 175 722 0 0 0
April 1 2,700 24,000 100 0 0 0 232
April 1 2,700 28,000 100 0 0 116 0
April 1 2,700 28,000 150 0 0 104 0
April 1 2,700 32,000 150 0 177 0 0
April 8 2,600 24,000 100 0 0 0 258
April 8 2,700 28,000 150 0 25 123 0
April 15 2,700 28,000 225 216 0 0 0
April 22 2,650 28,000 150 0 216 233 216
Yieldsa 163 156 152 146
Section 2. Soybean Management Practicesb
Risk Aversion Levels
Planting Maturity
Date Group Neutral Low Medium High
April 22 MG3 0 334 507 0
April 22 MG4 1,300 0 0 0
April 29 MG2 0 0 255 912
April 29 MG4 0 966 0 0
May 6 MG4 0 0 0 0
May 13 MG4 0 0 404 0
June 10 MG4 0 0 6 0
June 17 MG4 0 0 129 388
Yields 62 62 60 57
a
Yields for both corn and soybeans are in bu/acre.
b
Optimal plant population and row spacing was the same for all risk scenarios of 111,000 plants per acre and 15 inch row
spacing.
neutrality (occurred under sub-meter and RTK planting soybeans on June 10 as selected under
investigations) and soybeans planted May 6 under the RTK optimal solution or June 17 as selected
low risk aversion (occurred under RTK in- under the sub-meter optimal solution. Therefore,
vestigation). Furthermore, planting corn on March when the auto-steer systems were combined, the
25 with a maturity of 2,650 growing degree days, interactive production effects were seen. When
28,000 plants/acre, and nitrogen application rate operating with both auto-steer systems, both
of 150 lbs/acre was added to the optimal pro- planting dates were part of the optimal production
duction set (occurred under RTK investigation). set. However, the noticeable dominance of sub-
More importantly was the competing production meter auto-steer was evident by the minuscule
decision under medium risk aversion between amount of soybeans planted on June 10.Shockley, Dillon, and Stombaugh: Influence of Auto-Steer Navigation 71
Table 8. Expected Net Returns as well as Profitability of Auto-Steer above the Base Case (%) as
Input Prices Fluctuate by the Percentages Indicateda
Sub-Meterb RTKc Bothd
Herbicide Nitrogen Seed Herbicide Nitrogen Seed
20% 0.65 0.36 0.42 0.98 0.91 0.98
10% 0.60 0.34 0.37 0.92 0.88 0.92
0% 0.54 0.32 0.32 0.86 0.86 0.86
210% 0.48 0.32 0.27 0.80 0.85 0.80
220% 0.43 0.30 0.23 0.74 0.83 0.75
a
The percentages indicate the increase in net returns above the base scenario with the same increase in input price.
b
The impact of herbicide price fluctuations on the profitability of sub-meter auto-steer on the self-propelled sprayer when
compared with the base case.
c
The impact of nitrogen and seed price fluctuations on the profitability of RTK auto-steer on the tractor when compared with the
base case.
d
The impact of herbicide, nitrogen, and seed price fluctuations on the profitability of both sub-meter auto-steer on the self-
propelled and RTK auto-steer on the tractor when compared with the base case.
Impact of Input Price on the Profitability greater potential for profitability than herbi-
of Auto-Steer cide price decreases.
Sensitivity analyses were conducted to de- Conclusion
termine the impact of input price fluctuations on
the net returns and profitability of auto-steer A whole farm economic model is used to assess
under the risk neutral scenario. Three inputs three auto-steer scenarios for various risk aver-
were investigated, herbicide, nitrogen, and seed sion levels. First, a general investigation into the
prices. However, only the inputs that each auto- increase in net returns and return on investment
steer scenario impacted were analyzed. Specif- for auto-steer under various overlap and inward
ically, sub-meter auto-steer influences herbicide drift scenarios and hours worked per week are
costs because it was on the sprayer. Addition- conducted. Results indicate that at the lowest
ally, RTK auto-steer influences nitrogen and overlap scenario, sub-meter auto-steer is profit-
seed cost because it was on the tractor. When the able and the return on investment is always
technologies were combined, they influenced all substantially larger than the interest rate. How-
three inputs. Each input price was varied from ever, at the lowest inward drift scenario for RTK,
220% to 20% of the base, ceteris paribus. Net auto-steer was not profitable and the return on
returns for the base case and all three auto-steer investment was lower than the interest rate.
scenarios were observed when input prices were Therefore, if the inward drift of the implements
varied (Table 8). As input prices were varied, the was only 0.5 ft., RTK would not be an eco-
percent increase in profitability above the base nomically viable option for a farm size of 2,600
case due to the adoption of auto-steer was also acres. A base overlap, inward drift, and hours
calculated. For example, when herbicide price worked per day are assumed and a more thor-
increased by 20%, there was an increase of 0.65% ough investigation is conducted.
in profitability over the base scenario due to the The objectives of this study are to determine
adoption of sub-meter auto-steer. As the appro- the economic, risk, and production implications
priate input price increased, auto-steer became due to the adoption of auto-steer. It is sufficient
more profitable. Also, as input price decreased, that the input savings that occur due to the re-
auto-steer became less profitable. When op- duction in total overlapped area be greater than
erating with both auto-steer systems, seed the annual cost of auto-steer for it to be considered
price increases provided greater potential for profitable. For all risk levels, results indicate that
profitability than nitrogen price increases. all three auto-steer scenarios are profitable when
Conversely, nitrogen price decreases provided compared with the base, with the greatest average72 Journal of Agricultural and Applied Economics, February 2011
increase in expected net returns of 0.90% ($3.35/ use, extension personnel facilitating its adoption
acre) for the scenario with both auto-steer sys- and use, researchers analyzing its potential,
tems. In addition, the minimum and maximum net and the industry personnel in marketing of
returns that occur over 30 years are both higher auto-steer.
due to the adoption of auto-steer. Furthermore, the
break-even acreages under all scenarios are less [Received January 2010; Accepted September 2010.]
than 1,555 acres, with a payback period of no
more than 4.5 years. Also, the largest return on
References
investment is 82.5% for sub-meter auto-steer.
The results also demonstrate that regardless
Adamchuk, V.I., T.S. Stombaugh, and R.R. Price.
of the auto-steer scenario or risk aversion level, ‘‘GNSS-Based Auto-Guidance in Agriculture.’’
the coefficient of variation decreases in all but Technical Bulletin No. 46. South Dakota State
one scenario. Nonetheless, when coupled with an University: International Plant Nutrition In-
increase in net returns, auto-steer can be used to stitute, 2008.
manage production risk. However, reduced pro- Cerrato, M.E. and A.M. Blackmer. ‘‘Comparison of
duction risk is based mainly on the farm man- Models for Describing Corn Yield Response to
ager’s capability to alter production practices. Nitrogen Fertilizer.’’ Agronomy Journal 82(1990):
138–43.
The results of this investigation also indicate
Dillon, C.R. ‘‘Production Practice Alternatives for
that the adoption of auto-steer can impact optimal
Income and Suitable Field Day Risk Manage-
corn and soybean production practices. Soybean ment.’’ Journal of Agricultural and Applied
production is impacted the most by the addition Economics 31,2(1999):247–61.
of sub-meter auto-steer navigation. When ana- Dillon, C.R., T.S. Stombaugh, B.M. Kayrouz, J.
lyzing both auto-steer systems together, it is ev- Salim, and B.K. Koostra. An Educational Work-
ident that sub-meter dominates RTK auto-steer shop on the Use of Precision Agriculture as a Risk
when determining the optimal production prac- Management Tool. Precision Agriculture Refer-
tices. However, RTK does influence many facets eed Conference Proceedings, 2005, pp. 861–67.
Ehsani, M.R., M. Sullivan, and J.T. Walker. ‘‘A
of the optimal production set. This is due to the
Method of Evaluating Different Guidance
reduced overlap, reduced inward drift, and speed
Systems.’’ Paper presented at the American
increases which results in an increase in the field Society of Agricultural Engineers, Chicago,
capacity of sprayer and/or tractor. This in turn Illinois, July 26–27, 2002.
influences the competition between resources, Freund, R.J. ‘‘The Introduction of Risk into a Pro-
specifically suitable field hours. Changes in the gramming Model.’’ Econometrica 24(1956):
input price directly affect the expected net returns 253–64.
and the profitability of auto-steer. Although the Gan-Mor, S., R.L. Clark, and B.L. Upchurch.
increase in net returns are not substantial in ‘‘Implement Lateral Position Accuracy under
RTK-GPS Tractor Guidance.’’ Computers and
magnitude, there are other benefits of auto-steer
Electronics in Agriculture 59(2007):31–38.
that cannot be quantified easily (e.g., the ability
Griffin, T.W., D.M. Lambert, and J. Lowenberg-
for the operator to multi-task and less fatigue). DeBoer. ‘‘Economics of Lightbar and Auto-
There exists opportunities for future research in Guidance GPS Navigation Technologies.’’
quantifying these benefits, exploring how the Precision Agriculture Refereed Conference
quality of life of farmers is impacted by auto- Proceedings, 2005, pp. 581–87.
steer, and how this influences its adoption. ———. ‘‘Economics of GPS-Enabled Navigation
When coupled with the ability to reduce Technologies.’’ Paper presented at the Ninth
production risk, these benefits provide sufficient International Conference on Precision Agri-
culture, Denver, Colorado, July 20–23, 2008.
evidence that auto-steer is a sound investment
Jones, J.W., G. Hoogenboom, C.H. Porter, K.J.
for this case study. The results demonstrate the Boote, W.D. Batchlelor, L.A. Hunt, P.W.
potential of auto-steer to enhance net returns, Wilkens, U. Singh, A.J. Gijsman, and J.T.
reduce risk, and enable adjustments in opti- Ritchie. ‘‘The DSSAT Cropping System Model.’’
mal production practices. This has implica- European Journal of Agronomy 18(2003):
tions for farmers considering its adoption and 235–65.Shockley, Dillon, and Stombaugh: Influence of Auto-Steer Navigation 73 Keicher, R., and H. Seufert. ‘‘Automatic Guidance Schnitkey, G., and D. Lattz. Illinois Farm Business for Agricultural Vehicles in Europe.’’ Com- Management. Internet site: http://www.farmdoc. puters and Electronics in Agriculture 25(2000): uiuc.edu/manage/machinery/machinery_summary. 169–94. html (Accessed December 2008). Laughlin, D.H., and S.R. Spurlock. Mississippi Stombaugh, T.S. Personal Communication. Uni- State Budget Generator v6.0, 2007. versity of Kentucky Department of Biosystems Lowenberg-DeBoer, J. ‘‘Risk Management Po- and Agricultural Engineering, May 2009. tential of Precision Farming Technologies.’’ Stombaugh, T.S., B.K. Koostra, C.R. Dillon, T.G. Journal of Agricultural and Applied Economics Mueller, and A.C. Pike. ‘‘Implications of To- 31,2(1999):275–85. pography on Field Coverage When Using McCarl, B.A., and D. Bessler. ‘‘Estimating an GPS-Based Guidance.’’ Precision Agriculture Upper Bound of the Pratt Risk Aversion Co- Refereed Conference Proceedings, 2007, pp. efficient When the Utility Function is Un- 425–32. known.’’ Australian Journal of Agricultural Stombaugh, T.S., D. McLaren, and B. Koostra. Economics 33(1989):56–63. ‘‘The Global Positioning System.’’ Technical National Agricultural Statistics Service Kentucky Bulletin No. AEN-88. University of Kentucky Field Office. Kentucky Agricultural Statistics Cooperative Extension Service, 2005. 2007–2008 Bulletin. Washington, DC: U.S. Stombaugh, T.S., and S.A. Shearer. ‘‘DGPS-Based Department of Agriculture, 2008. Guidance of High-Speed Application Equipment.’’ Oriade, C.A., and M.P. Popp. ‘‘Precision Farming as Paper presented at the conference of American a Risk Reducing Tool: A Whole-Farm Inves- Society of Agricultural Engineers, Sacramento, tigation.’’ Paper presented at the Fifth In- California, July 29–August 1, 2001. ternational Conference on Precision Agriculture, University of Kentucky, College of Agriculture- Minneapolis, Minnesota, July 16–19, 2000. UKAg Weather Center. Internet site: http:// Palmer, R.J. ‘‘Techniques of Navigating in a Farm wwwagwx.ca.uky.edu (Accessed June, 2008). Field.’’ Navigation: Journal of The Institute of University of Kentucky Cooperative Extension Navigation 36,4(1989):337–44. Service Bulletins. AGR1, AGR129, AGR130, Pierce, J.S. Kentucky Farm Business Management AGR132, ID139 Bulletins. Internet site: http:// Program: Annual Summary Data 2008. Lex- dept.ca.uky.edu/agc/pub_area.asp?area5ANR ington, KY: University of Kentucky Cooperative (Accessed June 2008). Extension Service, 2008. U.S. Department of Agriculture-National Agri- Reid, J.F., Q. Zhang, N. Noguchi, and M. Dickson. cultural Statistics Service. 2007 Census of Ag- ‘‘Agricultural Automatic Guidance Research in riculture. Washington, DC: USDA, 2010. North America.’’ Computers and Electronics in U.S. Department of Agriculture-Natural Re- Agriculture 25(2000):155–67. sources Conservation Service. Internet site: Schmidt, J.P., A.J. DeJoia, R.B. Ferguson, R.K. http://www.nrcs.usda.gov (Accessed June, 2008). Taylor, R.K. Young, and J.L. Havlin. ‘‘Corn World Agricultural Outlook Board. ‘‘World Ag- Yield Response to Nitrogen at Multiple In- ricultural Supply and Demand Estimates.’’ Field Locations.’’ Agronomy Journal 94(2002): Technical Bulletin No. WASDE-463. U.S. 798–806. Department of Agriculture, 2008.
You can also read
