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 (4) Citing Articles via Google Scholar Google Scholar Articles by Zillmann, E. Articles by Claupein, W. Search for Related Content PubMed Articles by Zillmann, E. Articles by Claupein, W. Agricola Articles by Zillmann, E. Articles by Claupein, W. Related Collections Nitrogen Spatial Variability Nutrient Management Site-Specific Analysis

Nitrogen Management

Assessment of Cereal Nitrogen Requirements Derived by Optical On-the-Go Sensors on Heterogeneous Soils

^a Institute of Crop Production and Grassland Research (340), Univ. of Hohenheim, Fruwirthstr. 23, D-70599 Stuttgart, Germany
^b Mississippi State Univ., Agricultural and Biological Engineering, Howell Hall, P.O. Box 9632, Mississippi State, MS 39762

^* Corresponding author (graeff{at}uni-hohenheim.de)

Received for publication September 2, 2005.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Variable N management is one of the most promising practices of precision agriculture to optimize nitrogen-use efficiency (NUE) and decrease environmental impact of agriculture. The objective of this study was to test the performance of fertilization in winter wheat (Triticum aestivum L.) and triticale (Triticosecale Wittm.) determined by reflection measurements of on-the-go sensors under heterogeneous field conditions. In 2004 geo-referenced yield and N fertilization data were collected in four heterogeneous fields in southern Germany. Nitrogen demand of plants was determined throughout the growing season and the corresponding amount of N fertilizer was broadcast with the N-Sensor (Yara, Germany) in real-time. The sensor uses the red edge position (720–740 nm) as an indicator of crop N status and relates this to crop N demand. The sensor algorithm is designed to stimulate plant growth in areas with low biomass and reduce risk of lodging in areas with high biomass. Fertilization was evaluated by calculating site-specific N balance maps to delineate zones with N surplus in the soil. The results revealed some general limitations of this sensor approach in areas with yield-limiting factors other than N. Nitrogen surplus above 50 kg ha^–1 was calculated for subfield areas dominated by shallow soils. The results of this study indicated that sensor-based measurements can be used efficiently for variable N application in cereal crops when N is the main growth-limiting factor. However, the causes for variability must be adequately understood before sensor-based variable rate fertilization can safely be used to optimize N side-dressing in cereals.

Abbreviations: GIS, geographical information system • GPS, global positioning system • NDVI, normalized difference vegetation index • NIR, near infrared • NUE, nitrogen-use efficiency • TR, triticale • VIS, visible • WW, winter wheat

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
ADVANCES in information technology and their application in crop production, also known as precision agriculture, are creating the potential for substantial change in management and decision-making in agriculture. Over the last 10 yr, research has provided evidence of heterogeneous fields leading to non-uniformity in crop yields at the subfield scale. It has been the purpose of research to better understand the causes of inherent within-field variability and to offer appropriate sensor technologies to optimize inputs such as N throughout the growing season.

Nitrogen fertilization rates in cereal production systems are generally applied uniformly based on field-level average soil available N status and a specified N requirement based on the grain yield goal. It is hard to predict optimal N rates because they can be highly variable between seasons, depending on weather conditions and soil N supply. Because N is an important nutrient in determining crop growth and yield (van Keulen et al., 1989) many farmers apply excess N to avoid N deficiency. If N is applied in excess, plant biomass and plant density become too large, which might increase the risk of lodging and inefficient light interception, leading to lower yields in crops such as wheat (Stokes et al., 1998).

However, most fields contain a complex arrangement of spatial and temporal variability in soil properties and crop productivity. As N interacts more extensively with the biological, chemical, and physical properties of soils than any other nutrient and since it is used in greater quantities by crops, N management is a primary consideration when discussing site-specific management (Mulla and Schepers, 1997). Nitrogen demand of crops varies spatially across fields and landscapes due to spatial differences in soil conditions (LaRuffa et al., 2001). Hence, site-specific N application needs information on crop growth and N demand at a high spatial and temporal resolution. Tools that are able to deliver this information have to be efficient and low cost (Hoskinson et al., 1999). Spatially variable N management according to local soil conditions promises to increase crop productivity and NUE, while decreasing the risk potential for environmental pollution (Hatfield and Prueger, 2004). The potential economic and environmental benefits of site-specific N application have been widely reported (Haneklaus and Schnug, 2002).

The objective of variable N management is to adjust the amount of N to the varying yield potential of sub-areas of fields to optimize yield, farmer's profits, and NUE. Since existing methods of soil and plant analysis have proven to be too costly and time consuming (Long et al., 1998) to fulfill this requirement, the focus is shifting from map-based variable rate application to approaches using remote sensing technologies. Technology for on-the-go optical sensing and optimum N rate application was introduced in the USA by Stone et al. (1996) and in Germany by Heege and Reusch (1996) and is currently available commercially. Spectral measurements are taken in particular wavelengths of the visible (VIS) and near infrared (NIR) region of the electromagnetic spectrum, where the spectral characteristics of plants are good indicators of their health and N content in the tissues (Hinzman et al., 1986). Chlorophyll is the most important factor affecting reflectance in the VIS of most field crops, but it has no influence on the reflection properties in the NIR. Several mathematical combinations of spectral information have been found to be good descriptors of N uptake and chlorophyll concentration. The normalized difference vegetation index (NDVI) (Rouse et al., 1974) and especially the red edge inflection point (REIP) (Guyot et al., 1988) were highly correlated with N uptake (Heege and Reusch, 1996; Reyniers et al., 2004). Several studies have been made to use real-time sensor-based spectral measurements to derive N fertilizer requirements of crops (Solie et al., 2002; Link et al., 2002). These studies revealed promising results with regard to increasing yields, creating more homogenous protein content and crop stands as well as increasing NUE and reduced N surplus (e.g., Link et al., 2004). However, in some other studies, it has not been possible to determine positive effects (Jørgensen and Jørgensen, 2001; Kilian et al., 2001). A reason for this is probably the fact that current available sensor technologies rely on the assumption that the spectral signature of leaves and canopies can be related to one single influencing factor (Blackmer and White, 1996). However, other nutrient disorders (Osborne et al., 2002a), water (Osborne et al., 2002b), temperature, and diseases (Schnug et al., 1998) are just a few examples of possible stress factors that affect leaf chlorophyll content and thus spectral signatures. As current remote sensing technologies lack the ability to clearly discriminate plant stress symptoms, the use of these technologies might lead to excessive N applications within for example, water stress regions of the fields, if the system does not account for water stress.

Considering the political encouragement of farmers to use sensor technologies for variable rate N application to decrease environmental pollution, the objective of the presented study was to show, that under certain field conditions the utilization of sensor determined N fertilizer recommendations can lead to decreased NUE in cereals and consequently may lead to a high environmental pollution risk.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Site Description
Experiments were conducted in 2004 on four different fields (Lehmgrube, Mohren, Stockacker, Schafweide) with moderate soil texture variability at the experimental station Ihinger Hof (48° 44' N 8° 56' E, 693 mm average precipitation, 8.1°C mean annual temperature) of the University of Hohenheim, Stuttgart, Germany. The mean monthly temperature during 2004 was consistent with long-term mean temperatures (Fig. 1 ). The monthly sum of precipitation was up to 48% less than the long-term mean precipitation, especially during the main vegetation period from April to June.

View larger version (28K): [in this window] [in a new window]   Fig. 1. Monthly mean temperature and mean precipitation in the region of Ihinger Hof over a 28-yr period (1977–2004) compared to 2004. Data were collected from a weather station at the research farm (48°44' N, 8°56' E).

Nitrogen Fertilization and Yield Mapping
In 2003 three of the test fields were planted with triticale (TR) (Triticosecale Wittm.) cv. Lamberto and one field was planted with winter wheat (WW) (Triticum aestivum L.) cv. Dekan. Triticale was sown in all fields by end of September 2003 with a planting density of 250 seeds per m², while winter wheat was sown around mid of October 2003 with a density of 350 seeds per m². All fields were plowed and harrowed shortly before sowing. The field sizes varied between 4 and 10 ha and the relief was homogeneous with little elevation differences from field to field in a range of 7 to 18 m (Table 1).

Triticale and WW were harvested at the beginning/mid of August 2004. Grain yield was determined by a John Deere combine-mounted yield mapping system using differential GPS to ensure high spatial accuracy of yield determination. Yield values were recorded every 3 s at a speed of 6.4 kmh (1.9 m s^–1). Unrealistic yield values were removed according to the procedure of Haneklaus et al. (2000) before interpolation using inverse distance algorithm.

Post-harvest N balance maps were produced using a GIS to show regions of low NUE, expressed as grain production per unit of N applied, and high residual soil N within the field. The computed N surplus was used as an indicator for site-specific N loss potential. The site-specific N balance map was calculated as follows:

[1]

where N_s is the N left in soil after harvest, N_MF is the amount of mineral fertilizer, N_min is the assumed N content of soil at the beginning of the vegetation period in spring 2004 and N_D is the N deposition out of the atmosphere. The crop N uptake (N_p) was calculated using estimated values published in the literature (Table 3) or measured values if available. The N_min content was determined at the beginning of the growing season as 40 kg N ha^–1 over all fields, not considering in-field variability. Nitrogen deposition of 19 kg N ha^–1 was observed at the experimental station.

The position of the red edge (720–740 nm) is used as an indicator of a quantity called ‘chlorophyll-index’, to make estimates of the chlorophyll content and biomass of a crop. The software then relates this to N need of the crop using proprietary relationships to establish an N application rate (Reusch, 1997). The algorithm for determination of crop N demand is designed to stimulate plant growth in areas with low biomass and reduces risk of lodging in areas with high biomass. Thus, the N-Sensor system should help the farmer to ensure that the optimum N rate is achieved in all areas of the field, and minimize the variability in yield over a whole field (Hydro Agri Precise, 2000).

Statistics
Measured yield and N fertilization point data were interpolated by geo-statistical standard routines implemented in ArcGIS. Correlation of yield and N application maps was analyzed using the Pearson correlation coefficient (r).

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
The results of the study indicated substantial spatial variability of cereal yields within the fields in a range between 10.1 and 19.7% of the mean yield. Results of grain yield (t ha^–1) are shown in Table 4. Mean grain yields varied between 6.6 and 8.1 t ha^–1 with the highest mean yield in the field Lehmgrube, where 184 kg N ha^–1 had been applied to winter wheat. The lowest mean yield was obtained in the field Stockacker, where 138.9 kg N ha^–1 resulted in 6.6 t ha^–1 of TR. Among the fields cropped with TR the field Stockacker received the highest N rates. The field Stockacker had the highest coefficients of variation in grain yield over all fields and resulted in a high correlation between N amounts and grain yield. In contrast, the field Schafweide received an average amount of 112.5 kg N ha^–1 which was the lowest N rate over all fields, and resulted in the highest grain yield for TR.

View larger version (32K): [in this window] [in a new window]   Fig. 2. Yield monitor data [dt ha–1 (1 dt = 100 kg) of triticale 2004 on the test field (A) Schafweide corresponded only in a few spatial patterns with the (B) N application map. Post-harvest N surplus in the soil is (C) negligible.

Similar to the field Schafweide, the test field Lehmgrube also revealed a partly good determination of sufficient N supply at time of fertilization. A comparison of the spatial distribution of variable N amounts (Fig. 3B ) with those of winter wheat grain yield (Fig. 3A) indicated low fertilization in high yielding zones and high N fertilizer amounts in N deficient zones.

View larger version (36K): [in this window] [in a new window]   Fig. 3. Yield monitor data of winter wheat 2004 on test field (A) Lehmgrube indicated similar spatial patterns as the (B) N application map. Low yielding zones lead to (C) a high post-harvest N surplus in the soil.

In the southern part of the field, high amounts of N were applied due to low biomass development at the time sensor measurements were taken. However, final grain yields were low in this area, leading to low NUE and high N surplus (Fig. 3C). On 2.5 ha of the field more than 50 kg N ha^–1 were probably left in the soil. Obviously, N was not the main yield-limiting factor in these areas of the field. This observation suggests the assumption that problems might occur with N prescriptions via spectral sensor measurements if N is not the yield-limiting factor. For instance, in dry years, less N is needed in zones with high risk of drought, since optimum yields would be depressed by water limitations.

The test fields Stockacker (Fig. 4 ) and Mohren (Fig. 6) are highly variable in soil texture and partly dominated by shallow rendzina soils associated with limited water supply especially in dry years. The yield map of the test field Stockacker (Fig. 4A) revealed a within-field variability of 19.7% (Table 2). Compared to the N application map, it was noticeable that the high yielding zones near the field border accurately matched the spatial extent of zones with below average fertilization (Fig. 4B). In these zones, high early biomass development was detected accurately by the sensor resulting in an adjusted smaller amount of fertilization between 60 and 100 kg N ha^–1 to prevent lodging. The main part of the field was side-dressed homogeneously with a higher N amount (140–160 kg N ha^–1). However, a high variation in final grain yield was determined in these areas similar to those areas on a lower N level. Sensor measurements before fertilization detected reduced biomass development and chlorophyll concentration resulting in an increased N application rate to create homogeneous crop growth. Nevertheless, the target yield was not possible due to other yield-limiting factors. The spatial extent of zones with high amounts of biomass at the time of fertilizing (Fig. 4B) matched the spatial extent of zones with high yields (Fig. 4A).

View larger version (29K): [in this window] [in a new window]   Fig. 4. Yield monitor data of triticale 2004 on test field (A) Stockacker corresponded only in a few spatial patterns with the (B) N application map. Low yielding zones lead to (C) a high post-harvest N surplus in the soil.

View larger version (41K): [in this window] [in a new window]   Fig. 6. Yield monitor data of triticale 2004 on test field (A) Mohren showed significant spatial correlations between (B) low yielding zones and high N applications. (C) Post-harvest N surplus in the soil.

View larger version (17K): [in this window] [in a new window]   Fig. 5. Baseline soil information of field Stockacker determined by the state soil survey (Reichsbodenschätzung) in 1937.

These observations are confirmed by the results of the test field Mohren (Fig. 6 ). Long-term experiences showed that the within field variability of yield (Fig. 6A) seemed to be mainly driven by soil moisture conditions.

The spatial patterns of the yield map corresponded well with soil texture (Fig. 7 , R = 0.72), resulting water holding capacity (R = 0.61) and different rooting depths (Fig. 8 ; R = 0.66) within the field. Low sand and high silt levels lead to higher yields because of a higher amount of plant available water in these areas. The N application map showed some low fertilized areas in spatial correspondence with high yielding zones (Fig. 6B). Increasing N amounts for low yielding zones lead to high N surplus above 50 kg N ha^–1 in the soil (Fig. 6C). This was the case at almost 19% of the field area. In Fig. 9 the correlation between the measured and estimated N balance is given exemplarily for 16 sampling points in the Mohren field. The correlation of R = 0.87 verifies the goodness of the N balance maps and provides statistical strength to the estimate of resulting N surplus in the different areas of the field. In general, the results indicated that the higher the second application rate, the higher the N surplus at the end of vegetation period.

View larger version (22K): [in this window] [in a new window]   Fig. 7. Spatial distribution of soil texture on test field Mohren. Soil texture was analyzed at 135 sample points within the upper 30 cm. The map was interpolated by using the spline algorithm of ArcGIS Spatial Analyst.

View larger version (31K): [in this window] [in a new window]   Fig. 8. Rooting depth at 16 sampling points with additional map of N balance of field Mohren. Numbers indicate the determined rooting depth in cm.

View larger version (14K): [in this window] [in a new window]   Fig. 9. Correlation of N-balance (kg N ha–1) point data (measured and estimated) of the field Mohren.

The examples presented revealed that N recommendations should aim at adjusting application rates more closely to the local yield potential rather than making crops more homogeneous. Spectral measurements seem to be of little use for N management in generally low yielding zones. In these zones it can be assumed that in most cases N is not the main limiting factor, especially when the determined patterns of spatial yield variability are stable over years.

The results of the study revealed that spectral measurements performed with the N-Sensor predicted N demand of plants quite accurately when N was the growth limiting factor in a field. However, the results also suggest that spectral measurements performed with the N-Sensor are not useful when yield-limiting factors other than N might exist. The fertilizing strategy of the N-Sensor system mainly focused on creating a homogeneous yield all over the field, leading to an increased N fertilizer rate in areas of low biomass and corresponding low chlorophyll content. This procedure does not take into account, that N might often not be the main yield-limiting factor especially in areas with shallow soils where yield largely depends on soil moisture conditions. Thus, resulting recommendations might not be optimal, because unexpected water deficits could occur during the growing season. In that case the use of sensor-based prescriptions lead to low NUE, high residual N in the soil and an increased risk of groundwater pollution by N leaching. From an environmental viewpoint the sensor-based recommendations should aim at adjusting application rates more closely to local yield potential rather than making crops more homogeneous (Ebertseder et al., 2003). The results of this study are in line with those of Schnug et al. (1998), who concluded that the crucial problem of spectral reflection measurements of crop canopies is that differences in spectral characteristics can have more than one cause. On-the-go sensors determine the crop N demand by measuring the reflection properties of crop canopies which can be indirectly related to crop N status (Heege and Reusch, 1996; Reyniers et al., 2004). This method of N-deficiency detection is based on the color of crop leaves, knowing that color changes from yellow-green to blue-green with increased N uptake and chlorophyll concentration. Changes of the spectral signature can only be assumed to be caused by a variation of N supply in the soil, because factors such as drought, diseases, and deficiencies of other minerals also affect the green color of plants (Masoni et al., 1996; Haneklaus and Schnug, 2002). Thus, changes in leaf color due to for example, water stress and N deficiency will result in the same sensor-based N fertilizer recommendation and might be wrong especially in fields were spatial and temporal stable yield patterns have been observed over multiple years.

In such fields the true causes of yield variation are not identified and the subsequent management changes will be ineffective and unprofitable. Farmers should be aware of this fact and carefully decide on the use of these technologies in fields with spatially and temporally stable yield-limiting factors. Causes for spectral variability within the field must be adequately understood, before sensor-based variable rate fertilization can safely be used to reduce or optimize N side-dressings in cereals. A superior strategy for variable-rate N application would take the history of production of a field, yield maps, precipitation during a growing season or already to the farmer known yield-limiting attributes into account when implementing any N management scheme and combine these information with sensor measurements.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Even if spectral sensor readings are suitable for indirect detection of plant biomass development in cereals, they are beneficial only if the information they provide increase the certainty of correct decisions. The results of this study indicated that sensor-based measurements can be used efficiently for variable N application in cereal crops when N is the main growth-limiting factor. However, if yield patterns are spatially and temporally stable over multiple years, the causes for spectral variability within the field must be adequately understood before sensor-based variable rate fertilization can safely be used to reduce or optimize N side-dressing in cereals. In these cases optical sensor systems should be used carefully, because they can lead to unwanted increases in soil N surplus and thereby increase the risk of groundwater pollution.

	ACKNOWLEDGMENTS

 
We would like to acknowledge the Landesstiftung Baden-Wuerttemberg for funding of the research in the framework of the project "Internationale Spitzenforschung."

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Boden, A.G. 1982. Bodenkundliche Kartieranleitung. 3rd ed. Schweitzerbart'sche Verlagsbuchhandlung, Stuttgart, Germany.
• Aufhammer, W. 1998. Getreide und andere Körnerfruchtarten: Bedeutung, Nutzung und Anbau. p. 508, Ulmer Verlag, Stuttgart, Germany.
• Blackmer, T.M., and S.E. White. 1996. Remote sensing to identify spatial patterns in optimal rates of nitrogen fertilization. p. 33–42. In P.C. Robert et al. (ed.) Precision agriculture. Proc. 3rd Int. Conf. on Prec. Agric., Minneapolis, MN. 23–26 June 1996. ASA, CSSA, and SSSA, Madison, WI.
• Doerge, T.A. 1999. Yield map interpretation. J. Prod. Agric. 12:54–61.
• Ebertseder, T., R. Gutser, U. Hege, R. Brandhuber, and U. Schmidhalter. 2003. Strategies for site-specific nitrogen fertilization with respect to long-term environmental demands. p. 193–198. In A. Werner and A. Jarfe (ed.) Proc. 4th Eur. Conf. on Prec. Agric., 4th, Berlin, Germany. 15–19 June 2003. Wageningen Academic Publ., Wageningen, the Netherlands.
• Food and Agriculture Organization. 1998 World Soil Resources Report 84. World Reference Base for Soil Resources, Rome.
• Frede, H.G., and S. Dabbert. 1998. Handbuch zum Gewässerschutz in der Landwirtschaft. Ecomed Verlag, Landsberg, Germany.
• Guyot, G., F. Baret, and D.J. Major. 1988. High spectral resolution: Determination of spectral shifts between the red and the near infrared. Int. Arch. Photogr. Remote Sens. 11:750–760.
• Haneklaus, S., H. Lilienthal, K. Panten, E. Haveresch, and E. Schnug. 2000. Routines for efficient yield mapping. In P.C. Robert et al. (ed.) Proc. Int. Conf. on Prec. Agric., 5th, Minneapolis, MN. 16–19 July 2000. [CD-ROM]
• Haneklaus, S., and E. Schnug. 2002. An agronomic, ecological and economic assessment of site-specific fertilization. Landbauforsch. Völkenrode 52:123–133.
• Hatfield, J.L., and J.H. Prueger. 2004. Nitrogen over-use, under-use, and efficiency. In Proc. Int. Crop Sci. Congress, 4th, Brisbane, Australia. 26 Sept.–1 Oct. 2004. Available at www.cropscience.org.au (verified 3 Mar. 2006). The Regional Institute, Ltd.
• Heege, J.J., and S. Reusch. 1996. Sensor for on the go control of site specific nitrogen top dressing. ASAE Paper 961018. ASAE, St. Joseph, MI.
• Hinzman, L.D., M.E. Bauer, and C.S.T. Daughtry. 1986. Effects of nitrogen fertilization on growth and reflectance characteristics of winter wheat. Remote Sens. Environ. 19:47–61.
• Hoskinson, R.L., J.R. Hess, and R.S. Alessi. 1999. Temporal changes in spatial variability of soil nutrients. p. 61–70. In J.V. Stafford (ed.) Proc. Europ. Conf. on Prec. Agric., 2nd, Denmark. 11–15 July.
• Hydro Agri Precise. 2000. Precise: A sixth sense for agriculture. Hydro-Agri, Lincolnshire, UK.
• Jørgensen, J.R., and R.N. Jørgensen. 2001. Impact on grain quality parameters when nitrogen is "sensor applied" by the "Hydro Precise system". p. 929–934. In G. Grenier and S. Blackmore (ed.) Proc. Europ. Conf. on Prec. Agric., Montpellier, France. 18–20 June 2001. Agro-Montpellier, Montpellier, France.
• Kilian, B., T.M. Hurley, and G. Malzer. 2001. Economic aspects of precision agriculture: An economic assessment of different site-specific N-fertilization approaches. p. 521–526. In G. Grenier and S. Blackmore (ed.) Proc. Europ. Conf. on Prec. Agric., Montpellier, France. 18–20 June 2001. Agro-Montpellier, Montpellier, France.
• LaRuffa, J.M., W.R. Raun, S.B. Phillips, J.B. Solie, M.L. Stone, and G.V. Johnson. 2001. Optimum field element size for maximum yields in winter wheat, using variable nitrogen rates. J. Plant Nutr. 24:313–325.
• Link, A., J. Jasper, and H.W. Olfs. 2004. Variable nitrogen fertilization by tractor-mounted remote sensing. In Proc. Int. Conf. on Prec. Agric., 7th, Minneapolis, MN. 25–28 July 2004. [CD-ROM]
• Link, A., M. Panitzki, and S. Reusch. 2002. Hydro N-Sensor: Tractor-mounted remote sensing for variable nitrogen fertilization. p. 1012–1018. In Proc. Int. Conf. on Prec. Agric., 6th, Minneapolis, MN. 14–17 July 2002. [CD-ROM]
• Long, D.S., R.E. Engel, and G.R. Carlson. 1998. Grain protein sensing for precision N management: II. p. 56–61. In Proc. of Intensive Wheat Management Conf., Denver, CO. 4–5 March. Potash & Phosphate Inst., Brookings, SD.
• Masoni, A., L. Ercoli, and M. Mariotti. 1996. Spectral properties of leaves deficient in iron, sulfur, magnesium, and manganese. Agron. J. 88:937–943.[Abstract/Free Full Text]
• Mulla, D.J., and J.S. Schepers. 1997. Key processes and properties for site-specific soil and crop management. p. 1–18. In F.J. Pierce and E.J. Sadler (ed.) The site-specific management for agricultural systems. ASA, CSSA, and SSSA, Madison, WI.
• Osborne, S.L., J.S. Schepers, D.D. Francis, and M.R. Schlemmer. 2002a. Detection of phosphorus and nitrogen deficiencies in corn using spectral radiance measurements. Agron. J. 94:1215–1221.[Abstract/Free Full Text]
• Osborne, S.L., J.S. Schepers, D.D. Francis, and M.R. Schlemmer. 2002b. Use of spectral radiance to estimate in-season biomass and grain yield in nitrogen- and water-stressed corn. Crop Sci. 42:165–171.[Abstract/Free Full Text]
• Reusch, S. 1997. Entwicklung eines reflexionsoptischen Sensors zur Erfassung der Stickstoffversorgung landwirtschaftlicher Kulturpflanzen. Forschungsbericht Agrartechnik des Arbeitskreises Forschung und Lehre der Max-Eyth-Gesellschaft Agrartechnik im VDI (VDI-MEG) 303, Kiel, Germany.
• Reyniers, M., E. Vrindts, and J. De Baerdernaeker. 2004. Fine-scaled optical detection of nitrogen stress in grain crops. Opt. Eng. 43:3119–3129.
• Rouse, J.W., R.H. Haas, J.A. Schell, D.W. Deering, and J.C. Harlan. 1974. Monitoring the vernal advancement and retrogradation (greenwave effect) of natural vegetation. NASA/Goddard Space Flight Ctr. Type III Final Rep., Greenbelt, MD.
• Schnug, E., K. Panten, and S. Haneklaus. 1998. Sampling and nutrient recommendations: The future. Commun. Soil Sci. Plant Anal. 29:1455–1462.
• Solie, J.B., M.L. Stone, W.R. Raun, G.V. Johnson, K. Freeman, R. Mullen, D.E. Needham, S. Reed, and C.N. Washmon. 2002. Real-time sensing and fertilization with a field scale Grenseeker Applicator. In Proc. Int. Conf. on Prec. Agric., 6th, Minneapolis, MN. 14–17 July 2002. [CD-ROM]
• Stokes, D.T., R. Sylvester-Bradley, R.K. Scott, R. Clare, J. Hopkinson, G.F. Milford, and S.E. Salmon. 1998. An integrated approach to nitrogen nutrition for wheat. Home-Grown Cereals Authority Project Rep.159. Caledonia House, London.
• Stone, M.L., J.B. Solie, W.R. Raun, R.W. Whitney, S.L. Taylor, and J.D. Ringer. 1996. Use of spectral radiance for correcting in-season fertilizer nitrogen deficiencies in winter wheat. Trans. ASAE 39:1623–1631.
• van Keulen, H., J. Goudriaan, and N.G. Seligman. 1989. Modeling the effects of N on canopy development and crop growth. p. 83–104. In G. Rusell et al. (ed.) Plant canopies: Their growth, form and function. Cambridge Univ. Press, Cambridge, UK.
• Wollring, J., and S. Reusch. 1999. Stickstoff variabel düngen- Sensor misst Versorgung der Pflanzen. In DLZ Agrarmagazin, Sonderheft 10, BLV-Verlag, München, Germany.
• Zadoks, J.C., T.T. Chang, and C.F. Konzak. 1974. A decimal code for the growth stages of cereals. Weed Res. 14:415–421.[CrossRef]

This article has been cited by other articles:

				G. A. Blackburn Hyperspectral remote sensing of plant pigments J. Exp. Bot., March 1, 2007; 58(4): 855 - 867. [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 (4) Citing Articles via Google Scholar Google Scholar Articles by Zillmann, E. Articles by Claupein, W. Search for Related Content PubMed Articles by Zillmann, E. Articles by Claupein, W. Agricola Articles by Zillmann, E. Articles by Claupein, W. Related Collections Nitrogen Spatial Variability Nutrient Management Site-Specific Analysis