Introduction

This presentation will be restricted to a consideration of the application efficiency of boom sprayers. Other application techniques such as ropewicks, high volume or hand held sprayers or granule applicators will not be considered.

Herbicides have been used to control weeds in Australia since the late nineteenth century when common salt and kerosene were used to control certain weedy plants (Cuthbertson 1972). The boom sprayer was developed in the 1880s in France and the United States of America and was first used in Australia in the early 1900s (Combellack 1981a). These early horse-drawn units used hollow cone nozzles and utilized a ground drive positive displacement pump (metered sprayer). Later developments saw the pump motorised, the units drawn by tractors, larger spray tanks, larger booms and, in the l940s, the introduction of the fan nozzle, Though the use of herbicides has exploded since the 1940s, following the introduction of the phenoxy acetic acid herbicides (2,4-D and MCPA), boom sprayer developments have not been so dramatic. We have seen the introduction of fibreglass tanks, more efficient pumps, more durable and lighter spray lines, self-levelling booms and even some electronic components over recent years. Even so, one could only term such changes as fine tuning. Why have there been so few developments? What is the community's attitude to boom spraying? How efficient are the boom sprayers used by growers? Is it possible to improve the efficiency of the boom sprayers being used? These are some of the questions which will be considered.

Communities’ attitude to boom spraying

To maximise yields from crops in the Riverina it is invariably essential to use agricultural chemicals to control either weeds, fungi, insects or nematodes. The value of these chemicals is doubtless obvious to a user, however before answering any question on this topic one must consider the attitudes of all people toward agricultural chemicals. In general terms the public appear very concerned about the use of pesticides and people can become very emotive about the effects spray supposedly has on their health and well-being. Everyone is aware of the public’s attitude towards DDT and more recently 2,4,5-T. In view of these experiences, and let us remember that another agricultural chemical or group of chemicals could well come under similar scrutiny in the future, it is imperative that we apply all such chemicals in the most effective manner possible. We must, therefore, ensure that the minimum amount of pesticide is used to produce the desired level of control whilst ensuring that damage to the environment and, the community, both within and outside of the treated area, is minimised. Can we honestly say that this is being achieved where boom spraying is practised? In other words , how efficient are the boom sprayers being used?

The majority of growers would probably consider their boom sprayer is efficient because it gives acceptable results. Furthermore, because of this, the question is often asked as to why one should bother to consider changes to the unit. This is a real dilemma as research workers consider spraying to be a very inefficient operation (Himel 1974, Combellack 1981b). The reasons for this dilemma will be assessed in this presentation.

How efficient is the boom sprayer?

If boom spraying is to be efficient then distribution must be uniform across the swath when stationary. Tests conducted by Keith Turnbull Research Institute (KTRI) staff show there to be considerable variation across farmers’ booms when tested using a patternator with 75 mm channels (see Table 1). The results show that the lowest reading was 96% below the mean and the highest was 132% above. Such results obviously indicate lack of uniformity and show that the variability of all but a few booms was unacceptably high. Unacceptable booms are defined as having a coefficient of variation that is greater than 20% (Nordby 1978).

The above results relate to static booms. In tests carried out on moving booms, spray distribution has been found to be equally variable both across and along the swath (Maybank et al. 1974, Grover et al. 1978, Combellack 1982). In each of these studies spray deposition varied by at least +/- 25% of the intended dose rate. For example, the work done in Canada (Maybank et al. 1974) showed that deposits of 2 4-D varied from 20 - 60 mg/m² when the intended dose was 40 mg/m². A similar variation (36 - 192%) has been obtained at KTRI using (Bromoxynil + MCPA) and diclofop methyl (Combellack 1982). What does this mean? Obviously if our initial notion about the sprayer working is correct, and I have no reason to doubt that they do not work, then it is obvious that parts of the sprayed area are being overdosed and parts underdosed. Indeed, the Canadian workers (Maybank et al. 1978) have calculated that if one aims to apply a herbicide at 20 m~/m² over at least 95% of the sprayed area then at best 25 mg/m² and at worst 120 mg/m² must be applied. In other words, at least one-and-a-quarter to six times the desired dose rate must be applied. I hope it is obvious, therefore, that your sprayer only works because you are using higher dose rates than would be necessary if even distribution could be achieved.

What can be done to improve the evenness of distribution?

Basically six factors need to be considered:

1. Firstly, output needs to be related to ground speed. There are various methods of achieving this aim and the subject has been well reviewed by Amsden 1980 and Musillami 1979. In summary, each method has faults; the only common fault is the time lapse for the change in output to be related to the change in ground speed.

2. Secondly, the boom must be stable to minimise both yaw and whip. Various attempts have been made to reduce the effects of both of those parameters; the best attempt appears to be the Gimbal boom designed by Nation C1980). This design employs hydraulic and spring dampers to reduce vertical and horizontal movements and has significantly reduced such movements (Nation 1980).

3. Thirdly, one needs to use efficient atomisers. Measurements of flow rate, spray angle and distribution have been made at KTRI on the most commonly used nozzles. These tests, using new nozzles, showed that with respect to flow rate only 12.7% were within ± 2½% of that stated by the manufacturer, 11.3% within ± 2½ - 5.0% and 61.3% had a variation greater than ± 10%. Also, when nozzles were tested at a pressure of 200 kPa on a boom at a height and spacing recommended by the manufacturer, the results were disappointing, as can be seen from a summary of the results (Table 2). It can be seen that only 2.6% of the nozzles tested gave excellent distribution (coefficient of variation <10%) whilst 49.4% were found to be unusable (coefficient of variation >20%) (Nordby 1978). In summary, these tests show that the quality of nozzles being used is far from satisfactory.

4. Fourthly, droplet trajectory needs to be considered. When fan nozzles are pointed vertically at the target, droplet capture on vertical weeds is disadvantaged compared to that of flat weeds. Variations in trajectory, such as the spray sheet being projected horizontally (i.e. parallel to the ground) and away from the direction of travel, need to be considered. In this case droplets would be subject to a combination of sedimentary and horizontal movement (due to wind). This trajectory would probably improve droplet capture on vertical weeds.

5. Fifthly, droplet size needs to be considered as this will influence capture by the target. In theory, small droplets (100pm) are more efficiently distributed. Unfortunately such droplets are also more prone to drift which leads to a dilemma - better potential capture with greater drift hazard or vice versa. Until more research has been done on the optimum droplet size for sprays, small droplets ( 100pm) should be avoided to minimise the drift hazard.

6. Sixthly, the nozzles must be correctly set up on the boom. If used pointing vertically towards the target they should be offset 10-15⁰ to the boom. The height of the boom is critical as lowering the boom only a few centimetres below the optimum height will greatly increase distribution variation. Also, one must use the correct pressure; if the pressure is too low then even distribution will not be possible.

What of the future?

Minor improvements to boom sprayer design will continue. In particular, electronics will become more popular, for example the use of electric solenoids and flow meters. Better nozzles will become available, possibly injected moulded nylon. Anvil nozzles will probably become more popular to enable low volume (25-30 1/ha) spraying in multi-cropping areas. Boom design will improve both horizontal and vertical stability. Electrostatic sprayers will become available; one type designed at the National Institute of Agricultural Engineering (UK), which filters out the driftable droplets C 7Opm),will be of considerable value in areas where drift is a problem. Though the electrodyne system, which is being developed by Id, requires special oil based formulations, it will probably be useful in applying pre-emergent and minimum tillage type herbicides at very low volume rates (<51/ha). One other area of great interest will be the development of granular herbicides and applicators. The ultimate goal must be to include a suitable slow release granular herbicide with our seed or fertilizer at the time of planting. Such a herbicide could protect the crop from weeds until harvest,

TABLE 1. Summary of boom assessments

*1 SS = Spraying Systems

L = Lurmark

H = Hardi

*2 B = Brass
SS = Stainless steel
N = Nylon
Al = Scintered Alumina

*3 C.V. = Coefficient of variation. The lower the figure the more uniform the result.

TABLE 2. Distribution accuracy new nozzles

REFERENCES

1. AMSDEN, R.C. (1970). The metering and dispersing of granules and liquid concentrates. British Crop Protection Council Monograph No. 2, 124 - 129,

2. COMBELLACK, J,H. (1981a). History and trends in the design of ground boom sprayers for Australian conditions. Proceedings of the Sixth Australian Weeds Conference, 1, 87 - 91.

3. COMBELLACK, J,H, (1981b). The problems involved in improving spraying efficiency. Australian Weeds, 1, 2, 13 - 18.

4. COMBELLACK, J.H, (1982). Herbicide Application - A review of broadacre ground application techniques. Crop Protection (In press).

5. CUTHERBERTSON, E.G. (1972). The Early History of Chemical Weed Control in Australia. Proceedings of the Weed Science Society of NSW. 5, 31 - 40.

6. GROVER, R,, KERR, L.A., MAYBANK, J. AND YOSHIDA, K. (1978). Field measurement of droplet drift from ground sprayers. I. Sampling, analytical and data integration techniques. Canadian Journal of Plant Sciences, 58, 611 - 622.

7. HIMEL, C,M. (1974), Analytical methodology in ULV in Pesticide Application Methods. British Crop Protection Council Monograph No. 11, 112 - 119,

8. MAYBANK, J., YOSHIDA, K. AND GROVER, R. (1974). Droplet size spectra drift potential and ground deposition pattern of herbicide sprays. Canadian Journal of Plant Science, 54, 541 - 546.

9. MAYBANK, J., YOSHIDA, K. WALLACE, K. AND PETERS, M. (1978). Spray drift from agricultural pesticide applications. Air Pollution Control Association Journal, 28, 10, 1009 - 1014.

10. MUSILLAMI, 5. (.1979). Les systemes de regulation de debit des pulverisateurs a pression de liquide. Bulletin No. 267. Published by CNEEMA, Parc De Tourvoi, Paris, France.

11. NATION, H,J. (1980). The performance and stability of boom sprays. British Crop Protection Council Monograph, 24, 145 - 158.

12. NORDBY, A, (1978). Dyseposisjon pä spredebommer dysehøyde -arbeidstrykk - vaeskedordeling. N.J.F. Seminar Åkersproyter og åkersprøyting, Ås, ALH, Norge,