HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Miao, Y. Articles by Wiebers, M. Search for Related Content PubMed Articles by Miao, Y. Articles by Wiebers, M. Agricola Articles by Miao, Y. Articles by Wiebers, M. Related Collections Nitrogen Crop Models Nutrient Management Production Agriculture Site-Specific Analysis

Site-Specific Analysis and Management

Evaluating Management Zone Optimal Nitrogen Rates with a Crop Growth Model

^a Dep. of Soil, Water, and Climate, Univ. of Minnesota, St. Paul, MN 55108
^b Dep. of Agricultural and Biological Engineering, Mississippi State Univ., Mississippi State, MS 39762
^c Dep. of Biological and Agricultural Engineering, Univ. of Georgia, Griffin, GA 30223
^d Mosaic Crop Nutrition, 616 S Jefferson Ave., Paris, IL 61944

^* Corresponding author (ymiao{at}umn.edu)

Received for publication May 19, 2005.

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Determining MZ (management zone)-specific optimal N rate is a challenge in precision crop management. The objective of this study was to evaluate the potential of applying a crop growth model to simulate corn (Zea mays L.) yield at various N levels in different MZs and estimate optimal N rates based on long-term weather conditions. Three years of corn yield data were used to calibrate a modified version of the CERES-Maize (Version 3.5) model for a commercial field previously divided into four MZs in eastern Illinois. The model performance in simulating corn yield for two hybrids (33G26 and 33J24) at five N levels in two independent years was evaluated. Economically optimum N rates (EONRs) were estimated based on 15 yr of simulation (1989–2003). The model explained approximately 59 and 93% of yield variability during calibration and validation, respectively. The model performed well at non-zero N rates, with most of the simulation errors being <10%. Model-estimated EONR varied from 70 to 250 kg ha^–1. Economic analyses indicated that applying N fertilizer at year-, hybrid-, and MZ-specific EONR had the potential to increase net return by an average of US$49 (33G26) or US$52 (33J24) ha^–1 over a URN (uniform rate N) application at 170 kg ha^–1. Applying average hybrid- and MZ-specific EONRs across years did not consistently improve economic returns over URN application; however, applying the hybrid- and MZ-specific N rates that maximized long-term net returns would improve economic return by an average of US$22 (33G26) and US$14 (33J24) ha^–1.

Abbreviations: EONR, economically optimum nitrogen rate • MZ, management zone • PCM, precision crop management • URN, uniform rate N

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
DIVIDING a field into a few relatively uniform MZs is a practical and cost-effective approach to site-specific crop management with current technology and price relationships. Many approaches to MZ delineation have been proposed and evaluated (Mulla, 1991; Fleming et al., 2000, 2004; Blackmore, 2000; Fraisse et al., 2001a; Khosla et al., 2002; Diker et al., 2004; Chang et al., 2004; Miao et al., 2005); however, the biggest challenge facing the producer is how to manage inputs to optimize profit and reduce environmental contamination. Management zones are generally proxies for crop response zones. For the defined zones to be useful and practical for site-specific management, each zone should show different crop responses to nutrient inputs, and these responses need to be reliably estimated before making management decisions. This challenge has not been addressed very well in the literature due to variability in climate and the complexity of statistical methods required.

With site-specific N management, several approaches can be taken to determine the optimal fertilization rates in different MZs. The first approach involves determining the rate of fertilizers by applying current N recommendations based on soil fertility, moisture, and crop yield potential at the MZ scale (Hornung et al., 2003; Koch et al., 2004; Inman et al., 2005); however, it has been pointed out that current N recommendations may not be suitable for site-specific N management (Pan et al., 1997; Hergert et al., 1997; Anselin et al., 2004), and information on spatial crop response variability was not sufficiently used in developing such recommendations (Hurley et al., 2001; Swinton et al., 2002; Bullock et al., 2002). Current university and industry N recommendations are broad compromises intended for large-scale regional use (Pan et al., 1997). In addition to the difficulties in accurately estimating spatial yield goals, site-specific N recommendations based on soil organic matter have proven to be too simplistic to reflect within-field variability of N availability (Schmidt et al., 2002).

An alternative approach is to use N-rate strips, including a zero-N treatment, traversing a field to quantify N responses across different soil and landscape conditions and determine appropriate precision N management strategies in different MZs (Pierce and Nowak, 1999; Mamo et al., 2003; Hurley et al., 2004). This approach is useful in evaluating site-specific N management zones and variable N rates, but may not be practical for estimating zone-specific optimal N rates, which are affected by weather, cultivar, management, site characteristics, and their dynamic interactions. Many years of data may be required to reliably estimate the zone-specific N rates to be applied across years, even when the cultivars and management practices remain the same.

A third, and promising, approach is to use crop growth models to estimate optimal N rates in different MZs within a field, based on long-term simulations using different historical weather conditions. Process-oriented crop growth models, such as CERES-Maize (Jones and Kiniry, 1986) and CROPGRO (Hoogenboom et al., 1994), can simulate the impacts of genetics, weather, soil, management practices, and their dynamic interactions on crop growth, development, and yield based on C, N and water balance principles (Batchelor et al., 2002). They have been extensively validated and applied under a wide range of environmental conditions (Singh, 1985; Carberry et al., 1989; Jagtap et al., 1993; Kiniry et al., 1997; Garrison et al., 1999; Gungula et al., 2003) and for different purposes (Andresen et al., 2001; Jagtap and Abamu, 2003; Pang et al., 1998). In recent years, crop growth models have also been evaluated and applied in precision crop management (PCM), and have shown promising results (Paz et al., 1998, 1999, 2001, 2003; Moore and Tyndale-Biscoe, 1999; Braga, 2000; Basso et al., 2001; Booltink et al., 2001; Fraisse et al., 2001b; Seidl et al., 2001).

Although designed for homogeneous areas, several strategies have been developed in applying crop models for spatial analysis and modeling. The first approach divides a field into grids, and each grid cell is simulated, as demonstrated by Paz et al. (1999). The second approach runs simulations at different points within a field, and then interpolates the simulation results into surface maps for the whole field (Booltink and Verhagen, 1998). The third approach divides the field into a few MZs, and then runs simulations for each relatively homogeneous zone. For example, Basso et al. (2001) classified a Normalized Difference Vegetation Index derived from an aerial image taken in the summer (18 July) into three zones and ran simulations for each zone. Van Alphen (2002) used the WAVE (Water and Agrochemicals in soil and Vadose Environment) model to optimize N management in different MZs for winter wheat (Triticum aestivum L.). This third approach can greatly reduce the cost of collecting spatial model parameters and simplifies the application of crop growth models in PCM.

The objectives of this study were to: (i) calibrate the CERES-Maize model for within-field MZs with 3 yr of historical spatial yield data, (ii) evaluate the performance of the model in simulating yield for 2 yr independent of the calibration years, and (iii) estimate the optimum N rate using 15 seasons of historical weather data (1989–2003).

	MATERIALS AND METHODS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Study Site and Data Collection
This study was conducted on a relatively flat, 12.5-ha field in Paris, IL, managed using no-till practices and a corn–soybean [Glycine max (L.) Merr.] rotation since 1991. Relative elevation differed by only 1.97 m across the field. An Order 1 soil survey (1:8000) conducted by the USDA-NRCS in Illinois reveals that the dominant soils are Drummer silty clay loam (fine-silty, mixed, mesic Typic Endoaquoll), Brenton silt loam (fine-silty, mixed, mesic Aquic Argiudoll), and Raub silt loam (fine-silty, mixed, mesic Aquic Argiudoll). Manure from a hog house pit was applied to this field every other year between 1978 and 1996, and no P fertilizer was applied in most years during this period based on soil test results. The field used to be separated by a fencerow in the middle and was managed as two fields until the mid-1970s. Subsurface tile drainage was installed in this field. Grain yield has been measured since 1995 using a combine equipped with a differential global positioning system receiver and an AgLeader yield monitor (AgLeader Technology, Ames, IA). Corn was planted in 1995, 1997, 1999, 2001, and 2003.

Maps for relative elevation, organic matter, slope, electrical conductivity, spatial trends in yield, and temporal stability in yield were classified into four MZs (Fig. 1 ) using fuzzy cluster analysis as described by Miao et al. (2005). Key soil-landscape properties and spatial and temporal yield variability information in different zones are summarized in Table 1. More detailed information about data collection, processing, and MZ delineation can be found in Miao et al. (2005).

View larger version (65K): [in this window] [in a new window]   Fig. 1. Management zones delineated in the study field.

Model Input Data
Daily historical weather data (maximum and minimum temperature and precipitation) from 1989 to 2003 from the Waterworks weather station in Paris, IL, were obtained from the National Climatic Data Center (Asheville, NC). Solar radiation data for Champaign, IL, were obtained from the Illinois State Water Survey (Champaign, IL). Growing season rainfall data for the past 15 yr (1989–2003) are summarized in Fig. 2 . An Order 1 soil survey (1:8000) conducted by the Illinois USDA-NRCS was used to determine the dominant soil type in each management zone, and the Edgar County Soil Survey (NRCS, 2002) was used to determine soil texture and bulk density in each soil layer. Soil hydraulic properties were estimated according to Ratliff et al. (1983), which is a required input to the CERES-Maize model. The initial saturated hydraulic conductivity was set to an average value of 79.2 cm d^–1. A soil organic matter map was used to adjust soil organic C content in the first three soil layers of each MZ. Management data from 1995 to 2003 were provided by the producer, including planting dates and density, and N fertilizer application rates. Fertilizing and harvesting dates were also available for 2001 and 2003, but were estimated for 1995 to 1997. A constant N rate of 174 kg ha^–1 was sidedressed in early June during the calibration years. The planting dates for 1989 to 1994 were estimated according to April through May weather conditions in each year. The genetic coefficients of 33G26 and 33J24 were provided by Pioneer Hybrid International, Inc. (Johnston, IA). A generic cultivar was used to represent the hybrids used in 1995, 1997, and 1999, since no genetic coefficients for those hybrids were available. The initial soil water content was assumed to be 0.35 and 0.45 m³ m^–3 for the 0- to 90- and 90- to 195-cm soil depths, respectively. Initial soil NH₄ and NO₃ contents were assumed to be 0.5 and 0.3 mg kg^–1, respectively.

View larger version (23K): [in this window] [in a new window]   Fig. 2. Growing season (May–September) rainfall during 1989 to 2003 in Paris, IL.

On-farm N experiments were conducted in this field using a split-plot design, with four replications (blocks) in 2001 and 2003. The main plot consisted of five N rates: 0, 112, 168, 224, and 336 kg N ha^–1. The subplot treatments were two corn hybrids: Pioneer 33G26 (relative maturity 112 d) and 33J24 (relative maturity 112 d). The hybrids were planted side by side using the split-planter technique (Doerge and Gardner, 1999). Each strip (main plot) was 18.24 m wide (60 ft) and run across the whole field (200–400 m long due to the irregular shape of the field). The measured yields for the two hybrids at the five N rates in different MZs in 2001 and 2003 were used to evaluate the performance of the model at different N levels.

Model Simulation
The calibrated CERES-Maize model was used to simulate corn response to N fertilization for both hybrids in each MZ at small N rate increments (70–250 kg ha^–1 at 10 kg ha^–1 increments) for 15 yr (1989–2003) and the N rates that maximized net economic return were determined based on the following formula (Paz et al., 1999):

[1]

where Y is corn yield (kg ha^–1), Pc is the price of corn grain (US$0.09833 kg^–1), N is applied N rate (kg ha^–1), and Pn is the cost of N (US$0.46 kg^–1).

The year-, hybrid- and MZ-specific EONRs were estimated based on model simulation results. Partial economic analyses were performed to evaluate different precision N management strategies, with URN at 170 kg ha^–1 as a base line for comparison.

	RESULTS AND DISCUSSION

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Model Calibration
After adjusting four soil parameters including SCS curve number, effective tile drainage rate, saturated hydraulic conductivity, and available soil water, the model explained 58.7% of corn yield variability across the three calibration years (1995, 1997, and 1999), and the average simulated yield was 10% lower than measured yield. The model performed worst in 1995 (simulation errors >20%), and best in 1999 (simulation errors <10%, except Zone 2).

These results could probably be improved if more information were available concerning initial soil water, NH₄ and NO₃ content in the spring, and actual plant population in each MZ, which were assumed to be uniform across different zones. To determine how such variability may affect the model results for corn yield, the calibrated model's sensitivity to the variability of initial soil water (0.05–0.45 m³ m^–3 at 0.05 m³ m^–3 increments in the top 75 cm) and NH₄ content (0.1–3.0 mg kg^–1 at 0.5 mg kg^–1 increments) and crop population (6–9 plants m^–2 at 0.5 plants m^–2 increments) in each MZ were analyzed for normal, wet, and dry growing seasons. In normal or wet years, variation in the three model inputs was not a concern, because simulated corn yield generally varied <4%; however, in dry years, simulated yield could vary 10% or more. Simulated corn yield was more sensitive to plant population than to initial soil water and NH₄ content (simulated yield generally varied <2%). Management Zone 2 had the highest relative elevation and lowest electrical conductivity (Table 1), so it was drier and much more sensitive to initial soil water conditions than other zones.

Different corn hybrids were planted in the field: FS 6532 was planted in 1995, Asgr 760 and FS 6576 were planted in 1997 in the east and west halves of the field, respectively, and Agrigold 6527 was planted in 1999. Since genetic coefficients for these hybrids were not available, a generic 2700 to 2750 growing degree day hybrid was used in the model, thereby introducing potential errors in simulation results as well.

Model Evaluation
The calibrated model was then used to simulate corn yield at different N rates (0, 112, 168, 224, and 336 kg ha^–1) in each MZ for 2001 and 2003. Averaged across N rates (0–336 kg ha^–1), hybrids (33G26 and 33J24), years (2001 and 2003), and MZs (1–4), the model error during validation was 3.6% (average simulated yield was 9701 kg ha^–1, while the measured yield was 9361 kg ha^–1; Table 2) and the validated model explained >90% of the yield variability (Fig. 3 ). In general, the model performed well at non-zero N levels (112–336 kg ha^–1), with most of the average simulation errors being <10% (Table 2); however, the average simulation error varied from –21.9 to 25.1% at the 0 kg ha^–1 N rate. The model performed better in 2003 than in 2001, with 0.1 and –0.9% average simulation errors in 2003 for 33G26 and 33J24, respectively, compared with 7.3 and 10.9% average simulation errors in 2001 for 33G26 and 33J24, respectively.

View larger version (17K): [in this window] [in a new window]   Fig. 3. Simulated vs. measured corn yield across hybrids (33G26 and 33J24), years (2001 and 2003), N application rates (0–336 kg ha–1), and management zones.

Growing season (May–September) precipitation was normal during 2001 (519.9 mm) and wet during 2003 (640.1 mm), while it was dry during the growing seasons of 1999 (311.9 mm), 1997 (400.1 mm), and 1995 (429.5 mm). These precipitation differences may also contribute to an improvement in model performance for 2001 and 2003 compared with calibration years, because simulated corn water uptake of three crop growth models (CropSyst, CERES-Maize, and Erosion Productivity Impact Calculator [EPIC]) was found to be consistently more accurate under wet than dry conditions (Jara and Stockle, 1999).

Manure was applied in this field in 1994 and 1996, but the detailed application information was not available, and N contributions from manure application were not simulated in the model. This may explain the poor model performances in 1995 and 1997, since these 2 yr would be affected the most. The validation years (2001 and 2003) were at least 5 yr after any manure application and would not be significantly affected, which may be another reason for the improved validation results.

At the MZ level, simulated yield errors were larger when no N fertilizer was applied, varying from –34.1 to 111.6% (Table 2). At the other N rates, the simulated error was much smaller, varying from –11.6 to 20% (Table 2). The spatial variation in simulated yield error was expected, because the model was only calibrated for four soil hydrologic properties, without incorporating the spatial variation across the landscape in soil water redistribution, corn population, and soil residual N content. When no N fertilizer was applied, the model performance was more limited by information about spatial variability in N mineralization rate and soil residual N than when N was not limited, because these were the main sources of N under such situations. In MZ 2 and 4, simulated yields at 0 kg ha^–1 N were consistently higher in 2001, but consistently lower in 2003, compared with measured yield. There may be significantly higher N mineralization in 2003 than 2001, and the model may not simulate that variation well. Although the model has functions to adjust N mineralization rates according to soil water content and temperature (Godwin and Jones, 1991), they may need to be improved. The model consistently overestimated corn yield in MZ1, but underestimated yield in MZ3 (Table 2). Probably two other parameters, SLNF and SLPF, should also be optimized during the calibration process. The SLNF factor can be used to adjust the potential N mineralization rate for specific soils and the SLPF factor can be used to adjust potential daily canopy photosynthesis due to unknown soil or environmental factors that persist at a specific location (Tsuji et al., 1994).

Another important factor that may contribute to the simulation errors is the impact of a corn–soybean rotation, the beneficial effect of which has long been recognized. Increased N availability to corn has usually been identified as an important factor, in addition to other factors like reduced pest problems (Varvel and Wilhelm, 2003). The reported amount of N contribution from soybean to corn in rotations has varied from 30 to 75 kg ha^–1 N (Blevins et al., 1990; Ding et al., 1998; Varvel and Wilhelm, 2003). Oberle and Keeney (1990) reported that N contributions from soybean to first-year corn varied with soil type. Soybean yields have been found to vary both spatially and temporally within a field (Jaynes and Colvin, 1997), so N credits from soybean may also vary spatially and temporally based on the rule of thumb that 56 kg (1 bu) of soybean yield can contribute 0.454 kg (1 lb) of N to the subsequent corn crop (Pan et al., 1997). Although both CERES-Maize and CROPGRO-Soybean models are included in the APOLLO system, they cannot be run together to simulate the corn–soybean rotation effect yet. This feature needs to be included in future improvements.

Temporal Variability of Simulated Optimal Nitrogen Rate and Yield
Growing season (May–September) rainfall during the past 15 yr (1989–2003) varied from as low as 312 mm in 1999 to as much as 874 mm in 1989 (Fig. 1). Based on the Decile Classification Method developed by Gibbs and Maher (1967), 1999 had a very dry growing season, 6 yr (1991, 1992, 1994, 1995, 1997, and 2002) had dry growing seasons, 3 yr (1996, 2000, and 2001) had normal growing seasons, 2 yr (1990 and 2003) had wet growing seasons, while 3 yr (1989, 1993, and 1998) had very wet growing seasons. As a result, the model-simulated EONRs varied from year to year (Fig. 4 and 5) , with CVs varying from 21.4 to 30.0% for 33G26, and from 24.2 to 33.1% for 33J24 (Table 3). The EONRs were highest in 1996 (200–230 kg ha^–1 for 33G26, and 190–230 kg ha^–1 for 33J24), and lowest in 1998 (70–100 kg ha^–1 for 33G26, and 70–110 kg ha^–1 for 33J24). The general trend of EONR across years (Fig. 4 and 5) and the 15-yr average EONRs in each MZ were similar between the two hybrids, but in some years (1989 and 1990), EONRs for 33G26 were significantly higher than for 33J24. These results demonstrate the risks of estimating EONRs using only a few years of data. The estimated EONRs may be vastly different, depending on weather in the years chosen for study. For example, if EONRs in 1993, 1994, 1995, and 1996 were averaged to estimate the N rates to be applied in 1997 and 1998, the N rates would be overestimated by 52.5 to 122.5 kg ha^–1. The 15-yr average EONR for the two hybrids showed similar spatial patterns, with the lowest values in MZ 2 (134 and 133 kg ha^–1 for 33G26 and 33J24, respectively) and the highest values in MZ 3 (165 and 160 kg ha^–1 for 33G26 and 33J24, respectively), with a difference of 30 kg N ha^–1. This suggests that hybrid-specific N management may not be necessary for these two hybrids in most years or in some MZs. The EONRs showed the greatest differences among MZs in years with weather patterns showing the greatest departure from normal (1993, 1999, and 2002).

View larger version (17K): [in this window] [in a new window]   Fig. 4. Model-simulated economically optimal N rate (EONR) for hybrid 33G26 in each management zone (MZ) during the past 15 yr (1989–2003).

View larger version (17K): [in this window] [in a new window]   Fig. 5. Model-simulated economically optimal N rate (EONR) for hybrid 33J24 in each management zone (MZ) during the past 15 yr (1989–2003).

View larger version (17K): [in this window] [in a new window]   Fig. 6. Model-simulated yield at economically optimal N rate (EONR) for hybrid 33G26 in each management zone (MZ) during the past 15 yr (1989–2003).

View larger version (17K): [in this window] [in a new window]   Fig. 7. Model-simulated yield at economically optimal N rate (EONR) for hybrid 33J24 in each management zone (MZ) during the past 15 yr (1989–2003).

A third strategy for precision N management is to apply hybrid-specific N rates that could maximize long-term (15-yr) average marginal net returns in each MZ. Figure 8 shows that the 15-yr average returns of the two corn hybrids (33G26 and 33J24) at different N rates differed across MZs. The N rates that maximized the 15-yr average net economic returns were 170, 130, 180, and 160 kg ha^–1 for 33G26 in MZ 1, 2, 3, and 4, respectively, and 200, 150, 190, and 190 kg ha^–1 for 33J24 in the same four MZs, respectively. Averaged across MZs, this approach could increase marginal net return by an average of US$22 and US$14 ha^–1 over URN application for 33G26 and 33J24, respectively. Reducing N rates from 170 to 130 (33G26) and 150 (33J24) kg ha^–1 in MZ 2 increased the net return by an average of US$86 and US$28 ha^–1 for 33G26 and 33J24, respectively. None of the MZs had a 15-yr average net return lower than that for URN application. Therefore, this approach appears to be more profitable than the second approach, which involves applying average EONRs. Figure 8 also shows some unexplained bumps and bounces in average economic return responses to N rate, indicating that more years of simulation may be needed to get smooth responses with this model. More studies are needed to test the reliability of this approach and understand how these average N rates can be fine-tuned each year to further improve economic returns according to early growing season weather conditions, planting dates, or other situations that may have an impact on corn response to N fertilization.

View larger version (22K): [in this window] [in a new window]   Fig. 8. The 15-yr average marginal net return for two hybrids (33G26 and 33J24) in four management zones (MZ1–4) at different N rates based on model simulation.

	CONCLUSIONS

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

The EONRs for the two hybrids were estimated in each MZ during each of the past 15 yr (1989–2003). The EONR values varied from 70 to 250 kg ha^–1, with CVs varying from 21.4 to 30.0%. The 15-yr average EONRs and yield at EONR were very similar between the two hybrids in three of the four MZs. Economic analyses indicated that applying N fertilizer at year-, hybrid- and MZ-specific EONRs had the potential to increase marginal net return by an average of US$49 and US$52 ha^–1 over returns based on URN application at 170 kg ha^–1 for 33G26 and 33J24, respectively. Applying the long-term (15 yr) average hybrid- and MZ-specific EONRs across years did not consistently improve the 15-yr average marginal net return over URN application; however, applying the hybrid- and site-specific EONRs that maximized 15-yr average marginal net returns across years improved return over URN application by an average of US$22 and US$14 ha^–1 for 33G26 and 33J24, respectively. Simulation results consistently showed that MZ 2 would need lower N rates (130–150 kg ha^–1) for both hybrids than other MZs. We conclude that this crop growth model is a valuable tool that has the potential to improve corn N management and economic returns. The APOLLO system needs to be improved to simulate the corn–soybean rotation effect, which can influence optimal N rate estimation. More studies are needed to develop, test, and apply model-based precision N management strategies.

	ACKNOWLEDGMENTS

 
This study was funded by Cargill Crop Nutrition (now Mosaic), Cargill Dry Corn Ingredients, and Pioneer Hi-Bred International, and partly by a block travel grant offered by the Graduate School at the University of Minnesota. We would like to thank Mr. Ron Olson, Kirby Wuethrich, and other employees from the three funding companies for assistance in this study, local farmer Mr. Gene Barkley for his support and cooperation, and USDA-NRCS in Illinois for conducting an Order 1 soil survey of the study field.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Contribution of the Precision Agriculture Center, Univ. of Minnesota.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Andresen, J.A., G. Alagarswamy, C.A. Rotz, J.T. Ritchie, and A.W. LeBaron. 2001. Weather impacts on maize, soybean and alfalfa production in the Great Lakes region, 1895–1996. Agron. J. 93:1059–1070.[Abstract/Free Full Text]
• Anselin, L., R. Nongiovanni, and J. Lowenberg-DeBoer. 2004. A spatial econometric approach to the economics of site-specific nitrogen management in corn production. Am. J. Agric. Econ. 86:675–687.
• Basso, B., J.T. Ritchie, F.J. Pierce, J.W. Jones, and R.N. Braga. 2001. Spatial validation of crop models for precision agriculture. Agric. Syst. 68:97–112.
• Batchelor, W.D., B. Basso, and J.O. Paz. 2002. Examples of strategies to analyze spatial and temporal yield variability using crop models. Eur. J. Agron. 18:141–158.
• Batchelor, W.D., J. Paz, and K. Thorp. 2005. Development and evaluation of a decision support system for precision agriculture. In D.J. Mulla (ed.) Precision agriculture [CD-ROM]. Proc. Int. Conf., 7th, Minneapolis, MN. 25–28 July 2004. Univ. of Minnesota, St. Paul.
• Blackmore, S. 2000. The interpretation of trends from multiple yield maps. Comput. Electron. Agric. 26:37–51.
• Blevins, R.L., J.H. Herbek, and W.W. Frye. 1990. Legume cover crops as a nitrogen source for no-till corn and grain sorghum. Agron. J. 82:769–772.[Abstract/Free Full Text]
• Booltink, H.W.G., B.J. van Alphen, W.D. Batchelor, J.O. Paz, J.J. Stoorvogel, and R. Vargas. 2001. Tools for optimizing management of spatially-variable fields. Agric. Syst. 70:445–476.
• Booltink, H.W.G., and J. Verhagen. 1998. Using decision support systems to optimize barley management on spatial variable soil. p. 219–233. In M.J. Kropff et al. (ed.) Application of systems approaches at the field level. Kluwer Acad. Publ., Dordrecht, the Netherlands.
• Braga, R.N. 2000. Predicting the spatial pattern of corn yield under water limiting conditions. Ph.D. diss. Univ. of Florida, Gainesville.
• Bullock, D.S., J. Lowenberg-DeBoer, and S.M. Swinton. 2002. Adding value to spatially managed inputs by understanding site-specific yield response. Agric. Econ. 27:233–245.
• Calmon, M.A., J.W. Jones, D. Shinde, and J.E. Specht. 1999. Estimating parameters for soil water balance models using adaptive simulated annealing. Appl. Eng. Agric. 15:703–713.
• Carberry, P.S., R.C. Muchow, and R.L. McCown. 1989. Testing the CERES-Maize simulation model in a semi-arid tropical environment. Field Crops Res. 20:297–315.
• Chang, J., D.E. Clay, C.G. Carlson, C.L. Reese, S.A. Clay, and M.M. Ellsbury. 2004. Defining yield goals and management zones to minimize yield and nitrogen and phosphorus fertilizer recommendation errors. Agron. J. 96:825–831.[Abstract/Free Full Text]
• Diker, K., D.F. Heermann, and M.K. Brodahl. 2004. Frequency analysis of yield for delineating yield response zones. Precis. Agric. 5:435–444.
• Ding, W., D.J. Hume, T.J. Vyn, and E.G. Beauchamp. 1998. Nitrogen credit of soybean to a following corn crop in central Ontario. Can. J. Plant Sci. 78:29–34.
• Doerge, T.A. 2002. Variable-rate nitrogen management creates opportunities and challenges for corn producers. Available at www.plantmanagementnetwork.org/pub/cm/review/variable%2Dn/ (verified 25 Jan. 2005). Crop Manage. doi:10.1094/cm-2002-0905-01-RS.
• Doerge, T.A., and D.L. Gardner. 1999. On-farm testing using the adjacent strip comparison method. p. 603–609. In P.C. Robert et al. (ed.) Precision agriculture: Proc. Int. Conf., 4th, St. Paul, MN. 19–22 July 1998. ASA, CSSA, and SSSA, Madison, WI.
• Fleming, K.L., D.F. Heermann, and D.G. Westfall. 2004. Evaluating soil color with farmer input and apparent soil electrical conductivity for management zone delineation. Agron. J. 96:1581–1587.[Abstract/Free Full Text]
• Fleming, K.L., D. Westfall, W. Wines, and M.C. Brodahl. 2000. Evaluating farmer defined management zone maps for variable rate fertilizer application. Precis. Agric. 2:201–215.
• Fraisse, C.W., K.A. Sudduth, and N.R. Kitchen. 2001a. Delineation of site-specific management zones by unsupervised classification of topographic attributes and soil electrical conductivity. Trans. ASAE 44:155–166.
• Fraisse, C.W., K.A. Sudduth, and N.R. Kitchen. 2001b. Calibration of the CERES-Maize model for simulating site-specific crop development and yield on claypan soils. Appl. Eng. Agric. 17:547–556.[ISI]
• Garrison, M.V., W.D. Batchelor, R.S. Kanwar, and J.T. Ritchie. 1999. Validation of the CERES-Maize water and nitrogen balances under tile drained conditions. Agric. Syst. 59:189–200.
• Gibbs, W.J., and J.V. Maher. 1967. Rainfall deciles as drought indicators. Bureau of Meteorol. Bull. 48. Commonwealth of Australia, Melbourne, Australia.
• Godwin, D.C., and C.A. Jones. 1991. Nitrogen dynamics in soil–plant systems. p. 287–321. In J. Hanks and J. T. Ritchie (ed.) Modeling plant and soil systems. ASA, CSSA, and SSSA, Madison, WI.
• Goffe, W.L., G.D. Ferrier, and J. Rogers. 1994. Global optimization of statistical functions with simulated annealing. J. Econometrics 60:65–99.[CrossRef]
• Gungula, D.T., J.G. Kling, and A.O. Togun. 2003. CERES-Maize predictions of maize phenology under nitrogen-stressed conditions in Nigeria. Agron. J. 95:892–899.[Abstract/Free Full Text]
• Hergert, G.W., W.L. Pan, D.R. Huggins, J.H. Grove, and T.R. Peck. 1997. Adequacy of current fertilizer recommendations for site-specific management. p. 283–300. In F.J. Pierce and E.J. Sadler (ed.) The state of site-specific management for agriculture. ASA, CSSA, and SSSA, Madison, WI.
• Hoogenboom, G.J., J.W. Jones, P.W. Wilkens, W.D. Batchelor, W.T. Bowen, L.A. Hunt, N. Pickering, U. Sing, D.C. Godwin, B. Baer, K.J. Boote, J.T. Ritchie, and J.W. White. 1994. Crop models. P. 95–244. In G.Y. Tsuji et al. (ed.) DSSAT Version 3. Vol. 2. Univ. of Hawaii, Honolulu.
• Hornung, A., R. Khosla, R. Reich, and D.G. Westfall. 2003. Evaluation of site-specific management zones: Grain yield and nitrogen use efficiency. p. 297–302. In J.V. Stafford and W. Werner (ed.) Precision agriculture. Wageningen Acad. Publ., Wageningen, the Netherlands.
• Hurley, T., B. Kilian, and H. Dikici. 2001. The value of information for variable rate nitrogen applications: A comparison of soil test, topographical, and remote sensing information. Selected Paper, AAEA Annual Meeting, Chicago, IL. 5–8 Aug. 2001. Available at http://agecon.lib.umn.edu/cgi-bin/pdf_view.pl?paperid=2722&ftype=.pdf (accessed 14 Nov. 2005; verified 18 Jan. 2006). Am. Agric. Econ. Assoc., Ames, IA.
• Hurley, T.M., G.L. Malzer, and B. Kilian. 2004. Estimating site-specific nitrogen crop response functions: A conceptual framework and geostatistical model. Agron. J. 96:1331–1343.[Abstract/Free Full Text]
• Inman, D., R. Khosla, D.G. Westfall, and R. Reich. 2005. Nitrogen uptake across site specific management zones in irrigated corn production systems. Agron. J. 97:169–176.[Abstract/Free Full Text]
• Jagtap, S.S., and F.J. Abamu. 2003. Matching improved maize production technologies to the resource base of farmers in a moist savanna. Agric. Syst. 76:1067–1084.[ISI]
• Jagtap, S.S., M. Mornu, and B.T. Kang. 1993. Simulation of growth, development and yield of maize in the transition zone of Nigeria. Agric. Syst. 41:215–229.[ISI]
• Jara, J., and C.O. Stockle. 1999. Simulation of water uptake in maize, using different levels of process detail. Agron. J. 91:256–265.[Abstract/Free Full Text]
• Jaynes, D.B., and T.S. Colvin. 1997. Spatiotemporal variability of corn and soybean yield. Agron. J. 89:30–37.[Abstract/Free Full Text]
• Jones, C.A., and J.R. Kiniry. 1986. CERES-Maize: A simulation model of maize growth and development. Texas A&M Univ. Press, College Station.
• Khosla, R., K. Fleming, J.A. Delgado, T. Shaver, and D.G. Westfall. 2002. Use of site-specific management zones to improve nitrogen management for precision agriculture. J. Soil Water Conserv. 57:513–518.
• Kiniry, J.R., J.R. Williams, R.L. Vanderlip, J.D. Atwood, D.C. Reicosky, J. Mulliken, W.J. Cox, H.J. Mascagni, Jr., S.E. Hollinger, and W.J. Wiebold. 1997. Estimation of two maize models for nine U.S. locations. Agron. J. 89:421–426.[Abstract/Free Full Text]
• Koch, B., R. Khosla, W.M. Frasier, D.G. Westfall, and D. Inman. 2004. Economic feasibility of variable-rate nitrogen application utilizing site-specific management zones. Agron. J. 96:1572–1580.[Abstract/Free Full Text]
• Mamo, M., G.L. Malzer, D.J. Mulla, D.R. Huggins, and J. Strock. 2003. Spatial and temporal variation in economically optimum nitrogen rate for corn. Agron. J. 95:958–964.[Abstract/Free Full Text]
• Miao, Y., D.J. Mulla, and P.C. Robert. 2005. Combining soil-landscape and spatial-temporal variability of yield information to delineate site-specific management zones. p. 811–818. In J.V. Stafford (ed.) Precision Agriculture '05. Wageningen Acad. Publ., Wageningen, the Netherlands.
• Moore, G.A., and J.P. Tyndale-Biscoe. 1999. Estimation of the importance of spatially variable nitrogen application and soil moisture holding capacity to wheat production. Precis. Agric. 1:27–38.
• Mulla, D.J. 1991. Using geostatistics and GIS to manage spatial patterns in soil fertility. p. 336–345. In G. Kranzler (ed.) Proc. Automated Agric. 21st Century, Chicago, IL. 16–17 Dec. 1991. ASAE, St. Joseph, MI.
• NRCS. 2002. Soil survey of Edgar County, Illinois. NRCS, Champaign, IL, USA.
• Oberle, S.L., and D.R. Keeney. 1990. Factors influencing corn fertilizer N requirements in the northern U.S. Corn Belt. J. Prod. Agric. 3:527–534.
• Pan, W.L., D.R. Huggins, G.L. Malzer, C.L. Douglas, Jr., and J.L. Smith. 1997. Field heterogeneity in soil–plant nitrogen relationships: Implementations for site-specific management. p. 88–99. In F.J. Pierce and E.J. Sadler (ed.) The state of site-specific management for agriculture. ASA, CSSA, and SSSA, Madison, WI.
• Pang, X.P., S.C. Gupta, J.F. Moncrief, C.J. Rosen, and H.H. Cheng. 1998. Evaluation of nitrate leaching potential in Minnesota glacial outwash soils using the CERES-Maize model. J. Environ. Qual. 27:75–85.
• Paz, J.O., W.D. Batchelor, B.A. Babcock, T.S. Colvin, S.D. Logsdon, T.C. Kaspar, and D.L. Karlen. 1999. Model-based technique to determine variable rate nitrogen for corn. Agric. Syst. 61:69–75.
• Paz, J.O., W.D. Batchelor, T.S. Colvin, S.D. Logsdon, T.C. Kaspar, and D.L. Karlen. 1998. Analysis of water stress effects causing spatial yield variability. Trans. ASAE 41:1527–1534.
• Paz, J.O., W.D. Batchelor, and J.W. Jones. 2003. Estimating potential economic return for variable soybean variety management. Trans. ASAE 46:1225–1233.
• Paz, J.O., W.D. Batchelor, and G.L. Tylka. 2001. Method to use crop growth models to estimate potential return for variable-rate management in soybeans. Trans. ASAE 44:1335–1341.
• Pierce, F.J., and P. Nowak. 1999. Aspects of precision agriculture. Adv. Agron. 67:1–85.
• Press, W.H., S.A. Teukolsky, W.T. Vetterling, and B.P. Flannery. 1996. Numerical recipes in C: The art of scientific computing. Cambridge Univ. Press, New York.
• Ratliff, L.F., J.T. Ritchie, and D.K. Cassel. 1983. Field-measured limits of soil water availability as related to laboratory-measured properties. Soil Sci. Soc. Am. J. 47:770–775.[Abstract/Free Full Text]
• Royce, F.S., J.W. Jones, and J.W. Hansen. 2001. Model-based optimization of crop management for climate forecast applications. Trans. ASAE 44:1319–1327.
• Schmidt, J.P., A.J. Dejoia, R.B. Ferguson, R.K. Taylor, R.K. Young, and J.L. Harlin. 2002. Corn yield response to nitrogen at multiple in-field locations. Agron. J. 94:798–806.[Abstract/Free Full Text]
• Seidl, M.S., W.D. Batchelor, J.B. Fallick, and J.O. Paz. 2001. GIS-model based decision support system to evaluate corn and soybean prescriptions. Appl. Eng. Agric. 17:721–728.
• Singh, U. 1985. A crop growth model for predicting corn (Zea mays L.) performance in the tropics. Ph.D diss. Univ. of Hawaii, Honolulu.
• Swinton, S., Y. Liu, and N.R. Miller. 2002. Is yield response site-specific? Revisiting nitrogen recommendations on corn. In Proc. AAEA Annu. Meet., Long Beach, CA. 28–31 July 2002. Available at http://agecon.lib.umn.edu/cgi-bin/pdf_view.pl?paperid=4538&ftype=.pdf (accessed 14 Nov. 2005; verified 18 Jan. 2006). Am. Agric. Econ. Assoc., Ames, IA.
• Tsuji, G.Y., G. Uehara, and S. Balas (ed.). 1994. DSSAT Version 3. Univ. of Hawaii, Honolulu.
• Van Alphen, B.J. 2002. A case study for precision N management in Dutch arable farming. Nutr. Cycling Agroecosyst. 62:151–161.
• Vanotti, M.B., and L.G. Bundy. 1994. An alternative rationale for corn nitrogen recommendations. J. Prod. Agric. 7:243–249.
• Varvel, G.E., and W.W. Wilhelm. 2003. Soybean nitrogen contribution to corn and sorghum in western Corn Belt rotations. Agron. J. 95:1220–1225.[Abstract/Free Full Text]

This article has been cited by other articles:

				Q. Fang, L. Ma, Q. Yu, R. W. Malone, S. A. Saseendran, and L. R. Ahuja Modeling Nitrogen and Water Management Effects in a Wheat-Maize Double-Cropping System J. Environ. Qual., October 23, 2008; 37(6): 2232 - 2242. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Miao, Y. Articles by Wiebers, M. Search for Related Content PubMed Articles by Miao, Y. Articles by Wiebers, M. Agricola Articles by Miao, Y. Articles by Wiebers, M. Related Collections Nitrogen Crop Models Nutrient Management Production Agriculture Site-Specific Analysis