Assessment of Fertilizer Application Accuracy With The Use of Navigation Aids

 T.F. Burks, S.A. Shearer, and J.P. Fulton

Burks, T.F., S.A. Shearer, and J.P. Fulton.  2000.  Assessment of Fertilizer Application Accuracy With the Use of Navigation Aids. ASAE Paper No. 001154.  Annual International Meeting, Midwest Express Center, Milwaukee, Wisconsin, July 9-12.

INTRODUCTION

            The use of navigation aids for guiding sprayers and fertilizer spreaders is gaining in popularity, with the expectation that application uniformity is far superior to application based on dead reckoning.  Popular devices include foam markers and parallel tracking devices that use GPS.  Foam markers utilize an air pump to pressurize a tank containing the foaming agent, thereby causing the agent to flow to the accumulating chamber.  The foam collects in this chamber until the accumulated mass overcomes surface tension, causing a foam blob to fall to the ground.  Most often the foam accumulators are placed at the ends of the applicator boom, or alternately at the center of the applicator when booms were not utilized, as is the case with spinner disk granular applicators.  Equipment operators use the foam blobs left on the field surface as a navigation aid to know where the applicator has been.  GPS-based lightbars allow the operator to select an AB line by identifying start and end points.  Software allows the user to specify the sensitivity and distance between parallel passes.  A sequence of LEDs light up to guide the driver as steering corrections are warranted.  The expectation is that either foam markers or lightbars will improve the quality or application by improving the quality of spacing between parallel passes.  The purpose of the work detailed in this manuscript is to assess the perceived improvement in application accuracy.

 

BACKGROUND

Application accuracy is an important property to assess variable-rate technology (VRT) spinner spreader fertilizer application system. The coefficient of variation (CV) is typically used to characterize the quality of spread distribution by a spinner spreader. The lower the CV the more uniform the distribution pattern.  Typically, the CV varies from 5% to 10% for the transverse spread pattern of a spinner spreader, but this variation probably doubles (Parish, 1991) under field conditions such as terrain irregularities.  However, it expected that the CV could increase to 15% to 20% when conducting field tests (Sogaard and Kierkegaard, 1994).

 

The ASAE Standard (Procedure for Measuring Distribution Uniformity and Calibrating Granular Broadcast Spreaders. in ASAE, 1997) describes a uniform method of determining performance data on broadcast spreaders for granular materials, and provides a basic test procedure to compare spreader distribution patterns. Details, such as test setup, collection devices, test procedure, distribution, effective swath width, determination of application rates, are given clear definitions in the test procedure section in the standard.

 

Olieslagers et al. (1996) described the fertilizer distribution of a spinning disc spreader. Many parameters including orifice position and angular speed of disc impact the distribution pattern of disc spreader. VRT application, accomplished by changing the mass deposition rate on spinner discs, leads to a fluctuating spreader pattern that results in large deviation from the intended application rates. Olieslagers et al. (1997) suggested that continuous change to various spreader adjustments might be needed to maintain a uniform distribution pattern when changing rates on-the-go. They also stated that future work should be concentrated on the dynamic response of the fertilizer spread pattern when changing material rates on-the-go.

            Chaplin et al. (1995) investigated the distribution of dry material during field application. They described a methodology based on ASAE (1997), and did testing for a single-disk mounted fertilizer spreader.  Pettersen et al. (1991) investigated how the distribution pattern of a twin-disc spreader was influenced by fertilizer particle size. They provided a detailed test method for collecting fertilizer samples and used interpolation techniques and computer graphics to get continuous distribution patterns.

 Fulton et al. (1999) conducted an investigation to quantify the distribution of a spinner spreader equipped with VRT.  As suspected the distribution pattern changed substantially as increasing rate of granular materials were dropped onto the spinner discs.  Further, they were able to assess the rate of change in distribution when switching from low to high application rates.  Fulton et al. (2000) went on to describe application uniformity by simulating the “as applied” surface based on the knowledge gained in the Fulton et al. (1999) investigation.  Missing from either investigation was an analysis of the variations in pattern overlap.  Figure 1, adapted from Fulton et al. (2000), shows in part the effect of errors in pattern overlap, and the result non-uniformity for VRT application.  Non-uniform spacing between spreader passes contributes to the deviation of the application surface from the desired surface.  

 

OBJECTIVES

             The specific objectives of this investigation were: 1) to design and conduct an investigation to assess the quality of applicator parallel tracking with, and without, navigation aides, and 2) to approximate any improvement in application uniformity.

METHODS

             A 5.0 ha grass field was selected as the test course for this investigation.  The vehicle to be used for tracking was a traditional spinner spreader truck equipped with VRT capabilities.  Specific hardware and software selections for this study included a SATLOC LiteStar lightbar with the 2.08 version software for ground applications along with the SATLOC SLXg L-band receiver, and Falcon II (AgChem Equipment Co.) task computer running AgView (GSI Solutions, Inc.) software with an Ag 132 DGPS receiver with Coast Guard Radiobeacon correction (Trimble Navigation, Inc.).  The latter compliment was used for logging the tracking data for later analyses.  The applicator truck was also equipped with a ChemFarm foam marking system.

             The truck driver was instructed to navigate the 5.0 ha test field in the same manner he would normally make parallel passes to apply fertilizer.   Although feedback of the history of the traverse of the field was available at the Falcon II monitor, he was instructed to dim the monitor so that this information was not available.  The driver was instructed to drive the field at typical ground speeds (25 to 30 km/h), and at differing orientations between tests.  The tests to be conducted included: 1) no navigation aids, 2) navigation with foam marker, and 3) navigation using the GPS lightbar.  All tests were replicated twice.  The desired distance, to maintain application uniformity between adjacent passes, was 16.1 m.

             Tracking data were logged at 15 m intervals, the default option of the AgView software.  Essentially positions fixes were acquired at one-second intervals providing the minimum data-logging interval was achieved.  There was a sliding scale to allow more frequent logging at reduced ground speeds such as in turns.

 

RESULTS AND DISCUSSION

            Figures 2 show examples of the tracking data for individual test runs with, and without the use of navigation aids.  A cursory review of the data reveals that indeed navigation with the lightbar aid tended to be more collinear and parallel when compared the tracks with no navigation aids, or the foam marker track.  Unfortunately, there also appears to be greater differences in the mean difference between adjacent parallel passes.

             The data were compared using two analytical approaches.  Using the measurement tool within ArcView 3.2, and after projecting the tracks into State Plane coordinates (Kentucky North, NAD 83), 100 measurements (50 for either replicate) were collected for assessing the mean and variance of the distance between adjacent parallel passes.  Slight deviations (up to 3o) from the normal to the direction of travel for these measurements were deemed acceptable.  For each test case five to seven transects were made for collecting 50 measurements.

            Table 1 contains a summary of the data for assessing the distance between parallel passes.  The target distance between parallel passes was 16.1 m.  From this table it is somewhat obvious that perhaps the best navigation aid is the foam marker, as this produced a mean distance of 16.4 m, which is slightly greater than the 16.1 m target.  Navigation with the lightbar seemed to produce less desirable results, 18.5 m spacing versus the desired spacing of 16.1 m.  A statistical analysis using a Z-test for the comparison of means revealed no statistically significant differences between the navigation without aid, and navigation with foam marker means and the target.  The null hypothesis of the lightbar mean being equal to the target distance of 16.1 m was rejected at an alpha of 0.025.  The F-test was applied with an alpha of 0.10 to test for significant differences between means for the various navigation scenarios.  When comparing no navigation aids with use of a foam marker, a difference between means of 0.72 m was found to be significant.  Similarly when comparing the lightbar with no navigation aid, and then the lighbar with foam marking aid, there were significant differences of 0.06 and 1.44 m, respectively.

             A second, and perhaps equally interesting comparison was the variance between navigation approaches.  Using an F-test with the null hypothesis of the variances for each navigation method being the same, none of the variances were found to be significantly different at an alpha of 0.05.  A review of the information in Table 1 shows light bar to have a slightly reduced standard deviation when compared with the other two navigation approaches.

             To assess the co-linearity of the parallel passes 15 x-y coordinate pairs were extracted from one pass each for the no navigation aid test, and again for navigation with lightbar.  These two cases seemed t represent the extremes.  Figure 3 shows the graphs of either pass along with the linear regression statistics for fitting a straight line to either.  As one might expect the pass made using the lightbar exhibited better co-linearity as demonstrated by the r2 value of 1.0 versus the r2 of 0.997 for no navigation aid.  Arguably this difference is of minor importance.

             To estimate the effect of pattern overlap, pattern distributions from Fulton et al. (2000) were used to estimate the distribution of two overlapping parallel passes of a spinner spreader.  The low application rate distribution was selected for this example.  Figure 4 illustrates the patterns as described by Fulton et al. (2000).  Distribution curves created using average offsets for the foam maker and lightbar guidance approaches, as shown in Figure 5.  This represent the average extremes of the data collected in this investigation.  Most notable is the difference in application pattern in the 5 to 10 m overlap region, a 37.5% increase.  The role of proper overlap in fertilizer application is significant, and its effect must be considered when creating as applied surfaces.

 

SUMMARY

            With respect to variable-rate application, as with the example from Fulton et al. (2000) shown in Figure 1, overlap is a significant element of achieving the desired application surface.  From this investigation of guidance aids it is clear that the use of these devices produces mixed results.  While the lightbar guidance system seem to have less variability when compared with the use of foam markers, or no guidance aids, the mean distance between adjacent passes is much greater than expected.  This large difference may be attributed to errors in the set-up and operation of the lightbar system, although the errors are consistently graeter than expected.  The lightbar indicator sensitivity was set to 0.61 m (2.0 ft) per LED, as recommended in the owner’s manual.  It must also be noted that the truck was navigated using one DGPS receiver, while tracking was accomplished with a second DGPS receiver, possibly accounting for some of the difference.  It should also be noted that relative position errors are much less than the static horizontal accuracies reported by equipment manufacturers.  The results of such tracking errors have a significant effect on distribution patterns, which compound overall application errors for both fixed-rate and VRT applicatoions.  Perhaps the most significant result of this manuscript is the recognition of the role overlap plays in determining the “as applied” resultant, and that technologies such as lightbars do not necessarily eliminate these application errors.    

 

REFERENCES

 ASAE Standards, 44th Ed. 1997. 341.2.  Procedure for measuring distribution uniformity and calibrating broadcast
            spreaders.  ASAE, St. Joesph, MI.

 Chaplin, J. and E. Roytburg. 1995. Measuring the spatial performance of chemical applicators.  Site-Specific Management for Agricultural Systems.  pp.651-669.

 

Fulton, J.P., S.A Shearer, G. Chabra, and S.F Higgins.  1999.  Field evaluation of a spinner disc variable-rate fertilizer applicator.  Presented at the 1999 Annual International Meeting of the ASAE in Toronto, Canada.  ASAE Paper No. 991101.

 

 Fulton, J.P., S.A Shearer, T.F Burks, and S.F Higgins.  2000.  Simulated application errors for granular materials for fixed and variable-rate application.  Presented at the 2000 Annual International Meeting of the ASAE in Milwaukee, WI.  ASAE Paper No. 001153.

 

Linsley, C.M. and F.C Bauer.  1929.  Test your soil for acidity.  Circular 346. Univ. of Ill. Agric. Experiment Station, Urbana, IL.

 

Olieslagers, R., H. Ramon, and J. De Baerdemaeker. 1996.  Calculation of fertilizer distribution patterns from a spinning disc spreader by means of a simulation model.  J. Agr. Engr. Res., 63(2): 137-152.

 

Olieslagers, R., H. Ramon, and J. De Baerdemaeker. 1997. Performance of a continuously controlled spinning disc spreader for precision application of fertilizer.   Precision Agriculture 1997.   BIOS Publishers Ltd. pp. 661-668.

 

Parish, R.L. 1991. Effect of rough operating surface on rotary spreader distribution pattern.  Applied Engineering in Agriculture ASAE, 7(1):61-63.

 

Sogaard, H.T. and P. Kierkegaard.  1994.  Yield reduction resulting from uneven fertilizer distribution.  Trans. of ASAE. 37(6): 1749-1752.

 Whitney, R.W., D.L. Dludlu, L. J. Swenson, D.G. Hopkins, and W.C. Dahnke.  1993.  Variable fertilizer application based on yield goal, fertility, and soil map unit. J. Prod. Agr.  6(2): 255-261.