Using Crop Sensors to Improve Corn Nitrogen Management
Crop Insights by John Shanahan, Agronomy Research Manager
Crop Insights by John Shanahan, Agronomy Research Manager
Current nitrogen (N) management practices for corn production systems typically include significant quantities (e.g., 200 to 250 lbs/A) of pre-plant N applied at field-uniform rates. These conventional practices can sometimes result in sizeable fertilizer N losses, especially in extremely wet springs in the Corn Belt (Mathesius and Luce, 2009). In fact, only 30 to 50% of the applied N is recovered by the crop in many cases (Raun and Johnson, 1999). The lost N not only reduces grower profits (through lost fertilizer and reduced yields in N deficient areas), but can also lead to environmental contamination (i.e., nitrate leaching or greenhouse gaseous losses).
Risk of N losses has prompted growers to consider alternative N management strategies that have potential for improving crop N use efficiency (NUE). One such strategy involves the use of crop sensors to measure canopy N status and direct on-the-go and spatially-variable in-season N applications. A variation of this theme involves applying 1/3 to 1/2 of the projected needed N at or prior to planting and using sensors to direct the balance of crop N requirement as in-season applications (Figure 1).
Compared to current N management practices, sensors are better able to account for within field spatial variability and year-to-year changes in rainfall and the soil’s capacity to mineralize and supply N. This method has the potential to better match fertilizer N supply with crop N need. Use of sensors minimizes the potential for over- and underapplications of nitrogen. Research results suggest this approach not only allows the grower to maximize yields and profitability, but also leads to the highest crop NUE and reduced potential for environmental pollution (Hong et al., 2007; Shanahan et al., 2008). This Crop Insights will discuss the active canopy reflectance sensor systems ("crop sensors") commercially available and their potential for improving N management in corn production.
Presently, there are three crop sensor reflectance monitoring systems commercially available to growers: 1) the GreenSeeker® system, 2) the OptRx® system, and 3) the CropSpec® system. All three systems use active sensor technology to measure crop reflectance. This measurement is used to assess canopy greenness and biomass to determine crop N requirements.
Nitrogen is a major constituent of chlorophyll molecules (green pigments in plants used in photosynthesis). Therefore, determining chlorophyll content by canopy reflectance measurements also estimates crop N content. (USDA)
Because each plant expresses the effects of the surrounding soil conditions, it is the best indicator of soil nitrogen status in its microenvironment. Plant sensing, then, provides the best opportunity for characterizing and differentiating the spatial variability of crop N need. Another benefit of using this approach for making within-season N management decisions is that there is little time delay between assessment and interpretation. This offers a significant advantage over more traditional N management approaches involving soil sampling and analysis. A primary disadvantage of using the plant for assessing N need is that it narrows the window of time when N applications can take place to the period from about knee-high to just before tasseling.
By definition, crop canopy reflectance is the fraction of incoming light reflected by the crop canopy. Chlorophyll present in leaves absorbs light in the visible light wavelengths (450-700 nanometers (nm)), with more blue (450-520 nm) and red (630-680 nm) light being absorbed than green light (520-600 nm). This results in higher reflectance in the green band, and is why plants appear green to the human eye. Hence, sensing canopy reflectance at visible light wavelengths can provide a relative measure of leaf chlorophyll content and crop N status.
Compared to visible light, plants absorb much less near-infrared (NIR) light. In other words, plants reflect more light in NIR wavelengths of 700-1400 nm, with percent NIR reflectance increasing as crop biomass increases (another measure of crop chlorophyll / N status). These reflectance characteristics for visible and NIR light of crop canopies are the basis for the development of numerous vegetative indices. One such index is the Normalized Difference Vegetation Index (NDVI) which is calculated using light reflectance of the red and NIR bands. The formula for calculating NDVI is as follows:
NDVI = (NIR - Red) / (NIR + Red).
Values for NDVI range from -1.0 to +1.0. In typical sensing operations output ranges from 0.1 to 0.9, with values ranging from 0.1 to 0.2 for soil surfaces and 0.2 to 1.0 for crop canopies (NDVI values increase as both crop biomass and greenness increase).
The GreenSeeker system (Figure 2) was developed by NTech Industries of Ukiah, CA and is distributed by Trimble Ag of Sunnyvale, CA. The OptRx system (Figure 3) was developed by Holland Scientific of Lincoln, NE and is sold by Ag Leader Technology of Ames, IA. The CropSpec system (Figure 4) is distributed by Topcon Positioning Systems, Incorporated of Olathe, KS.
Active sensors function by emitting their own source of modulated light onto the crop canopy and then measuring the percent of modulated light reflected from the canopy back to the sensor (Figure 3). More specifically, the GreenSeeker and OptRx use light emitting diodes and the CropSpec uses laser diodes for emitting light; reflectance is measured by all sensors using photodiodes.
The reason sensors use modulated light is to distinguish natural sunlight from their own emitted light. This unique feature, accomplished with electrical circuits, allows the sensors to function equally well in conditions ranging from darkness to full sunlight. Operationally, these sensors can be mounted on N fertilizer applicators equipped with computer processing and variable rate controllers so that sensing and fertilization are done in one pass.
Each of these sensors measures canopy reflectance in both visible and NIR wavelengths (i.e., red and NIR bands) from which vegetative indices can be calculated (see NDVI equation). Refer to sensor manufacturers’ supporting documentation for important design features and specific visible and NIR spectral bands which each sensor uses to measure canopy reflectance. Following these instructions will enable the operator to obtain the optimal performance for each sensor.
The GreenSeeker and OptRx sensors have been designed to be mounted and operated so they are level and aligned directly over corn rows as shown in Figures 5 and 6, respectively. Follow the operator’s manual for specific instructions, but generally, distance from the sensors to the crop canopy should be about 2 to 3 feet. This should result in a zone of illumination or "sensor's footprint" on the target of approximately 2 feet wide centered over a given row and aligned perpendicular to the direction of travel (Figure 5).
The width of the variable-rate application should be considered when deciding how many sensors should be installed on a given applicator. A minimum of 2 or 3 sensors on different rows should be used for representing the entire applicator swath width, with more sensors needed as applicator width increases (Roberts et al., 2009).
The CropSpec uses a two-sensor configuration that installs one sensor positioned at an oblique 45° angle on each side of the top of the applicator cab (Figure 7). The sensor footprint is larger (6 to 10 feet) and integrates more crop area (2 to 3 rows wide) than that of the other two commercial sensors. Hence, only one CropSpec sensor configuration is required per fertilizer applicator.
For some of these sensor types, determining the rate of sidedress N requires sensor measurements from an N-sufficient reference, an area of corn plants that has been well fertilized since planting. The general premise for sensor operation is this: the greater the difference in sensor measurements between N-sufficient reference corn and unfertilized or deficiently-fertilized corn, the more N fertilizer is needed. Without this reference to determine a relative difference, there is little basis for making N rate recommendations.
|Establishing an N-Sufficient Reference Area
Because leaf structure and color along with canopy architecture are sometimes defining characteristics of hybrids, reflectance patterns can vary between hybrids even under adequately fertilized conditions. For these reasons, great care should be taken when comparing sensor data between fields that have different cropping histories, growth stages, and/or hybrids. Normalizing data to an adequately fertilized area within a field that has only received a little extra N should make it possible to make a reliable comparison between fields, hybrids, etc.
Sensor-based N applications are not generally made until the V6 growth stage or later. For this reason, a primary goal for field N management under high yield conditions should be to limit N stress from planting until this time. One means of achieving this goal is to apply 1/3 to 1/2 of the projected needed N at or prior to planting and use sensors to direct the balance of crop N requirement as in-season applications. Suggested pre-plant N rates include 40 to 60 lbs of N/acre for corn after soybeans and 60 to 80 lbs N/acre for corn after corn.
1. Sensor Check
Follow manufacturer’s instructions for installation and operation of sensors.
2. Fertilizer Applicator Check
A high-clearance vehicle configured with sensors that can straddle over the top of the corn crop is the preferred type of applicator (Figure 1). This vehicle, equipped with a computer that can process the sensor information and direct a variable-rate N controller serves as a one-pass sense and variable-rate N application system. If the system allows, test for appropriate delay settings (i.e., time of sensing, calculation, and fertilizer application response). Calibrate for the targeted application rates before going to the field for application. Ensure that the equipment is capable of delivering the range of N rates likely required and at the speed of field operations.
3. Scan N reference Strip with Sensors
At the time of side-dress N fertilization, sensor readings are first taken from the N- sufficient reference strip(s). Convert sensor reading to the appropriate vegetation index as per manufacturer recommendations. Reference values are stored in the computer for on-the-go calculations. The applicator then drives over the rest of the field sensing, calculating, and applying N variably in one pass. Sensor values will change if corn leaves are wet (e.g., due to morning dew or light rain) versus when they are dry. Other factors such as temperature changes throughout a day may cause slight changes in sensor values. For this reason, it is strongly recommended that the N-sufficient reference value be refreshed after each 2 to 3 hours of N fertilizer application.
4. Calculations for N Fertilization
The current procedures used to convert sensor readings into N application rates vary for the different sensor systems (Raun et al., 2002, Schmidt et al., 2009, Kitchen et al., 2010, Solari et al., 2010, Holland and Schepers, 2010). However, all involve comparing sensor readings from N-sufficient reference plants to the remainder of the field receiving N application.
The approach outlined by Solari et al. (2010) and Holland and Schepers (2010) are highlighted here for the sake of illustration. They utilized the sufficiency index (SI) concept of Biggs et al. (2002) as the means for converting sensor readings into N application rates where: SI = sensor reading of target area divided by sensor value for N-sufficient reference plants. For example, the equation developed by Solari et al. (2010) illustrated in Figure 9 shows that the appropriate in-season N application rate for corn at a sensor-determined SI value of 0.8 would be around 120 lbs N/acre. This figure also shows that as sensor SI approaches a value of 1, the crop is under less and less N stress and requires smaller amounts of N application to relieve N stress.
It should be emphasized that while this illustration denotes the general concept of sensor-directed N applications, procedures for the different sensors do vary. Consequently, users are encouraged to refer to the manufacturers’ supporting documentation for specific instructions on use of the different sensors.
5. Field Results
Results from several on-farm studies show modest to significant economic benefits associated with using these sensor technologies to direct in-season N applications. Benefits resulted from reduced N fertilizer costs and/or increased yields in N deficient field areas.
Active canopy reflectance sensing technology has the potential to improve yield, nitrogen use efficiency, and profit in corn production systems. This is especially true in situations where excessive precipitation has increased the potential for N losses. Most traditional N management schemes are designed to provide an N rate recommendation for application prior to or shortly after planting. These estimates are static and are not designed to react to in-season variations in weather. Such variations can alter N mineralization from organic matter and crop residue, and change the level of N loss from the system.
Sensor technologies, on the other hand, offer the opportunity to optimize tradeoffs among yield, profit, and environmental protection. This is because of their potential to achieve synchrony between N supply and crop demand, while accounting for spatial and year-to-year variability in soil N.
Nevertheless, it should be recognized there are areas where the sensor-based approach may not be appropriate. One example is rain-fed areas where precipitation from the proposed time of in-season N applications to the end of the growing season is low and/or erratic. Under these circumstances, in-season N applications made to the soil surface may be unavailable for crop uptake and thus likely to limit yields. In such cases, current pre-plant N applications would likely be more suitable.
|Unique Features of Sensor Systems
Potential users of sensor technologies are encouraged to explore the unique features of these three in-season sensor systems that include options to:
1. Substitute a virtual reference strip for a high-N reference strip
2. Auto-calibrate sensors while driving to simplify operator involvement
3. Base N recommendation on potential yield
4. Cut back on the recommended N rate in areas where complete yield remediation is unlikely
5. Choose between several visible wavebands to optimize sensitivity depending on growth stage
6. Consider different management zones within a field
7.Use imagery or maps as a driver for changing N rates.
Biggs, G.L., T.M. Blackmer, T.H. Demetriades-shah, K.H. Holland, J.S. Schepers, and J.H. Wurm. 2002. Method and apparatus for real-time determination and application of nitrogen fertilizer using rapid, non-destructive crop canopy measurements. United States Patent No. 6393927.
Holland, K. H. and J. S. Schepers. 2010. Derivation of a variable rate nitrogen application model for in-season fertilization of corn. Agron. J. 102:1415-1424.
Hong, N., P.C. Scharf, J.G. Davis, N.R. Kitchen, and K.A. Sudduth. 2007. Economically optimal nitrogen rate reduces soil residual nitrate. J. Environ. Qual. 36, 354-362.
Kitchen, N.R., K.A. Sudduth, S.T. Drummond, P.C. Scharf, H. Palm, D.F. Roberts, and E.D. Vories. 2010. Groundbased canopy reflectance sensing for variable-rate nitrogen corn fertilization. Agron. J. 102:71-84.
Mathesius, J. and G. Luce, 2009. Assessing and Managing Nitrogen Losses in Corn. Crop Insights, Vol. 19, No. 8. Pioneer Hi-Bred, Johnston, IA.
Raun, W.R. and G.V. Johnson. 1999. Improving nitrogen use efficiency for cereal production. Agron. J. 91: 357- 351.
Raun, W.R., J.B. Solie, G.V. Johnson, M.L. Stone, R.W. Mullen, K.W. Freeman, W.E. Thomason, and E.V. Lukina. 2002. Improving nitrogen use efficiency in cereal grain production with optical sensing and variable rate application. Agron. J. 94:815-820.
Roberts, D. F., V.I. Adamchuk, J. F. Shanahan, R.B. Ferguson, and J. S. Schepers. 2009. Optimization of crop canopy sensor placement for measuring nitrogen status in corn. Agron. J. 101:140-149.
Shanahan, J.F., N.R. Kitchen, W.R. Raun, and J.S. Schepers. 2008. Responsive in-season nitrogen management for cereals. Computer & Electronics in Agric. 61:51-62.
Solari, F., J.F. Shanahan, R.B. Ferguson, and V. I. Adamchuk. 2010. An active sensor algorithm for corn N applications based on a chlorophyll meter sufficiency index framework. Agron. J. 102:1090-1098.
Schmidt, J.P., A.E. Dellinger, and D.B. Beegle. 2009. Nitrogen recommendations for corn: An on-the-go sensor compared with current recommendation methods. Agron. J. 101:916-924.
NOTE: Links with this symbol will take you outside of pioneer.com. Pioneer.com does not own or control the content on sites other than its own.