INTRODUCTION - Oklahoma State University–Stillwater

  • Doc File 2,681.00KByte



Determination of optimum resolution for predicting corn grain yield using sensor measurements

BYUNGKYUN CHUNGa, KEFYALEW GIRMAa, KENT L. MARTINb, BRENDA S. TUBAÑAc, DARYL B. ARNALLa, & WILLIAM R. RAUNa*[1]

aDepartment of Plant and Soil Sciences, Oklahoma State University, Stillwater, OK 74078; bDepartment of Agronomy, Kansas State University, Manhattan, KS66506 ; cSchool of Plant, Environmental and Soil Sciences, Louisiana State University, Baton Rouge, LA,70803

.

Contribution from the Oklahoma Agricultural Experiment Station.

Running Title: Optimum Resolution for Predicting Corn Grain Yield

Determination of optimum resolution for predicting corn grain yield using sensor measurements

Abstract

Identifying the optimum resolution where differences in corn (Zea mays L.) grain yields are detectable could theoretically improve nitrogen (N) management, thereby resulting in economic and environmental benefits for producers and the public at large. The objective of this study was to determine the optimum resolution for prediction of corn grain yield using indirect sensor measurements. Corn rows, 15-30 m long, were randomly selected at three locations where the exact location of each plant was determined. In 2005 and 2006, four of eight rows at each location were fertilized with 150 kg N ha-1 as urea ammonium nitrate (28% N). A GreenSeeker™ optical sensor was used to determine average Normalized Difference Vegetation Index (NDVI) across a range of plants and over fixed distances (20, 40, 45.7, 60, 80, 91.4, 100, 120, 140, 160, 180, 200, 220, and 240 cm). Individual corn plants were harvested and grain yield was determined. Correlation of corn grain yield versus NDVI was evaluated over both increasing distances and increasing number of corn plants. Then, the squared correlation coefficients (rcc2) from each plot (used as data) were fitted to a linear plateau model for each resolution treatment (fixed distance and number of corn plants). The linear-plateau model coefficient of determination (rlp2) was maximized when averaged over every 4-plants in 2004 and 2006, and over 11-plant in 2005. Likewise, rlp2 was maximized at a fixed distance of 95, 141, and 87 cm in 2004, 2005, and 2006, respectively. Averaged over sites and years, results from this study suggest that in order to treat spatial variability at the correct scale, the linear fixed distances should likely be 15.3*** |0.96 |

| |Efaw |V10 |9 |Y= 0.0799+0.0395x, when X < 9; Y=0.44, when X > 9*** |0.83 |

| |Lake Carl Blackwell |V10 |15 |Y= 0.0055+0.0135x, when X > 14.7; Y=0.2, when X< 14.7*** |0.95 |

|2006 |Efaw |V6 |2 |Y= 0.1921 + 0.069x, when X > 2; Y=0.33, when X < 2 |0.22 |

| | | | | | |

| |Lake Carl Blackwell |V6 |6 |Y= -0.00787 + 0.0099x, when X> 6; Y=0.05, when X < 6* |0.47 |

| |Efaw |V8 |2 |Y= 0.1975+0.0989x, when X < 2; Y= 0.4, when X < 2 * |0.40 |

| |Lake Carl Blackwell |V8 | |╪ | |

| |Efaw |V10 |2 |Y= 0.0943+0.1498x, when X> 2.5; Y= 0.47, when X < 2.5*** |0.71 |

| |Lake Carl Blackwell |V10 |6 |Y= 0.0205+0.0169x, when X> 6.2; Y=0.13, when X < 6.2** |0.65 |

*, ** and *** model significant at the 0.05, 0.01, and 0.001 levels of probability, respectively.

╪ Critical plant number exceeded data boundary.

¶ Critical point (number of corn plants) determined by defining the linear and plateau phases of the model separately.

Table 4. Linear-plateau model with critical fixed distance resolution (joint) and coefficient of determination (rlp2) derived from the correlation between NDVI and corn grain yield versus fixed distance at three growth stages at Efaw, Hennessey and Lake Carl Blackwell, 2004-2006.

|Year |Location |Growth Stage |Joint ¶ |Model |rlp2 |

|2004 |Efaw |V8 |93 |Y= -0.063 + 0.0043x, when X < 93.4; Y=0.33, when X > 93.4*** |0.95 |

|2004 |Lake Carl Blackwell |V8 |92 |Y= 0.0729 + 0.0026x, when X 92*** |0.66 |

|2004 |Hennessy |V8 |100 |Y= -0.0698 + 0.0033x, when X < 100; Y=0.26, when X > 100** |0.61 |

|2005 |Efaw |V6 | |╪ | |

|2005 |Lake Carl Blackwell |V6 |140 |Y= -0.0077 + 0.0007x, when X < 140;Y=0.09, when X > 140** |0.63 |

|2005 |Efaw |V8 |171 |Y= -0.0146 +0.0024x, when X < 171.1; Y=0.39, when X > 171.1*** |0.90 |

|2005 |Lake Carl Blackwell |V8 |133 |Y= -0.0093 +0.0007x, when X < 132.9; Y=0.09, when X > 132.9*** |0.70 |

|2005 |Efaw |V10 |156 |Y= 0.0212+0.002x, when X < 156.4; Y=0.34, when X > 156.4*** |0.95 |

|2005 |Lake Carl Blackwell |V10 |106 |Y= -0.0094+0.0011x, when X > 106.2; Y=0.11, when X< 106.2*** |0.72 |

|2006 |Efaw |V6 |92 |Y= 0.1573+0.0027x, when X > 92.5; Y=0.41, when X < 92.4*** |0.75 |

|2006 |Lake Carl Blackwell |V6 |100 |Y= 0.0014+0.00001x, when X> 100; Y=0.01, when X < 100 |0.15 |

|2006 |Efaw |V8 |66 |Y= 0.1158+0.0049x, when X < 66.1; Y=0.44, when X < 66.1*** |0.94 |

|2006 |Lake Carl Blackwell |V8 | |╪ | |

|2006 |Efaw |V10 |66 |Y= 0.0677+0.006x, when X> 65.9; Y=0.47, when X < 65.9*** |0.87 |

|2006 |Lake Carl Blackwell |V10 |111 |Y= 0.0887+0.0019x, when X> 110.6; Y=0.29, when X < 110.6*** |0.71 |

*, ** and *** model significant at the 0.05, 0.01, and 0.001 levels of probability, respectively.

╪ Critical plant number exceeded data boundary.

¶ Critical point (distance in cm) determined by defining the linear and plateau phases of the model separately.

[pic]

Figure 1. Framework for the bicycle and the adjustable pole that holds the sensor parallel and directly above the corn row. The shaft encoder is used to determine the distance at which NDVI is recorded is shown on the rear tire of the bicycle.

-----------------------

* William R. Raun, Oklahoma State University, Plant and Soil Sciences, 044 Agricultural Hall, Stillwater, OK, USA. E-mail: bill.raun@okstate.edu

................
................

Online Preview   Download