- Email: [email protected]
Buffer zone widths for honeybees from ground and aerial spraying of insecticides
Environmental Pollution 63 (1990) 247-259
Buffer Zone Widths for Honeybccs from Ground and Aerial Spraying of Insecticides B. N. K. D a v i s & C. T. W i l l i a m s * NERC, Institute of Terrestrial Ecology, Monks Wood Experimental Station, Abbots Ripton, Huntingdon PEI7 2LS, UK (Received 21 July 1989; revised version received 27 September 1989; accepted 28 September 1989) ABSTRACT The relative field hazards of insecticides to honeybees have been estimated by considering intrinsic toxicity levels andfield application rates. This approach is extended here to a consideration of buffer zones downwind of sprayed areas by estimating the distance at which bees would encounter an LDso dose from spray drift. 'LDso distances' are determined for both ground and aerial spraying of ground crops in Britain using published data on spray deposition under various weather conditions. For ground spraying at low wind speeds ( < 3 m s- 1), this zone of risk is up to 5 m for thegreat majority of compounds. Aerial spraying in unstable atmospheric conditions appears to produce drift deposits of about the same order of magnitude as from ground spraying at wind speeds of about 4 m s - l , with maximum LDso distances of <4Ore for chlorpyrifos, .fenitrothion and triazophos. For aerial spraying in stable atmospheric conditions these distances would be much greater. Pieris brassicae larvae are contrasted with honeybees in their relative sensitivities to insecticides and consequent LDso distances.
INTRODUCTION U n d e r the Control o f Pesticides Regulations (1986), substances are classified into four categories in relation to their toxicity to honeybees (as in the former Pesticide Safety Precaution Scheme), viz. 'extremely d a n g e r o u s ' n L D s o * Present address: Dodo Creek Research Station, Ministry of Agriculture and Lands, PO Box G l3, Honiara, Solomon Islands. 247 Environ. Pollut. 0269-7491/90/$03-50 © 1990 Elsevier Science Publishers Ltd, England. Printed in Great Britain
248
B. N. K. Davis, C. T. Williams
<0"1 ~tg bee- 1; ,dangerous, LDso 0.1-1.0; 'harmful'--LDso 1.0-10-0; and not classified--LDso >10tag bee -1 The actual danger from any compound, however, depends also on the extent of exposure in the field. As Smart and Stevenson (1982) said: 'This hazard is a function of intrinsic toxicity of the pesticide, the field rate (weight of active ingredient per hectare) at which it is applied, the proportion of the dose which is available for transfer to the bees at risk, and the behaviour of the bee itself.' This statement is true for any non-target organism, and the danger from pesticides is not necessarily confined to the crop itself since spray drift can affect non-target areas. In a few cases, buffer zones have been set to prevent harmful effects of insecticide spray drift on beneficial insects or wildlife. In Israel, for example, aerial spraying of cotton is not permitted within 200 m of citrus orchards because of the risk to natural predators whose elimination has caused pest outbreaks in the orchards (Harpaz, 1979). In Canada, where permethrin is used to control forest insect pests, buffer zones have been determined for 'worst case' scenarios, so as to minimise the risk to aquatic invertebrates and fish (Payne et al., 1988). A 20 m swath width was found to be adequate for limiting mortality to acceptable levels during ground-based applications. The use of unsprayed 'conservation headlands' 6 m wide in cereal fields in Britain is advocated by the Game Conservancy Trust to increase the insect fauna for the benefit of partridge chicks, and this measure also reduces pesticide drift into hedgerows (Cuthbertson & Jepson, 1988). The Nature Conservancy Council is concerned with the hazards of spray drift to wildlife generally, especially from aerial spraying. The Control of Pesticide Regulations (I 986) places an obligation on operators to consult the NCC if aerial spraying is to be undertaken within 0.75 nautical mile of a Site of Special Scientific Interest. However, more detailed knowledge is needed on buffer zones for both ground and aerial spraying of pesticides, and to this end, Sinha et al. (1990) have estimated ds0 values in bioassay studies of spray drift, i.e. the distance at which organisms will receive an LDso dose under given conditions. This paper compares the environmental hazards of 20 insecticides used in Britain for ground and/or aerial spraying of agricultural crops by estimating the respective 'LDso distance' values for honeybees from published data.
SPRAY D R I F T DEPOSITION The extent of spray drift depends on three sets of factors: (a) meteorological factors such as wind speed, turbulence and atmospheric stability, (b) the 'roughness length' of the surface over which the wind is flowing, and (c)
Buffer zone widths for hone),bees
249
operator-controlled variables, such as application method, type of nozzle, number of swaths and height of release (Elliott & Wilson, 1983; Joyce, 1985; Quantick, 1985; Gilbert & Bell, 1988). Briefly, atmospheric conditions are said to be unstable, neutral or stable depending on the temperature profile and vertical movement of air. Unstable atmospheric conditions occur when the ground and vegetation heats up on a sunny, clear day, producing a rising column of warm air (thermals), and sucking in cooler air near ground level. Stable conditions develop when the ground cools faster than the air above (by radiating heat to the sky) and then cools the adjacent layer ofair; they are typical of late evenings to early mornings with clear skies and little or no wind. Neutral conditions are when there is no tendency for air to rise or fall vertically. Pasquill (1961) has defined five categories, from A, very unstable, to F, very stable, which are generally used by agrometeorologists in Britain. However, this paper uses a "stability ratio" index in which values less than zero indicate unstable conditions, those between c. -0.1 and 4-0" 1 neutral stability, those between 0.1 and 1.2 moderately stable, and those above 1.2 very stable conditions (Yates et al., 1981). Stable conditions, with low wind speeds and reduced turbulence, increase pesticide drift hazards because a slowly diffusing plume of small drops can drift well off the target area in an unpredictable way. Roughness length is a parameter determined by the vegetation height and density which affects the frictional turbulence of the air flow. Spray particles are more likely to drift over short, smooth surfaces, such as short grass, than over tall, rough vegetation, such as a wheat crop which creates vertical eddies. For commonly used ground crop sprayers, fitted with hydraulic pressure nozzles and using recommended procedures, wind speed is considered the most important factor affecting drift (Nordby & Skuterud, 1974; Goering & Butler, 1975; Byass & Lake, 1977; Maybank et al., 1978; Lloyd & Bell, 1982; Arvidsson, 1985). Aerial spraying generally produces considerably more drift, with estimates of losses from the target area ranging from 20-40% (Maybank et al., 1978; Shires & Bennett, 1985) to > 50% (Ware et al., 1970; Joyce, 1985) of the total amount applied. It is most important to avoid spraying in stable atmospheric conditions. In general, wind speed is a less dominant influence for aerial sprayers than it is for ground sprayers, probably because of the overriding influence of the aircraft wake. Nevertheless, aerial spraying should only be carried out in wind speeds of <10 knots ( < 5 m s -1) (MAFF/HSE, 1989). Operator precision, in spray release height and in switching nozzles on and off at the beginning and end of spray runs, is especially important. Evaporation of water-based sprays is a particular
250
B. N. K. Davis, C. T, Wilfiams
problem because droplets take longer to reach the crop than for ground sprayers, and can produce aerosols consisting of almost neat pesticide.
Drift from ground spraying Several studies of drift from ground sprayers using conventional hydraulic systems give estimates of deposits downwind of multiple-swath applications, or provide enough single-swath data for such estimates to be calculated. The results for five studies are illustrated by Marrs et al. (1989) (from Williams et al., 1987). They show that concentrations of spray deposits at any particular distance downwind could vary by between one and two orders of magnitude; deposits were greater at higher wind speeds and on narrow cylindrical targets rather than on flat targets. Curves of maximum drift deposits on cylindrical targets are shown here in Fig. 2, lines A and B for wind speeds of 2.5-3 and 4 m s- 1, respectively. Figure 2, line B' shows drift deposits on fiat targets at the higher wind speed for comparison. Nordby and Skuterud (1974) obtained drift deposit values for a still lower wind speed of 1-5 m s-~, the curve (4a in Marrs et aL, 1989) having a similar but slightly lower trajectory to that shown in Fig. 2, line A. The experimental conditions for these three studies used to construct Fig. 2 are summarized in Table 1.
Drift from aerial spraying There is a considerable amount of data on drift from aerial spraying with standard hydraulic pressure nozzles but all measurements of deposits have been made with fiat, horizontal targets which have low collection efficiency for small droplets. Most of the studies have been undertaken in southern USA but examples are drawn also from Britain and Canada. Table 2 and Fig. 1 summarize the spraying conditions and illustrate the results of eleven trials. A subset of these data, representing the curves for m a x i m u m deposits under three sets of contrasting conditions of atmospheric stability (Fig. 2, lines C, D and E) are used in distance estimation.
Hazard indices and LD5o distances The field application rate of a pesticide can be compared with the topical LDso dose for a honeybee (in/zg per bee) to obtain a notional index of field hazard. If the target area of a bee in a spray cloud is taken to be 1 cm 2, and it is assumed to receive an instantaneous dose, a sprayer application rate of 100g a.i. h a - 1 would produce a deposit on a bee of 1/zg (since 100gha- 1 = 1/ag c m - 2). One can thus calculate the factor, or hazard index HI, by which the application rate R exceeds the LDso dose D, from H I = R/IOOD. The
4-2 (2)
4 (2)
2-5-3 (2)
Wind speed (ht) (m s- t) (m)
104
104
70
Field width (m)
short grass
stubble
? bare arable
Field surface
polyethylene pins 3 x 50 mm high (ground) vertical rods 3 x 50ram high (55) petri dishes 140ram diam. (ground)
Targets (cm above ground)
B'
B
A
Fig. 2 reference
2.9 (?) 2"4 (1"5)
I. Shires & Bennett (1985) 2. Ware et at,. (1984) 3. Yates et aL (1967) (a) (b) (c) (d) ,r 4. Yates et al. (1974) (a) (b) 5. Yates et al. 0978) (a) (b) 6. Riley et al. (1989)
3.8-7.2 (2-4) 3"2-4"4 (2"4) 4-2 (5) 3' I (5) 2"8-3'5 (1-0)
1-3-4.2 (I-2)
Wind speed (ht) (In s- i) (m)
Reference
?0 -0-5 0-1-1"2 - 1.7-0 1.0 0 -0"1-0-1 0"24)'7 -0.48 0"28 'very stable'
Atmospheric stability ratio ! I 5 3 I I 6 3 ! I I
Number of trials
alfalfa short grass potatoes
100 800
mature wheat cotton alfalfa
Field surface
200
c. 250 137 200
Field width (m)
D E
C
Fig. 2 reference
TABLE 2 Hydraulic Aerial Spraying. Experimental Conditions During Six Spray Drift Assessment Studies used in Constructing Fig. I and Fig. 2. A Singleengine, Fixed-wing Aircraft was Used in all Studies Except 5b where a Helicopter was Used
Grover et al. (1978)
Nordby & Skuterud (1974)
Arvidsson (1985)
Reference
TABLE I Hydraulic Ground Spraying. Experimental Conditions During Three Spray Drift Assessment Studies Used in Constructing Fig. 2
bO L~
1"4
B. N. K. Davis, C. T. Williams
252
10-
5cO
¢0 ¢D
1.0-', ¢o
m
0.5-
2
C
~
0.1-
0
~ 0
0.05 -
'\
0w 0
\
\
\,
0
o
0.01 -
"3b
I
I
I
I
I
10
50
100
500
1000
Distance downwind (m)
Fig. 1. Relationships between concentration of drift deposits and distance downwind of ground crops sprayed from the air with hydraulic equipment. ( - - ) = stable and moderately stable conditions; (. . . . ) = neutral stability: ( - - - ) = moderately unstable conditions. Reference numbers as in Table 2.
reciprocal of tTaisvalue gives the fraction of the application rate required just to produce an LDs0 dose. This spray concentration will occur at a particular distance downwind depending on the extent of spray drift. For example, the hazard index for fenitrothion at an application rate of 700 g a.i. h a - ~ and LD~0 of0.018 is 700/1.8 = 389, and 0.26% of this application would produce the LDso dose. The LDso distance is read offthe appropriate curve in Fig. 2 for ground or aerial spraying. It should be noted that downwind distances are sometimes measured from the edge of the target area (Ware et al., 1984; Shires & Bennett, 1985) and sometimes from the centre line of the last swath i.e. 5 m from the edge of the target area in Yates et al. (1974, 1978), and 17 m
Buffer zone widthsfor honeybees
+'°I
\
\,
¢D
•\
253
\
",,
\
m
5
a o
../ 0.5
0 O. @
o
\
0.1
0.05
1
5
10
50
100
500
Distance downwind (m)
Fig. 2. Relationships between concentration of drift deposits and distance downwind of areas sprayed from the ground (A, B, B') and from the air (C, D, E) with hydraulic nozzles. A and B represent low (_< 3 m s-~) and high (4 m s - t ) wind speeds respectively with deposits measured on narrow cylindrical targets: B' shows deposits at high wind speed on flat targets. C, D and E represent contrasting atmospheric conditions with deposits measured on flat targets. See Tables l and 2 for details.
from the edge in Riley et al. (1989). These distances must, therefore, be deducted from the values read off the curves i n Fig. 2 to make them comparable. RESULTS Table 3 gives the LDso values for 20 insecticides, and the range of ground application rates for six crops in Britain. Except for phosalone, the rates for
254
B. N. K. Davis, C. T. Williams
TABLE 3 Contact Toxicity of 20 Insecticides to Honeybees, and Application Rates in Hydraulic Spraying to Ground Crops from Ground and Air, and Hazard Index from Ground Spraying Compound
LDso (Itg/bee) ~
Application rate (g a.i. ha- 1) Ground b
Azinphos-methyl Carbaryl Chiorfenvinphos Chlorpyrifos Cypermethrin Deltamethrin Demeton-S-methyl Diflubenzuron Dimethoate Endosulfan Fenitrothion Fenvalerate HCH Malathion Mevinphos Permethrin Phosalone Pirimicarb Pirimiphos-methyl Triazophos
0-063 1.3 4.1 0"059 0.035* 0.035* 0"26 > 30 0.51" 7.1 0.018 0.209 0.33* 0.27 0.07 0.075* 8.9 > 54 0.39 0.055
310(W)--450(B) 850(B)-2 000(P) 672(W)--2 350(C) 480(C)-720(B) 12-5(B)-75(P) 6.3(W)-7.5(P) 122(W}--325(B) 100(B) 335(WJ--400(B) 420(R) 700(P) 15(B)-30(P) 11.2(P)-I 120(W) 1 260(S) 140--275(B) 40(PW)-100(B) 490(W)--700(R) 140(S)-210(B) 900(R)--2 100(C) 350(B)-1 050(C)
Hazard index (ground)
Air c
280~: 1 700 --25 7-5§ 244 -160 -700 30 ---25 700a 140 -353
49.2-71.4 6-5-15.4 1.6--5.7 81.4-122 3.6-21.4 1.8-2-1 4.7-12.5 < 0.03 6-6-7-8 0.6 389 0.7-1.4 0.3-33.9 46.7 20.0-39.3 5.3-13.3 0.6-0.8 <0-04 23.1-53-8 63.6-191
° From Stevenson (1978), Smart and Stevenson (1982), Worthing (1983) or Hill (1985); * means or mid-values of a range. b From Scopes and Stables (1989). B = brassicas, C = carrots, P = peas, R = oilseed rape, S = several crops, W = wheat/barley. c Maximum use on peas, from Processors and Growers Research Institute (1988 pets. comm.); ~ only available with demeton-S-methyl sulphone at 84g a.i. ha- 1; § used only with heptenophos at 120g a.i. ha- 1. d Manufacturer's recommended rate for brassica crops.
aerial a p p l i c a t i o n s a p p l y to p e a c r o p s ( P r o c e s s o r s a n d G r o w e r s R e s e a r c h O r g a n i s a t i o n , pers. c o m m . ) t h o u g h v e r y little c a r b a r y l is used, a n d a z i n p h o s m e t h y l is u s e d o n l y in m i x t u r e s w i t h d e m e t o n - S - m e t h y l s u l p h o n e o n peas. T h e h a z a r d i n d e x v a l u e s in T a b l e 3 s p a n o v e r f o u r o r d e r s o f m a g n i t u d e f r o m < 0 . 0 4 f o r p i r i m i c a r b to 389 f o r f e n i t r o t h i o n . ( F e n i t r o t h i o n c a n b e u s e d at a r a t e o f 1050 g a.i. h a - 1 f o r a p p l e s a n d s o f t fruit giving a h a z a r d i n d e x o f 583.) T h e i n d e x is the s a m e as t h a t d e v i s e d b y S m a r t a n d S t e v e n s o n (1982) b u t d i v i d e d b y 100 so t h a t a v a l u e o f 1"0 r e p r e s e n t s the L D s o dose. T a b l e 4 s h o w s t h a t the m a x i m u m L D s o z o n e o f risk to h o n e y b e e s f r o m
255
Buffer zone widths for honeybees
TABLE 4 Maximum LDso Distances (m) for Honeybees Receiving Insecticide Drift from Spraying of Ground Crops. Ground Spraying at Low and High Wind Speeds, Aerial Spraying in Different Atmospheric Conditions. Range of Values Indicates Range of Dose Rates Wind/atmospheric conditions Curve in Fig. 2
Azinphos-methyl Carbaryl Chlorfenvinphos Chlorpyrifos Cypermethrin Deltamethrin Demeton-S-methyl Diflubenzuron Dimethoate Endosulfan Fenitrothion Fenvalerate HCH Malathion Mevinphos Permethrin Phosalone Pirimicarb Pirimiphos-methyi Triazophos
Ground spraying 2"5-3 m s - ~ A
2-5 <1 <1 6-9 <1 < 1 < 1 0 <1 0 21 0 0-1 2 <1 <1 0 0 < 1-3 4-14
4 m s- ~ B
23-33 < 1-4 < 1 36-46 < I-7 < l < 1-2 0 <1 0 (72)" < I < 1-15 22 6-19 < 1-3 0 0 8-25 29-56
Aerial spraying unstable C
(17) (7) ~ -<5 <5 (5) -<5 -61 <5 ---<5 0 0 -21
neutral D
stable E
49 20 --(12) <5 (I 6) -(5) -215 <5 ---(6) 0 0 -61
95 32 --21 (8) 26 -(l I ) -755 (5) ---( 11 ) 0 0 -135
a Distances in parentheses are read from extrapolated curves (shown by dashes) in Fig. 2. b __ not approved for aerial spraying on crops.
g r o u n d s p r a y i n g at l o w w i n d s p e e d (_< 3 m s - 1 ) is u p t o 5 m f o r t h e g r e a t m a j o r i t y o f c o m p o u n d s a n d d o s e rates; o n l y c h l o r p y r i f o s , f e n i t r o t h i o n a n d t r i a z o p h o s at t h e h i g h e r d o s e r a t e s p o s e h a z a r d s b e y o n d t h e i m m e d i a t e field boundary. At higher wind speeds (_>4ms-l), azinphos-methyl, HCH, malathion, mevinphos and pirimiphosmethyl start to pose serious drift h a z a r d s , w i t h m a x i m u m L D s o d i s t a n c e s o f _> 1 5 m , w h i l e c h l o r p y r i f o s , f e n i t r o t h i o n a n d t r i a z o p h o s m a y h a v e L D s o d i s t a n c e s o f > 4 0 m. A e r i a l s p r a y i n g in u n s t a b l e a t m o s p h e r i c c o n d i t i o n s w o u l d a p p e a r t o r e s u l t in s p r a y d r i f t d e p o s i t s o f t h e s a m e o r d e r o f m a g n i t u d e , b e t w e e n a b o u t 1 0 m a n d 1 0 0 m d o w n w i n d , as f r o m g r o u n d s p r a y i n g a t t h e h i g h e r w i n d s p e e d ( c o m p a r e c o l u m n s 2 a n d 3 in T a b l e 4 a n d lines B a n d C in Fig. 2). However, these spray drift deposition curves from aerial spraying may
256
B. N. K. Davis, C. T. Williams
greatly underestimate the risk to flying insects which are extremely efficient collectors of spray droplets and are likely to receive higher concentrations at a given distance than the broad flat targets used in those studies. Curves B and B' in Fig. 2 show about an order of magnitude difference in downwind distances for narrow and flat targets, respectively, receiving the same deposition from ground spraying at almost the same wind speed (Table 1). This factor is particularly relevant to unstable conditions which often coincide with dry, sunny weather when evaporation from spray droplets is rapid. As mentioned earlier, the most serious dangers from aerial spraying evidently arise under stable atmospheric conditions when a cloud of fine particles can drift for considerable distances. Under these conditions, even relatively safe compounds, such as carbaryl and demeton-S-methyl, could apparently have m a x i m u m LDso distances of > 25 m, azinphos-methyl and triazophos could have the same effects at distances of 100 m or more, while fenitrothion could severely affect areas a few fields away (755 m).
DISCUSSION Figs 1 and 2 are plotted on log log axes and the ordinates show % sprayer emission (application) rates in order to plot all the curves on the same basis. These curves will not normally pass through the 100% level at zero distance from the edge of the target area because of the cumulative drift from other upwind swaths. Gilbert and Bell (1988) give multiplication factors for 1-10 swaths showing, for example, that the total drift 8 m downwind of a 104 m wide sprayed area will be 3.3 times that from a single 12m wide sprayed swath. Aerial spraying also produces more turbulence than ground spraying does, and this results in very variable deposition within about 20 m of the sprayed area. Shires and Bennett (1985) found that average deposits were 9% of emission rate at ground level on ditch banks immediately downwind of sprayed wheat fields ( < 10m), but they recorded deposits of 29% at similar distances upwind. Ware et al. (1972) used aspirated air samples to measure vertical concentration profiles in the drift cloud produced by an aerial application. At 2 5 m downwind, they found that the drift concentration was 5"6 times greater at 15 m than at 0"3 m above the ground. Insects at tree height close to a sprayed area are therefore likely to be at much higher risk from aerial spraying than they are from ground spraying. The curves in Fig. 2 have been selected to provide worst case scenarios. If the results from Riley et al. (1989) are taken as standard for stable atmospheric conditions (Fig. 1, line 6), then the honeybee LDso distances for
Buffer zone widthsfor honeybees
257
azinphos-methyl, triazophos and fenitrothion spray drift would be 12, 21 and 163 m respectively. Values for all other compounds would be < 1 m. The measurements of spray drift clearly depend on a great many weather, site and operator-controlled variables. The lack of comparability in methodology, e.g. in width and nature of sprayed surface, and in type of targets used for collecting drift deposition, makes it difficult to compare results, especially between ground and aerial spraying. The maximum LDs0 distance values estimated here are based on available data but should not be considered as providing more than a provisional basis for determining the widths of buffer zones to minimise the risk to honeybees. It is well known that different insects show very different sensitivities to particular insecticides, and one would therefore expect quite different results for other beneficial and non-target species from spray drift. Hassan et al. (1987) have compared the sensitivities of a range of beneficial species when exposed to residues of 22 pesticides in standardized laboratory and field tests. Although the results are not closely comparable with those for bees they provide clear examples of such different sensitivities: deltamethrin is highly toxic ('harmful') to the predaceous mite Phytoseiulus persimilis but 'harmless' to the ichneumonid Coccygomimus turionellae and ground beetle Pterostichus cupreus whereas fenitrothion has quite the reverse toxicity ratings for these species. Sinha et al. (1990) have obtained LDs0 values for topical application of eight insecticides to 2-day-old Pieris brassicae larvae. In comparison with bees, these compounds range from 6200 times less toxic to P. brassicae for dimethoate, to 250 times more toxic for diflubenzuron. Assuming direct exposure to spray drift, simplistic extrapolation for this species would indicate effects only from pirimiphos-methyl and fenitrothion. The LDso distance for pirimiphos-methyl at the higher application rate would thus be 25 m. For fenitrothion, the LDso distances for ground and aerial spraying would similarly be 4, (8), 24 and 37 m for curves B, C, D and E in Fig. 2, respectively. In practice, initial field trials indicate a lower hazard from fenitrothion than predicted by this means, and a higher hazard from diflubenzuron. Stomach poisoning from ingesting residues may have played an important role in these trials and this again points to the need for caution in extrapolating from the results in Table 4. ACKNOWLEDGEMENTS This work was funded by both the Nature Conservancy Council and the Department of the Environment (PECO 7/2/50). We are grateful to Dr A. S. Cooke, Dr J. H. Stevenson and referees for helpful comments on the manuscript.
258
B. N. K. Davis, C. T. Williams REFERENCES
Arvidsson, T. (1985). Phytotoxic damage and residues of MCPA in spring rape caused by wind drift. In Weeds and Weed Control, 1, 26th Swedish Weed Conference, Uppsala, pp. 302-10. Byass, J. B. & Lake, J. R. (1977). Spray drift from a tractor powered field sprayer. Pestic. Sci., 8, 117-26. Cuthbertson, P. S. & Jepson, P. C. (1988). Reducing pesticide drift into the hedgerow by the inclusion of an unsprayed field margin. Proceedings BCPC Conference (Pests and Diseases), Vol. 2, Brighton, pp. 747-51. Elliott, J. G. & Wilson, B. J. (eds) (1983). The Influence of the Weather on the Efficiency and Safety of Pesticide Application: The Drift of Herbicides. BCPC occasional publication No. 3, BCPC, London. Gilbert, A. J. & Bell, G. J. (1988). Evaluation of the drift hazards arising from pesticide spray application. Aspects Appl. Biol., 17, 363-76. Goering~ C. E. & Butler, B. J. (1975). Paired field studies of herbicide drift. Trans. Am. Soc. Agric. Engrs., 18, 27-34. Grover, R., Kerr, L. A., Maybank, J. & Yoshida, K. (1978). Field measurements of droplet drift from ground sprayers: I. Sampling, analytical and data integration techniques. Can. J. Pl. Sci., 58, 611-22. Harpaz, I. (1979). Regulatory measures adopted in Israel regarding the screening and field application of pesticides within the context of integrated control. International Pest Control., 21, 28-30. Hassan, S. A., Albert, R., Bigler, F., Blaisinger, P. et al. (1987). Results of the third joint pesticide testing programme by the IOBC/WPRS-Working Group 'Pesticides and Beneficial Organisms'. J. Appl. Ent., 103, 92-107. Hill, I. R. (1985). Effects on non-target organisms in terrestrial and aquatic environments. In The Pyrethroid Insecticides, ed. J. P. Leahey, Taylor and Francis, London, pp. 151-262. Joyce, R. J. V. 0985). Application from the air. In Pesticide Application: Principles and Practice, ed. P. T. Haskell, Clarendon, Oxford, pp. 118-52. Lloyd, G. A. & Bell, G. J. (1982). Relative contamination of humans and the environment by spray drift from use of a tractor mounted drift sprayer (the Ulvamast) and conventional hydraulic equipment. Proceedings of the 1982 British Crop Protection Conference--Weeds, Vol. 3, pp. 1045-54. MAFF/HSE (1989). Pesticides 1989. MAFF reference book 500, HMSO, London. Marrs, R. H., Williams, C. T., Frost, A. J. & Plant, R. A. (1989). Assessment of the effects of herbicide spray drift on a range of species of conservation interest. Environ. Pollut., 59, 71-86. Maybank, J., Yoshida, K. & Grover, R. (1978). Spray drift from agricultural pesticide applications. J. Air Pollution Control Ass., 28, 1009-14. Nordby, A. & Skuterud, R. (1974). The effects of boom height, working pressure and wind speed on spray drift. Weed Res., 14, 385-95. Pasquill, F. (1961). The estimation of the dispersion of windborne material. Met. Mag., Lond., 90, 33-49. Payne, N. J., Helson, B. V., Sundaram, K. M. S. & Fleming, R. A. (1988). Estimating buffer zone widths for pesticide applications. Pestic. ScL, 24, 147-61. Quantick, H. R. (1985). Aviation in Crop Protection, Pollution and Insect Control. Collins, London.
Buffer zone widthsfor honeybees
259
Riley, C. M., Wiesner, C. J: & Ernst, W. R. (1989). Off-target deposition and drift of aerially applied agricultural sprays. Pestic. Sci., 26, 159-66. Scopes, N. & Stables, L. (eds) (1989). Pest andDisease Control Handbook. 3rd edn, BCPC, London. Shires, S. W. & Bennett, D. (1985). Contamination and effects in freshwater ditches resulting from an aerial application ofcypermethrin. Ecotoxicol. Environ. Safe., 9, 145-58. Sinha, S. N., Lakhani, K. H. & Davis, B. N. K. (1990). Studies on the toxicity of insecticidal drift to the first instar larvae of the large white butterfly Pieris brassicae L. (Lepidoptera: Pieridae). Ann. Appl. Biol. 116 (in press). Smart, L. E. & Stevenson, J. H. (1982). Laboratory estimation of toxicity of pyrethroid insecticides to honeybees: relevance to hazard in the field. Bee World, 63, 150-2. Stevenson, J. H. (1978). The acute toxicity of unformulated pesticides to worker honeybees (Apis mellifera L.). Plant Path., 27, 38--40. Ware, G. W., Cahill, W. P., Gerhardt, P. D. & Witt, J. M. (1970). Pesticide drift. IV. On-target deposits from aerial applications of insecticides. J. Econ. Ent., 63, 1982-3. Ware, G. W., Estesen, B. J., Cahill, W. P. & Frost, K. R. (1972). Pesticide drift. V. Vertical drift from aerial spray applications. J. Econ. Ent., 65, 590-2. Ware, G. W., Buck, N. A. & Estesen, B. J. (1984). Deposit and drift losses from aerial ultra-low-volume and emulsion sprays in Arizona. J. Econ. Ent., 77, 298-303. Williams, C. T., Davis, B. N. K., Marrs, R. H. & Osborn, D. (1987). Impact of Pesticide Drift. NERC contract report to NCC, Institute of Terrestrial Ecology, Huntingdon. Worthing, C. R. (ed.) (1983). The Pesticide Manual, 7th edn. BCPC, Croydon. Yates, W. E., Akesson, N. B. & Coutts, H. H. (1967). Drift hazards related to ultralow-volume and diluted sprays applied by agricultural aircraft. Trans. Am. Soc. Agric. Engrs., 10, 628-32. Yates, W. E., Akesson, N. B. & Cowden, R. E. (1974). Criteria for minimizing drift residues on crops downwind from aerial sprays. Trans. Am. Soc. Agric. Engrs., 17, 627-32. Yates, W. E., Akesson, N. B. & Bayer, D. E. (1978). Drift of glyphosate sprays applied with aerial and ground equipment. Weed Sci., 26, 597-604. Yates, W. E., Akesson, N. B. & Brazelton, R. W. (1981). Systems for the safe use of pesticides. Outl. Agric., 10, 321-6.
Buffer Zone Widths for Honeybccs from Ground and Aerial Spraying of Insecticides B. N. K. D a v i s & C. T. W i l l i a m s * NERC, Institute of Terrestrial Ecology, Monks Wood Experimental Station, Abbots Ripton, Huntingdon PEI7 2LS, UK (Received 21 July 1989; revised version received 27 September 1989; accepted 28 September 1989) ABSTRACT The relative field hazards of insecticides to honeybees have been estimated by considering intrinsic toxicity levels andfield application rates. This approach is extended here to a consideration of buffer zones downwind of sprayed areas by estimating the distance at which bees would encounter an LDso dose from spray drift. 'LDso distances' are determined for both ground and aerial spraying of ground crops in Britain using published data on spray deposition under various weather conditions. For ground spraying at low wind speeds ( < 3 m s- 1), this zone of risk is up to 5 m for thegreat majority of compounds. Aerial spraying in unstable atmospheric conditions appears to produce drift deposits of about the same order of magnitude as from ground spraying at wind speeds of about 4 m s - l , with maximum LDso distances of <4Ore for chlorpyrifos, .fenitrothion and triazophos. For aerial spraying in stable atmospheric conditions these distances would be much greater. Pieris brassicae larvae are contrasted with honeybees in their relative sensitivities to insecticides and consequent LDso distances.
INTRODUCTION U n d e r the Control o f Pesticides Regulations (1986), substances are classified into four categories in relation to their toxicity to honeybees (as in the former Pesticide Safety Precaution Scheme), viz. 'extremely d a n g e r o u s ' n L D s o * Present address: Dodo Creek Research Station, Ministry of Agriculture and Lands, PO Box G l3, Honiara, Solomon Islands. 247 Environ. Pollut. 0269-7491/90/$03-50 © 1990 Elsevier Science Publishers Ltd, England. Printed in Great Britain
248
B. N. K. Davis, C. T. Williams
<0"1 ~tg bee- 1; ,dangerous, LDso 0.1-1.0; 'harmful'--LDso 1.0-10-0; and not classified--LDso >10tag bee -1 The actual danger from any compound, however, depends also on the extent of exposure in the field. As Smart and Stevenson (1982) said: 'This hazard is a function of intrinsic toxicity of the pesticide, the field rate (weight of active ingredient per hectare) at which it is applied, the proportion of the dose which is available for transfer to the bees at risk, and the behaviour of the bee itself.' This statement is true for any non-target organism, and the danger from pesticides is not necessarily confined to the crop itself since spray drift can affect non-target areas. In a few cases, buffer zones have been set to prevent harmful effects of insecticide spray drift on beneficial insects or wildlife. In Israel, for example, aerial spraying of cotton is not permitted within 200 m of citrus orchards because of the risk to natural predators whose elimination has caused pest outbreaks in the orchards (Harpaz, 1979). In Canada, where permethrin is used to control forest insect pests, buffer zones have been determined for 'worst case' scenarios, so as to minimise the risk to aquatic invertebrates and fish (Payne et al., 1988). A 20 m swath width was found to be adequate for limiting mortality to acceptable levels during ground-based applications. The use of unsprayed 'conservation headlands' 6 m wide in cereal fields in Britain is advocated by the Game Conservancy Trust to increase the insect fauna for the benefit of partridge chicks, and this measure also reduces pesticide drift into hedgerows (Cuthbertson & Jepson, 1988). The Nature Conservancy Council is concerned with the hazards of spray drift to wildlife generally, especially from aerial spraying. The Control of Pesticide Regulations (I 986) places an obligation on operators to consult the NCC if aerial spraying is to be undertaken within 0.75 nautical mile of a Site of Special Scientific Interest. However, more detailed knowledge is needed on buffer zones for both ground and aerial spraying of pesticides, and to this end, Sinha et al. (1990) have estimated ds0 values in bioassay studies of spray drift, i.e. the distance at which organisms will receive an LDso dose under given conditions. This paper compares the environmental hazards of 20 insecticides used in Britain for ground and/or aerial spraying of agricultural crops by estimating the respective 'LDso distance' values for honeybees from published data.
SPRAY D R I F T DEPOSITION The extent of spray drift depends on three sets of factors: (a) meteorological factors such as wind speed, turbulence and atmospheric stability, (b) the 'roughness length' of the surface over which the wind is flowing, and (c)
Buffer zone widths for hone),bees
249
operator-controlled variables, such as application method, type of nozzle, number of swaths and height of release (Elliott & Wilson, 1983; Joyce, 1985; Quantick, 1985; Gilbert & Bell, 1988). Briefly, atmospheric conditions are said to be unstable, neutral or stable depending on the temperature profile and vertical movement of air. Unstable atmospheric conditions occur when the ground and vegetation heats up on a sunny, clear day, producing a rising column of warm air (thermals), and sucking in cooler air near ground level. Stable conditions develop when the ground cools faster than the air above (by radiating heat to the sky) and then cools the adjacent layer ofair; they are typical of late evenings to early mornings with clear skies and little or no wind. Neutral conditions are when there is no tendency for air to rise or fall vertically. Pasquill (1961) has defined five categories, from A, very unstable, to F, very stable, which are generally used by agrometeorologists in Britain. However, this paper uses a "stability ratio" index in which values less than zero indicate unstable conditions, those between c. -0.1 and 4-0" 1 neutral stability, those between 0.1 and 1.2 moderately stable, and those above 1.2 very stable conditions (Yates et al., 1981). Stable conditions, with low wind speeds and reduced turbulence, increase pesticide drift hazards because a slowly diffusing plume of small drops can drift well off the target area in an unpredictable way. Roughness length is a parameter determined by the vegetation height and density which affects the frictional turbulence of the air flow. Spray particles are more likely to drift over short, smooth surfaces, such as short grass, than over tall, rough vegetation, such as a wheat crop which creates vertical eddies. For commonly used ground crop sprayers, fitted with hydraulic pressure nozzles and using recommended procedures, wind speed is considered the most important factor affecting drift (Nordby & Skuterud, 1974; Goering & Butler, 1975; Byass & Lake, 1977; Maybank et al., 1978; Lloyd & Bell, 1982; Arvidsson, 1985). Aerial spraying generally produces considerably more drift, with estimates of losses from the target area ranging from 20-40% (Maybank et al., 1978; Shires & Bennett, 1985) to > 50% (Ware et al., 1970; Joyce, 1985) of the total amount applied. It is most important to avoid spraying in stable atmospheric conditions. In general, wind speed is a less dominant influence for aerial sprayers than it is for ground sprayers, probably because of the overriding influence of the aircraft wake. Nevertheless, aerial spraying should only be carried out in wind speeds of <10 knots ( < 5 m s -1) (MAFF/HSE, 1989). Operator precision, in spray release height and in switching nozzles on and off at the beginning and end of spray runs, is especially important. Evaporation of water-based sprays is a particular
250
B. N. K. Davis, C. T, Wilfiams
problem because droplets take longer to reach the crop than for ground sprayers, and can produce aerosols consisting of almost neat pesticide.
Drift from ground spraying Several studies of drift from ground sprayers using conventional hydraulic systems give estimates of deposits downwind of multiple-swath applications, or provide enough single-swath data for such estimates to be calculated. The results for five studies are illustrated by Marrs et al. (1989) (from Williams et al., 1987). They show that concentrations of spray deposits at any particular distance downwind could vary by between one and two orders of magnitude; deposits were greater at higher wind speeds and on narrow cylindrical targets rather than on flat targets. Curves of maximum drift deposits on cylindrical targets are shown here in Fig. 2, lines A and B for wind speeds of 2.5-3 and 4 m s- 1, respectively. Figure 2, line B' shows drift deposits on fiat targets at the higher wind speed for comparison. Nordby and Skuterud (1974) obtained drift deposit values for a still lower wind speed of 1-5 m s-~, the curve (4a in Marrs et aL, 1989) having a similar but slightly lower trajectory to that shown in Fig. 2, line A. The experimental conditions for these three studies used to construct Fig. 2 are summarized in Table 1.
Drift from aerial spraying There is a considerable amount of data on drift from aerial spraying with standard hydraulic pressure nozzles but all measurements of deposits have been made with fiat, horizontal targets which have low collection efficiency for small droplets. Most of the studies have been undertaken in southern USA but examples are drawn also from Britain and Canada. Table 2 and Fig. 1 summarize the spraying conditions and illustrate the results of eleven trials. A subset of these data, representing the curves for m a x i m u m deposits under three sets of contrasting conditions of atmospheric stability (Fig. 2, lines C, D and E) are used in distance estimation.
Hazard indices and LD5o distances The field application rate of a pesticide can be compared with the topical LDso dose for a honeybee (in/zg per bee) to obtain a notional index of field hazard. If the target area of a bee in a spray cloud is taken to be 1 cm 2, and it is assumed to receive an instantaneous dose, a sprayer application rate of 100g a.i. h a - 1 would produce a deposit on a bee of 1/zg (since 100gha- 1 = 1/ag c m - 2). One can thus calculate the factor, or hazard index HI, by which the application rate R exceeds the LDso dose D, from H I = R/IOOD. The
4-2 (2)
4 (2)
2-5-3 (2)
Wind speed (ht) (m s- t) (m)
104
104
70
Field width (m)
short grass
stubble
? bare arable
Field surface
polyethylene pins 3 x 50 mm high (ground) vertical rods 3 x 50ram high (55) petri dishes 140ram diam. (ground)
Targets (cm above ground)
B'
B
A
Fig. 2 reference
2.9 (?) 2"4 (1"5)
I. Shires & Bennett (1985) 2. Ware et at,. (1984) 3. Yates et aL (1967) (a) (b) (c) (d) ,r 4. Yates et al. (1974) (a) (b) 5. Yates et al. 0978) (a) (b) 6. Riley et al. (1989)
3.8-7.2 (2-4) 3"2-4"4 (2"4) 4-2 (5) 3' I (5) 2"8-3'5 (1-0)
1-3-4.2 (I-2)
Wind speed (ht) (In s- i) (m)
Reference
?0 -0-5 0-1-1"2 - 1.7-0 1.0 0 -0"1-0-1 0"24)'7 -0.48 0"28 'very stable'
Atmospheric stability ratio ! I 5 3 I I 6 3 ! I I
Number of trials
alfalfa short grass potatoes
100 800
mature wheat cotton alfalfa
Field surface
200
c. 250 137 200
Field width (m)
D E
C
Fig. 2 reference
TABLE 2 Hydraulic Aerial Spraying. Experimental Conditions During Six Spray Drift Assessment Studies used in Constructing Fig. I and Fig. 2. A Singleengine, Fixed-wing Aircraft was Used in all Studies Except 5b where a Helicopter was Used
Grover et al. (1978)
Nordby & Skuterud (1974)
Arvidsson (1985)
Reference
TABLE I Hydraulic Ground Spraying. Experimental Conditions During Three Spray Drift Assessment Studies Used in Constructing Fig. 2
bO L~
1"4
B. N. K. Davis, C. T. Williams
252
10-
5cO
¢0 ¢D
1.0-', ¢o
m
0.5-
2
C
~
0.1-
0
~ 0
0.05 -
'\
0w 0
\
\
\,
0
o
0.01 -
"3b
I
I
I
I
I
10
50
100
500
1000
Distance downwind (m)
Fig. 1. Relationships between concentration of drift deposits and distance downwind of ground crops sprayed from the air with hydraulic equipment. ( - - ) = stable and moderately stable conditions; (. . . . ) = neutral stability: ( - - - ) = moderately unstable conditions. Reference numbers as in Table 2.
reciprocal of tTaisvalue gives the fraction of the application rate required just to produce an LDs0 dose. This spray concentration will occur at a particular distance downwind depending on the extent of spray drift. For example, the hazard index for fenitrothion at an application rate of 700 g a.i. h a - ~ and LD~0 of0.018 is 700/1.8 = 389, and 0.26% of this application would produce the LDso dose. The LDso distance is read offthe appropriate curve in Fig. 2 for ground or aerial spraying. It should be noted that downwind distances are sometimes measured from the edge of the target area (Ware et al., 1984; Shires & Bennett, 1985) and sometimes from the centre line of the last swath i.e. 5 m from the edge of the target area in Yates et al. (1974, 1978), and 17 m
Buffer zone widthsfor honeybees
+'°I
\
\,
¢D
•\
253
\
",,
\
m
5
a o
../ 0.5
0 O. @
o
\
0.1
0.05
1
5
10
50
100
500
Distance downwind (m)
Fig. 2. Relationships between concentration of drift deposits and distance downwind of areas sprayed from the ground (A, B, B') and from the air (C, D, E) with hydraulic nozzles. A and B represent low (_< 3 m s-~) and high (4 m s - t ) wind speeds respectively with deposits measured on narrow cylindrical targets: B' shows deposits at high wind speed on flat targets. C, D and E represent contrasting atmospheric conditions with deposits measured on flat targets. See Tables l and 2 for details.
from the edge in Riley et al. (1989). These distances must, therefore, be deducted from the values read off the curves i n Fig. 2 to make them comparable. RESULTS Table 3 gives the LDso values for 20 insecticides, and the range of ground application rates for six crops in Britain. Except for phosalone, the rates for
254
B. N. K. Davis, C. T. Williams
TABLE 3 Contact Toxicity of 20 Insecticides to Honeybees, and Application Rates in Hydraulic Spraying to Ground Crops from Ground and Air, and Hazard Index from Ground Spraying Compound
LDso (Itg/bee) ~
Application rate (g a.i. ha- 1) Ground b
Azinphos-methyl Carbaryl Chiorfenvinphos Chlorpyrifos Cypermethrin Deltamethrin Demeton-S-methyl Diflubenzuron Dimethoate Endosulfan Fenitrothion Fenvalerate HCH Malathion Mevinphos Permethrin Phosalone Pirimicarb Pirimiphos-methyl Triazophos
0-063 1.3 4.1 0"059 0.035* 0.035* 0"26 > 30 0.51" 7.1 0.018 0.209 0.33* 0.27 0.07 0.075* 8.9 > 54 0.39 0.055
310(W)--450(B) 850(B)-2 000(P) 672(W)--2 350(C) 480(C)-720(B) 12-5(B)-75(P) 6.3(W)-7.5(P) 122(W}--325(B) 100(B) 335(WJ--400(B) 420(R) 700(P) 15(B)-30(P) 11.2(P)-I 120(W) 1 260(S) 140--275(B) 40(PW)-100(B) 490(W)--700(R) 140(S)-210(B) 900(R)--2 100(C) 350(B)-1 050(C)
Hazard index (ground)
Air c
280~: 1 700 --25 7-5§ 244 -160 -700 30 ---25 700a 140 -353
49.2-71.4 6-5-15.4 1.6--5.7 81.4-122 3.6-21.4 1.8-2-1 4.7-12.5 < 0.03 6-6-7-8 0.6 389 0.7-1.4 0.3-33.9 46.7 20.0-39.3 5.3-13.3 0.6-0.8 <0-04 23.1-53-8 63.6-191
° From Stevenson (1978), Smart and Stevenson (1982), Worthing (1983) or Hill (1985); * means or mid-values of a range. b From Scopes and Stables (1989). B = brassicas, C = carrots, P = peas, R = oilseed rape, S = several crops, W = wheat/barley. c Maximum use on peas, from Processors and Growers Research Institute (1988 pets. comm.); ~ only available with demeton-S-methyl sulphone at 84g a.i. ha- 1; § used only with heptenophos at 120g a.i. ha- 1. d Manufacturer's recommended rate for brassica crops.
aerial a p p l i c a t i o n s a p p l y to p e a c r o p s ( P r o c e s s o r s a n d G r o w e r s R e s e a r c h O r g a n i s a t i o n , pers. c o m m . ) t h o u g h v e r y little c a r b a r y l is used, a n d a z i n p h o s m e t h y l is u s e d o n l y in m i x t u r e s w i t h d e m e t o n - S - m e t h y l s u l p h o n e o n peas. T h e h a z a r d i n d e x v a l u e s in T a b l e 3 s p a n o v e r f o u r o r d e r s o f m a g n i t u d e f r o m < 0 . 0 4 f o r p i r i m i c a r b to 389 f o r f e n i t r o t h i o n . ( F e n i t r o t h i o n c a n b e u s e d at a r a t e o f 1050 g a.i. h a - 1 f o r a p p l e s a n d s o f t fruit giving a h a z a r d i n d e x o f 583.) T h e i n d e x is the s a m e as t h a t d e v i s e d b y S m a r t a n d S t e v e n s o n (1982) b u t d i v i d e d b y 100 so t h a t a v a l u e o f 1"0 r e p r e s e n t s the L D s o dose. T a b l e 4 s h o w s t h a t the m a x i m u m L D s o z o n e o f risk to h o n e y b e e s f r o m
255
Buffer zone widths for honeybees
TABLE 4 Maximum LDso Distances (m) for Honeybees Receiving Insecticide Drift from Spraying of Ground Crops. Ground Spraying at Low and High Wind Speeds, Aerial Spraying in Different Atmospheric Conditions. Range of Values Indicates Range of Dose Rates Wind/atmospheric conditions Curve in Fig. 2
Azinphos-methyl Carbaryl Chlorfenvinphos Chlorpyrifos Cypermethrin Deltamethrin Demeton-S-methyl Diflubenzuron Dimethoate Endosulfan Fenitrothion Fenvalerate HCH Malathion Mevinphos Permethrin Phosalone Pirimicarb Pirimiphos-methyi Triazophos
Ground spraying 2"5-3 m s - ~ A
2-5 <1 <1 6-9 <1 < 1 < 1 0 <1 0 21 0 0-1 2 <1 <1 0 0 < 1-3 4-14
4 m s- ~ B
23-33 < 1-4 < 1 36-46 < I-7 < l < 1-2 0 <1 0 (72)" < I < 1-15 22 6-19 < 1-3 0 0 8-25 29-56
Aerial spraying unstable C
(17) (7) ~ -<5 <5 (5) -<5 -61 <5 ---<5 0 0 -21
neutral D
stable E
49 20 --(12) <5 (I 6) -(5) -215 <5 ---(6) 0 0 -61
95 32 --21 (8) 26 -(l I ) -755 (5) ---( 11 ) 0 0 -135
a Distances in parentheses are read from extrapolated curves (shown by dashes) in Fig. 2. b __ not approved for aerial spraying on crops.
g r o u n d s p r a y i n g at l o w w i n d s p e e d (_< 3 m s - 1 ) is u p t o 5 m f o r t h e g r e a t m a j o r i t y o f c o m p o u n d s a n d d o s e rates; o n l y c h l o r p y r i f o s , f e n i t r o t h i o n a n d t r i a z o p h o s at t h e h i g h e r d o s e r a t e s p o s e h a z a r d s b e y o n d t h e i m m e d i a t e field boundary. At higher wind speeds (_>4ms-l), azinphos-methyl, HCH, malathion, mevinphos and pirimiphosmethyl start to pose serious drift h a z a r d s , w i t h m a x i m u m L D s o d i s t a n c e s o f _> 1 5 m , w h i l e c h l o r p y r i f o s , f e n i t r o t h i o n a n d t r i a z o p h o s m a y h a v e L D s o d i s t a n c e s o f > 4 0 m. A e r i a l s p r a y i n g in u n s t a b l e a t m o s p h e r i c c o n d i t i o n s w o u l d a p p e a r t o r e s u l t in s p r a y d r i f t d e p o s i t s o f t h e s a m e o r d e r o f m a g n i t u d e , b e t w e e n a b o u t 1 0 m a n d 1 0 0 m d o w n w i n d , as f r o m g r o u n d s p r a y i n g a t t h e h i g h e r w i n d s p e e d ( c o m p a r e c o l u m n s 2 a n d 3 in T a b l e 4 a n d lines B a n d C in Fig. 2). However, these spray drift deposition curves from aerial spraying may
256
B. N. K. Davis, C. T. Williams
greatly underestimate the risk to flying insects which are extremely efficient collectors of spray droplets and are likely to receive higher concentrations at a given distance than the broad flat targets used in those studies. Curves B and B' in Fig. 2 show about an order of magnitude difference in downwind distances for narrow and flat targets, respectively, receiving the same deposition from ground spraying at almost the same wind speed (Table 1). This factor is particularly relevant to unstable conditions which often coincide with dry, sunny weather when evaporation from spray droplets is rapid. As mentioned earlier, the most serious dangers from aerial spraying evidently arise under stable atmospheric conditions when a cloud of fine particles can drift for considerable distances. Under these conditions, even relatively safe compounds, such as carbaryl and demeton-S-methyl, could apparently have m a x i m u m LDso distances of > 25 m, azinphos-methyl and triazophos could have the same effects at distances of 100 m or more, while fenitrothion could severely affect areas a few fields away (755 m).
DISCUSSION Figs 1 and 2 are plotted on log log axes and the ordinates show % sprayer emission (application) rates in order to plot all the curves on the same basis. These curves will not normally pass through the 100% level at zero distance from the edge of the target area because of the cumulative drift from other upwind swaths. Gilbert and Bell (1988) give multiplication factors for 1-10 swaths showing, for example, that the total drift 8 m downwind of a 104 m wide sprayed area will be 3.3 times that from a single 12m wide sprayed swath. Aerial spraying also produces more turbulence than ground spraying does, and this results in very variable deposition within about 20 m of the sprayed area. Shires and Bennett (1985) found that average deposits were 9% of emission rate at ground level on ditch banks immediately downwind of sprayed wheat fields ( < 10m), but they recorded deposits of 29% at similar distances upwind. Ware et al. (1972) used aspirated air samples to measure vertical concentration profiles in the drift cloud produced by an aerial application. At 2 5 m downwind, they found that the drift concentration was 5"6 times greater at 15 m than at 0"3 m above the ground. Insects at tree height close to a sprayed area are therefore likely to be at much higher risk from aerial spraying than they are from ground spraying. The curves in Fig. 2 have been selected to provide worst case scenarios. If the results from Riley et al. (1989) are taken as standard for stable atmospheric conditions (Fig. 1, line 6), then the honeybee LDso distances for
Buffer zone widthsfor honeybees
257
azinphos-methyl, triazophos and fenitrothion spray drift would be 12, 21 and 163 m respectively. Values for all other compounds would be < 1 m. The measurements of spray drift clearly depend on a great many weather, site and operator-controlled variables. The lack of comparability in methodology, e.g. in width and nature of sprayed surface, and in type of targets used for collecting drift deposition, makes it difficult to compare results, especially between ground and aerial spraying. The maximum LDs0 distance values estimated here are based on available data but should not be considered as providing more than a provisional basis for determining the widths of buffer zones to minimise the risk to honeybees. It is well known that different insects show very different sensitivities to particular insecticides, and one would therefore expect quite different results for other beneficial and non-target species from spray drift. Hassan et al. (1987) have compared the sensitivities of a range of beneficial species when exposed to residues of 22 pesticides in standardized laboratory and field tests. Although the results are not closely comparable with those for bees they provide clear examples of such different sensitivities: deltamethrin is highly toxic ('harmful') to the predaceous mite Phytoseiulus persimilis but 'harmless' to the ichneumonid Coccygomimus turionellae and ground beetle Pterostichus cupreus whereas fenitrothion has quite the reverse toxicity ratings for these species. Sinha et al. (1990) have obtained LDs0 values for topical application of eight insecticides to 2-day-old Pieris brassicae larvae. In comparison with bees, these compounds range from 6200 times less toxic to P. brassicae for dimethoate, to 250 times more toxic for diflubenzuron. Assuming direct exposure to spray drift, simplistic extrapolation for this species would indicate effects only from pirimiphos-methyl and fenitrothion. The LDso distance for pirimiphos-methyl at the higher application rate would thus be 25 m. For fenitrothion, the LDso distances for ground and aerial spraying would similarly be 4, (8), 24 and 37 m for curves B, C, D and E in Fig. 2, respectively. In practice, initial field trials indicate a lower hazard from fenitrothion than predicted by this means, and a higher hazard from diflubenzuron. Stomach poisoning from ingesting residues may have played an important role in these trials and this again points to the need for caution in extrapolating from the results in Table 4. ACKNOWLEDGEMENTS This work was funded by both the Nature Conservancy Council and the Department of the Environment (PECO 7/2/50). We are grateful to Dr A. S. Cooke, Dr J. H. Stevenson and referees for helpful comments on the manuscript.
258
B. N. K. Davis, C. T. Williams REFERENCES
Arvidsson, T. (1985). Phytotoxic damage and residues of MCPA in spring rape caused by wind drift. In Weeds and Weed Control, 1, 26th Swedish Weed Conference, Uppsala, pp. 302-10. Byass, J. B. & Lake, J. R. (1977). Spray drift from a tractor powered field sprayer. Pestic. Sci., 8, 117-26. Cuthbertson, P. S. & Jepson, P. C. (1988). Reducing pesticide drift into the hedgerow by the inclusion of an unsprayed field margin. Proceedings BCPC Conference (Pests and Diseases), Vol. 2, Brighton, pp. 747-51. Elliott, J. G. & Wilson, B. J. (eds) (1983). The Influence of the Weather on the Efficiency and Safety of Pesticide Application: The Drift of Herbicides. BCPC occasional publication No. 3, BCPC, London. Gilbert, A. J. & Bell, G. J. (1988). Evaluation of the drift hazards arising from pesticide spray application. Aspects Appl. Biol., 17, 363-76. Goering~ C. E. & Butler, B. J. (1975). Paired field studies of herbicide drift. Trans. Am. Soc. Agric. Engrs., 18, 27-34. Grover, R., Kerr, L. A., Maybank, J. & Yoshida, K. (1978). Field measurements of droplet drift from ground sprayers: I. Sampling, analytical and data integration techniques. Can. J. Pl. Sci., 58, 611-22. Harpaz, I. (1979). Regulatory measures adopted in Israel regarding the screening and field application of pesticides within the context of integrated control. International Pest Control., 21, 28-30. Hassan, S. A., Albert, R., Bigler, F., Blaisinger, P. et al. (1987). Results of the third joint pesticide testing programme by the IOBC/WPRS-Working Group 'Pesticides and Beneficial Organisms'. J. Appl. Ent., 103, 92-107. Hill, I. R. (1985). Effects on non-target organisms in terrestrial and aquatic environments. In The Pyrethroid Insecticides, ed. J. P. Leahey, Taylor and Francis, London, pp. 151-262. Joyce, R. J. V. 0985). Application from the air. In Pesticide Application: Principles and Practice, ed. P. T. Haskell, Clarendon, Oxford, pp. 118-52. Lloyd, G. A. & Bell, G. J. (1982). Relative contamination of humans and the environment by spray drift from use of a tractor mounted drift sprayer (the Ulvamast) and conventional hydraulic equipment. Proceedings of the 1982 British Crop Protection Conference--Weeds, Vol. 3, pp. 1045-54. MAFF/HSE (1989). Pesticides 1989. MAFF reference book 500, HMSO, London. Marrs, R. H., Williams, C. T., Frost, A. J. & Plant, R. A. (1989). Assessment of the effects of herbicide spray drift on a range of species of conservation interest. Environ. Pollut., 59, 71-86. Maybank, J., Yoshida, K. & Grover, R. (1978). Spray drift from agricultural pesticide applications. J. Air Pollution Control Ass., 28, 1009-14. Nordby, A. & Skuterud, R. (1974). The effects of boom height, working pressure and wind speed on spray drift. Weed Res., 14, 385-95. Pasquill, F. (1961). The estimation of the dispersion of windborne material. Met. Mag., Lond., 90, 33-49. Payne, N. J., Helson, B. V., Sundaram, K. M. S. & Fleming, R. A. (1988). Estimating buffer zone widths for pesticide applications. Pestic. ScL, 24, 147-61. Quantick, H. R. (1985). Aviation in Crop Protection, Pollution and Insect Control. Collins, London.
Buffer zone widthsfor honeybees
259
Riley, C. M., Wiesner, C. J: & Ernst, W. R. (1989). Off-target deposition and drift of aerially applied agricultural sprays. Pestic. Sci., 26, 159-66. Scopes, N. & Stables, L. (eds) (1989). Pest andDisease Control Handbook. 3rd edn, BCPC, London. Shires, S. W. & Bennett, D. (1985). Contamination and effects in freshwater ditches resulting from an aerial application ofcypermethrin. Ecotoxicol. Environ. Safe., 9, 145-58. Sinha, S. N., Lakhani, K. H. & Davis, B. N. K. (1990). Studies on the toxicity of insecticidal drift to the first instar larvae of the large white butterfly Pieris brassicae L. (Lepidoptera: Pieridae). Ann. Appl. Biol. 116 (in press). Smart, L. E. & Stevenson, J. H. (1982). Laboratory estimation of toxicity of pyrethroid insecticides to honeybees: relevance to hazard in the field. Bee World, 63, 150-2. Stevenson, J. H. (1978). The acute toxicity of unformulated pesticides to worker honeybees (Apis mellifera L.). Plant Path., 27, 38--40. Ware, G. W., Cahill, W. P., Gerhardt, P. D. & Witt, J. M. (1970). Pesticide drift. IV. On-target deposits from aerial applications of insecticides. J. Econ. Ent., 63, 1982-3. Ware, G. W., Estesen, B. J., Cahill, W. P. & Frost, K. R. (1972). Pesticide drift. V. Vertical drift from aerial spray applications. J. Econ. Ent., 65, 590-2. Ware, G. W., Buck, N. A. & Estesen, B. J. (1984). Deposit and drift losses from aerial ultra-low-volume and emulsion sprays in Arizona. J. Econ. Ent., 77, 298-303. Williams, C. T., Davis, B. N. K., Marrs, R. H. & Osborn, D. (1987). Impact of Pesticide Drift. NERC contract report to NCC, Institute of Terrestrial Ecology, Huntingdon. Worthing, C. R. (ed.) (1983). The Pesticide Manual, 7th edn. BCPC, Croydon. Yates, W. E., Akesson, N. B. & Coutts, H. H. (1967). Drift hazards related to ultralow-volume and diluted sprays applied by agricultural aircraft. Trans. Am. Soc. Agric. Engrs., 10, 628-32. Yates, W. E., Akesson, N. B. & Cowden, R. E. (1974). Criteria for minimizing drift residues on crops downwind from aerial sprays. Trans. Am. Soc. Agric. Engrs., 17, 627-32. Yates, W. E., Akesson, N. B. & Bayer, D. E. (1978). Drift of glyphosate sprays applied with aerial and ground equipment. Weed Sci., 26, 597-604. Yates, W. E., Akesson, N. B. & Brazelton, R. W. (1981). Systems for the safe use of pesticides. Outl. Agric., 10, 321-6.