| [Image: graph of granular losses] |
| [Image: graph of granular losses] |
| [Image: graph of granular losses] |
Nontarget losses of granular material applied over widely spaced containers were reduced from 87% with a broadcast rotary applicator to 72%-86% with a drop-type spreader to 48%-75% with a drop spreader modified to band apply the material.
Plant species and container spacing configuration had a significant effect on material loss. With a drop spreader, losses ranged from a low of 10% with closely spaced juniper to 86% with widely spaced liriope. With a rim-to-rim hexagonal configuration, the losses varied from 10.2% with juniper to 19.9% with liriope. With a rim-to-rim square configuration, the losses varied from 15.1% with azalea to 31% with liriope. There were no significant differences in loss with the wide-spaced configuration.
Introduction
Application of herbicides formulated on granular carriers is a common practice in container nurseries. Depending on the container arrangement and plant species, a high percentage of the material applied may not be retained in the containers - particularly when the material is applied with a broadcast rotary spreader, as is a common practice. The material not retained in the containers represents a significant unproductive cost and can contribute to surface or groundwater pollution. This work demonstrated the increased efficiency of application possible with a drop-type spreader. The differences in application efficiency among different plant species are also noted.
Review of Literature
Container production of ornamental plants is a major industry in the United States. Controlling weeds in the containers requires herbicides. In many cases, the herbicides are applied in a granular formulation.
In Alabama, growers make an average of three applications per year (2). Rotary broadcast spreaders are frequently used to apply the granular herbicides, and losses from such applications can be significant. Gilliam et al. (1) documented losses ranging from 23% - 30% with closely spaced containers to 79% - 80% with containers spaced 30 cm on center. Their tests were conducted by sprinkling a measured amount of granules directly over the plants, so they did not measure the additional losses occurring because of the tapering pattern and end-of-bed effects common with typical rotary broadcast spreaders.
Several approaches to this problem have been suggested. Verma (5) developed slow-release herbicide tablets that could be placed individually in the containers, thus eliminating application losses. Parish et al. (4) developed a precision metering system that could be mounted on a potting machine to apply a discrete charge of granular material to the container of potting medium. The system placed the granular material in the medium, not on the surface as is needed for many herbicides. Parish et al. (3) built an applicator that straddled beds of containers and applied discrete charges of granular material to individual pots in multiples across the bed of containers. The applicator was originally designed to make a dibble hole for the plants and then place the granules in the hole, but the machine could easily be modified to apply granules to the surface of planted containers. Labor efficiency with that machine was far lower than with broadcast application.
The objectives of the current study were: (1) to quantify the granular material losses from rotary broadcast application to several plant species, taking pattern feathering into account, (2) compare those losses with the losses in three container configurations with a drop-type broadcast applicator and (3) then evaluate the improvement in efficiency possible with a drop-type spreader modified to apply a band of granular material.
Species Used in this Study
liriope (Liriope muscari Decne.), prostrate juniper (Juniperus prostrata L.), monkey grass [Ophiopogon japonicus (Thunb.) Ker-Gawl.], azalea (Rhododendron X 'Carror' ), dwarf gardenia (Gardenia radicans Thunb.)
Materials and Methods
Attapulgite clay granules (Florex LVM 20/40 from Floridin, Quincy, Fla.), a base material for granular pesticide formulations, were used for this experiment. The granules had a nominal sieve size of -20+40. Blank granules were used to avoid any operator toxicity or environmental problems that might have arisen from the use of active herbicide granules in the test program.
Five common species of ornamental plants were used for this study. Plant materials used were one growing-season-old liriope, prostrate juniper, monkey grass, azalea and dwarf gardenia, and two growing-seasons-old azalea. These plant species represent a wide range of plant canopy openness. All of the plants were in #1 nursery containers that were 16.5 cm (6.5 in.) in outside diameter and 16.2 cm (6.375 in.) high.
Three container configurations were evaluated:
Configuration 1. Containers were placed rim to rim in a hexagonal pattern with a straight line of containers in the direction of spreader travel.
Configuration 2. Containers were placed rim to rim in a square pattern.
Configuration 3. Containers were placed in a square pattern on 30.5 cm (12 in.) centers in each direction.
The first part of the study determined the amount of material lost when a typical broadcast application was made using a hand-cranked rotary applicator (Earthway Model 2700A). Two plant species, azalea and monkey grass, were used, with only the "extreme" configurations being evaluated - configuration 1 for monkey grass and configuration 3 for azalea. The containers were arranged in beds approximately 1.8 m (6 ft) wide. Two strips of polyethylene film 1.2 m (4 ft.) wide were placed under the containers, perpendicular to the bed and to the direction of spreader travel. A third strip of polyethylene film was laid down parallel to and between the first two strips, but without any plants or containers. The polyethylene film strips were long enough to catch the full width of the spreader patterns. The plant containers overlapped the polyethylene film strips in the direction of spreader travel. The spreader was operated down each side of the bed of plant containers with the bed to the left of the operator in each case and was angled in such a way that the spreader threw primarily to the left (the normal mode of operation for a spreader of this type in a container nursery application). Operating in this manner normally results in two complementary feathered patterns that approximate a uniform pattern when added together. Preliminary runs were made without plants to verify that the spreader patterns did indeed overlap to provide a reasonably uniform pattern in the bed area(s).
After operating the spreader over the containers and polyethylene film strips, the containers were removed and the material on each polyethylene film strip was collected and weighed. To arrive at the percentage of material lost, we divided the weight of the material on the strips under the plants by the weight of the material caught on the strip without plants. Operating the spreader over all strips on each test run eliminated the need for separate calibration. The polyethylene film strip without plants served as a calibration for each test run. The test was replicated four times.
For the second objective of the study, a special laboratory fixture was built to facilitate the evaluation of the drop-type spreader. An elevated wooden track approximately 0.6 m (2 ft.) high and 4.9 m (16 ft.) long ran over a wire mesh "table" on which the plants were placed. A sheet of polyethylene film 0.9 m (3 ft.) long and wider than the bed of plants was placed under the wire mesh to collect all granular material not retained in the containers or on the plant leaves. Plants were placed on the wire mesh in each of the three configurations. This was repeated for each species of plant. The number of plants used was adequate to extend beyond the test application and collection area in all directions.
A spreader forward speed of about 3.2 km/h (2 mile/h) was used. Since it was difficult to control spreader forward speed between runs, a set of four empty containers was placed under the test track just ahead of the plants. The material falling into those containers divided by the area of those containers represented the total application rate made by the spreader for each of the test runs. Thus, each test run had its own independent calibration. The amount of material falling onto the polyethylene film sheet under the plants could then be compared directly with the amount falling into the calibration containers to determine the percent of material not retained by the plants and containers for each individual test run. Each test was replicated four times. The delivery rate for this series of runs averaged 347 kg/ha (310 lb/A).
A Gandy model 24H12 spreader was used. The ports on the spreader were 2.86 cm (1.125 in.) apart. This resulted in discrete bands of granules being dropped. This problem was corrected by adding a steel deflector under the ports at a 45E angle to vertical. The deflector caused the material to spread and blend into a uniform band across the full 60.9 cm (24 in.) width of the spreader.
After the basic runs were completed with empty containers and all species of plants at each of the three container configurations, the spreader was modified to band the material over the containers by closing off the unneeded delivery ports on the spreader with masking tape. Since the greatest loss of granular material occurred with widely spaced containers, only configuration 3 was used for this series of test runs. This test was replicated four times. The delivery rate for this series of tests averaged 116 kg/ha (104 lb/A).
Results and Discussion
Table 1 compares the material losses from the rotary broadcast spreader and the drop-type broadcast spreader for two species of plants. With the monkey grass in a close hexagonal configuration (configuration 1), significantly more granular material was lost from the rotary broadcast application. A rotary spreader does not restrict the distribution pattern to a discrete band, but allows the pattern to taper off gradually on each side of the pattern. To achieve a full application rate on the entire bed of containers, some material is necessarily thrown outside the bed. As indicated in Table 1, this loss increased the total loss significantly with closely spaced plants where the loss between plants was relatively small. With the azaleas on a wider spacing (configuration 3), the loss again was higher with the rotary spreader, but the difference was not statistically significant.
Table 1. Comparison of granular material losses between a rotary-type broadcast spreader and a drop spreader. The figures shown are percentages of material not retained in containers.
|
|
Rotary Spreader |
Rotary Spreader |
Drop Spreader |
Drop Spreader |
| Species |
Configuration |
Mean % |
CV, % |
Mean % |
CV, % |
| Monkey grass |
1 |
37.2 |
18.5 |
12.9 |
6.3 |
| Azalea |
2 |
86.8 |
14.5 |
75.5 |
12.5 |
Table 2 compares the granular material losses from five plant species and empty containers arranged in the three container configurations. The table also shows the theoretical losses for each container configuration. The actual losses were similar to the theoretical losses, but the plant material did have an effect on the loss of granules. Granular material lost when applied to juniper was the lowest because the juniper canopy was low in stature and did not cover the surface of its container. Therefore, more of the granules reached the medium surface and were retained in the container. Liriope exhibited increased material losses, indicating that the smooth leaves that covered the container surface deflected granules away from the containers.
Table 2. Comparison of granular material losses with five different plant species and empty containers using three container configurations. The figures shown are percentage of material not retained in containers.
|
Conf. |
#1 |
Conf. |
#2 |
Conf. |
#3 |
| Species |
Mean % |
CV, % |
Mean % |
CV, % |
Mean % |
CV, % |
| Theoretical |
9.3a |
---- |
21.5b |
---- |
78.3a |
---- |
| Empty |
12.3b |
3.6 |
27.7c |
7.2 |
76.2a |
5.5 |
| Liriope |
19.9c |
7.7 |
31.0c |
13.9 |
86.0a |
8.4 |
| Juniper |
10.2a |
11.7 |
20.1b |
16.6 |
72.1a |
9.0 |
| Monkey Grass |
12.9b |
6.3 |
30.5c |
6.8 |
81.4a |
12.3 |
| Azalea |
10.5a |
7.8 |
15.1a |
10.7 |
75.5a |
12.5 |
| Gardenia |
13.4b |
12.9 |
21.4b |
23.2 |
80.2a |
2.9 |
Table 3. Comparison of granular material losses between a full-width drop spreader and a drop spreader modified to band the material. Five different plant species and empty containers were used. All plants are spaced in each direction on 30.5 cm (12 in.) centers. The figures shown are percentage of material not retained in containers.
An increase in material loss was noted among configurations 1, 2 and 3. As the containers were spaced farther apart, the losses increased since the containers covered a smaller percentage of the application area. Application efficiency was better with closely spaced plants.
The improvement in application efficiency possible with a more controlled application pattern is shown in Table 3. For all plant species except juniper and azalea, a significant reduction in granular material loss was noted with the banding spreader compared with the full-width drop spreader. Material was still lost between the containers in a line parallel to the direction of travel, but losses between containers perpendicular to the line of travel were greatly reduced in most cases.
This study first demonstrated the extremely low application efficiency of rotary broadcast application on widely spaced plant containers. The losses in that case were as high as 87% of the granular material applied. Using a close container spacing reduced the losses with a rotary spreader to 37%. Using a drop-type spreader reduced the losses somewhat with widely spaced containers to 72%-86% and reduced the losses with close container spacing to 10%-20%. Modifying a drop spreader to apply discrete bands further reduced losses with the widely spaced containers to 48%-75%.
Efficacy of granular material application under typical nursery conditions is very poor. This study demonstrates that application efficiency can be significantly improved by the use of different, but commercially available, application equipment. The authors suggest that nurseries consider using drop-type spreaders rather than rotary spreaders for granular application. Although a small, low clearance spreader was used for this test, large high-clearance spreaders with similar operating characteristics are commercially available in widths up to 3.7 m (12 ft.). Using such a spreader in a broadcast mode for closely spaced containers and modifying it to band on widely spaced containers would significantly improve the efficiency of granular herbicide application. This improvement in application efficiency would reduce costs and environmental contamination from lost herbicide.
Gilliam, C. H., D. C. Fare, and A. Beasley. 1992. Nontarget herbicide losses from application of granular Ronstar to container nurseries. J. Environ. Hort. 10(3):175-176.
Gilliam, C.H., W.J. Foster, J.L. Adrain and R.L. Shumack. 1990. A survey of weed control costs and strategies in container production nurseries. J. Environ. Hort. 8:133-135.
Parish, R. L., P. P. Chaney, W. A. Meadows, and D. L. Fuller. 1988. A dibble fertilizer applicator for containers in nursery beds. J. Environ. Hort. 6(2):63-66.
Parish, R. L., R. J. Constantin, W. L. Brown, and D. W. Wells. 1986. Development of an automated fertilizer applicator for potting machines. J. Environ. Hort. 4(3):91-94.
Verma, B. P. 1978. Slow release herbicide tablets for container nursery. Trans. of the ASAE 21(6):1054-1059.