Estimation of the in-field variation in fertiliser application

Variation in field application from a centrifugal fertiliser spreader has, in the past, been difficult to assess due to the intensive field testing required. However, with the application of current technologies such as Global Positioning Systems and Geographic Information Systems, the assessment of field application variation is now possible. The aim of this research was to develop a method to assess field application variation using basic transverse spread pattern test and vehicle tracking data. The information was used to measure and compare the effect of spread pattern, driving accuracy, driving method and paddock shape. Two differently shaped paddocks were used for analysis, one rectangular, the other triangular. A simple analysis method was initially used to calculate which areas of the paddock received nil, single, or double application. An advanced assessment method was also developed to measure application variation within the paddock and to calculate field application variation. Transverse spread pattern data was averaged at 2 m increments and linked to GPS tracking data from the spreading vehicle using ArcGI S 9.1(C). The dataset was then analysed and a field application map produced. The field coefficient of variation (CV) was then calculated from the application map. Three simulations of perfect driving width and perfect application pattern were also conducted as a means of comparison to the actual field data. Simple analysis results showed, in both paddocks, the area receiving the correct application rate was between 70 and 80%, with 5-15% of the area receiving nil application. Using the advanced analysis method, the calculated field CV was 32.9 and 43.0% for the triangular and rectangular paddocks, respectively. However, when simulated "perfect driving" tracking data were used, the average field CV was reduced to 24.8 and 23.5%, respectively. When both perfect driving and a perfect spread pattern were used field application variation could be reduced to less than 20% for either shaped paddock. The results concluded that there is far greater variation in field application of fertiliser than that measured from a single transverse test, with the single biggest gain in calculated field CV being from driver accuracy and driving method.