SERG International

SERG International

A partnership in forest pest management research


Terms of

Field Testing


Graduate Student
Award Application



SERG International Guidelines for Spray Characterization and Product Efficacy Testing

This document provides guidelines for spray characterization and field testing of aerially applied pest control products intended for use in forest management. Where possible, SERG International has adopted international standards. For SERG International product trials, PMRA Guidelines (PMRA DIR2003-04) should be followed.

1. Wind Tunnel Characterization of Atomizer Sprays.

Purpose: To characterize drop size distributions from atomizers being used in deposit characterization and product efficacy studies.

Wind tunnel tests should be guided by standards ASME799, ASME1296 and ASME1458. Atomizer/ tank mix combinations should be wind tunnel tested in order to establish the drop-size distribution that will be used during field trials. Wind tunnel conditions should reflect operational conditions, namely: aircraft speed, atomizer type, nozzle angle or cage RPM, flow range and tank mix. When dilute tank mixes are evaluated, atomizer characterization should also include undiluted formulations (if used operationally) for comparison. Protocols should be documented based upon accepted standards (ASTM being drafted, SDTF SOP).

2. Calibration and Deposit Pattern Testing of Aerial Application Equipment

Purpose: To determine the swath for atomizer/aircraft configuration.

ASAE S386.2 DEC98 standard should be followed when calibrating and establishing a deposit pattern for spray equipment. For into-wind trials using atomizers producing sub-150μm VMD spray distributions, spray lines should be extended upwind at least 20x the application height (i.e. for a spray height of 20m, the spray line should extend at least 400m upwind of the crosswind sampling grid) in order to ensure sufficient distance is provided for small drops to reach sampling surfaces. For crosswind trials, the sample line should be extended to distances that allow for deposit offset from the spray line and extended downwind deposit due to drift. Sampling distances can be estimated using the AGDISP model to characterize downwind deposit for different aircraft heights, crosswind speeds and drop size distributions. For crosswind sprays, the spray line should be centered on the sample line and at least twice the sample-line length.

3. Product Efficacy Testing

  1. To demonstrate or estimate the operational efficacy of a single product
  2. To compare the operational efficacy of several different products, formulations or application rates
  3. To establish a relationship between the amount of product applied and efficacy

i) Definition of efficacy

The meaning of efficacy may vary from one trial to another depending on the objective of the pesticide application. The two most common expressions of efficacy in forestry are:

1. Reduction of the pest population This may be measured directly from pre- and postspray population estimation, or indirectly through surrogate measurements.

2. Reduction in damage (e.g. defoliation, cone destruction) or surrogate measurements (e.g. tree color, whole-tree visual assessment)

Efficacy should be explicitly defined in the final SERG International report.


ii) Measurement of efficacy

There are two main issues that must be addressed when measuring efficacy:

1. Timing of the post-treatment sample. Often, efficacy is measured as soon as possible after treatment to prevent unexpected sources of mortality (unrelated to pesticide deposit) from confounding or mitigating spray results. However, this is not necessarily good practice. In general, product toxicity is established well before field trials are attempted and the question addressed by field trials is usually not whether the pesticide can reduce populations and thus protect trees, but how much and how fast under field conditions. In general, efficacy measurements should show whether or not the application has an overall effect

  1. at the end of the insect's current generation when the objective is population reduction or
  2. at the end of the current damage-causing period when the objective is damage reduction.

2. Surrogate measurements are by definition an indirect measurement. Care must be taken to ensure that the surrogate measurement is closely related to the efficacy variable that it is intended to replace. For example, frass drop is related to insect
density, but also to feeding activity, the age and development rate of the population. However, feeding rate can be influenced by larval stage, weather conditions, food quality and sub-lethal effects of some pesticides (e.g. feeding inhibition in SBW). The researcher has the responsibility of unequivocally relating such surrogates to efficacy, and experimental protocols should address this issue at the onset whenever a surrogate measure is proposed.

Field documentation should include:

1. Product potency (toxicity to the target pest species).

1.1 Source (manufacturer, batch)
1.2 Age (storage conditions, shelf life)
1.3 Quantification of active ingredient

2. Treatment Program.
2.1 Aircraft

  • Type (Model, registration)
  • Speed
  • Height above the canopy
  • Flight line spacing (track)
  • Flight line direction
  • GPS flight record

2.2 Atomization

  • Atomizer setting (type, description, number, position and orientation with respect to the wings, flow rate, spinning rate)
  • Formulation physics (viscosity, volatile fraction)

2.3 Spray weather

  • Wind (speed, direction)
  • Relative Humidity
  • Temperature
  • Pre- and post-spray rain

2.4 Target condition

  • Topography
  • Map of block (shape, area, location)
  • Stand conditions (density, height, defoliation, species composition)
  • Foliage development (leaf/needle spread)

2.5 Deposit

  • GPS positions relative to spray lines
  • Deposit quantification at efficacy sites

3. Dose acquisition and expression.
3.1 Tree condition
3.2 Insect population susceptibility

  • Genetics
  • Development

3.3 Pre-spray population level

3.4 Post-spray conditions

  • Natural mortality (compensation)
  • Weather (temperatures, rain)
SERG International field trials should be properly documented. Application variables (aircraft, atomization, weather and target condition) that influence deposit should be logged during product trials. GPS should be used to document both flight lines and sampling locations. Sufficient replicates should be completed to statistically justify conclusions reached.

4. Experimental designs

i) Gradient Design

To obtain a dose-response rather than a yes/no answer to the efficacy question, it may be possible to use a design where dose (application rate or actual deposit) is varied systematically across blocks. The analysis then becomes one of covariance between efficacy and a measure of deposit. This does NOT alleviate completely the need for replication. However, it does reduce considerably the number of blocks needed for a trial.

There may be several methods to obtaining significant deposit gradients in the field, however, one such method is the overlaying of flight lines at the upwind edge of blocks. An adequate number of distinct samples (where efficacy and deposit are measured simultaneously) can then be taken along transverse lines that start well upwind of the flight line (to provide zero-deposit measurements) and extend downwind to the lowdeposit side. A reliable deposit assessment method must be available for this approach to be useful. As much as possible, error in the deposit measurement should be kept low relative to error in efficacy.

This type of design offers several key advantages. First, the need for replication is reduced. Secondly, controls are not distinct blocks: they are inherent in the upwind deposit segment of the traverse line. Thus, it is important to design the applications and sampling so that no-deposit areas are sampled and high and low deposits are obtained.

In the deposit-gradient trial, the question being addressed is: what dosage (deposit) level is "efficacious". Implicitly, if such a dosage is found to exist at the end of the trial, efficacy has been demonstrated. The added bonus is the ability to recommend a deposit rate and perhaps an application rate (given an adequately documented relationship between application and deposit).

ii) Block Design - Deposit optimization considerations

Efficacy trials are often carried out on blocks ranging in size from 35 to 50ha resulting in block widths ranging from 300-500m. Within these blocks, deposit needs to reflect levels that would be characteristic of operational treatments to larger areas. Downwind deposit from a single line reflects application characteristics and the impact of local meteorology on the released pesticide. For Instance, with increased release height, increased winds andreduced emitted droplet size (Vmd), the peak deposit will occur at increasing distances (deposit offset) downwind of the flight line. Reduction in peak deposit is closely linked with increased deposit offset. Therefore, for small blocks, application techniques need to be adopted that achieve deposits that minimize this effect and are representative of largerblock operational programs.

An example of design considerations is given in Figure 1. Forestry insecticide treatments generally utilize relatively small droplet size distributions (Vmd = 50-100μm) that reflect

Figure 1

Figure 1. Example of predicted deposit using AGDISP.

the need to maximize drop density on foliage at levels that provide a lethal dose to the target pest. A single treatment line applied in a moderate crosswind along the upwind edge of a block can produce deposits that are substantially less than application rates (Fig 1). Model predictions (AGDISP) suggest a maximum deposit that is less than 15% of application rate can result from a single-line aerial treatment when the emitted drop-size distribution is near 80μm, release height is 15m above canopy and winds are 10kph at the aircraft height. Peak deposit occurs nearly 50m (2 swath offset) downwind of the flight line. Under these conditions, an operational treatment would achieve deposit levels nearing application rate only after several lines (downwind deposit is 70% of application rate after 11 swaths) have been applied (i.e. significant distance (250m) into the treatment block).

Without optimization strategies, deposit within research blocks of 35-50ha receiving a normal treatment would span a significant range due to the nature of the deposit profileon the upwind side of the block. However, optimization techniques involving flight line offset and multiple passes on upwind lines can provide a more uniform deposit across the sampling grid. In this example, shifting treatment lines by two swaths and making three passes on the upwind line would provide more uniform deposit within the research block. While a standard 10-line treatment would result in maximum deposit downwind of the block, optimization techniques would shift peak deposit to lie within block boundaries, more closely reflecting average deposit within larger operational blocks. Despite multiple passes on a single line, predicted deposit does not exceed label rate. Deposit on the downwind side can be modified with the addition of extra treatment lines. Alternately, the downwind portion of the block can be removed from the sampling regime.

Table 1 gives treatment considerations for a range of wind speeds at aircraft height. Table results reflect an aircraft (Cessna 188) height that is 15m above canopy using a track spacing (swath) that is 23m. Emission VMD is 80μm. Two design strategies for efficacy trials are considered: Gradient design (multi-pass on a single line) and Block design. Block width is 500m. For the Gradient design, number of passes to achieve a deposit that is equivalent to 70% of application rate and downwind peak-deposit location are tabled. Distance from the peak deposit location to the location where deposit has fallen to 10% of peak deposit is also listed. For the Block design, two scenarios are considered (see Fig
1): Normal treatment and Optimized treatment. For the Normal treatment, swath number associated with deposit that is 70% of application is noted for different wind speeds. For the Optimized treatment, flight offset and multiple passes are calculated to produce the most uniform deposit pattern with highest average deposit within the target block. Flight line offset and multiple passes are tabled for varying wind conditions.

Gradient Spray Block Spray
Normal Optimized
# of passes Peak
(# Swaths)
(# Swaths)
# of
2 2 16 80 2.3 0 2
5 4 28 155 6.5 1 2
10 7 45 187 12 1 2
15 11 60 435 17.3 2 5

Table 1. Deposit characteristics for Gradient and Block spray trials.


Using AGDISP, a table of spray strategies for different wind speeds should be developed as a reference for proper experimental design to achieve deposit requirements. Where possible, treatment blocks should be expanded to ensure edge-effect deposit does not impact field results.


iii) Block separation requirements to minimize cross-block contamination.

Efficacy trials on a number of different blocks should strive to minimize cross-block contamination by ensuring block separation is adequate. In Table 2, block separations are tabulated for a Cessna 188 treatment of a 500m wide block. Spray height is 15m above canopy. Emission VMD is 80μm. Crosswinds at aircraft height range from 2 to 15kph. Distances from the downwind edge of the block have been calculated using
AGDISP. Off-target deposit is expressed as a fraction of the applied active. Allowing field trials in winds up to label maximum (16kph) requires block separations to distances greater than 800m if 5% contamination is acceptable. For trials employing multiple rate strategies, low rate combinations should be restricted to locations upwind of surrounding blocks.

Wind Speed at
Aircraft Height
Block Separation Requirements (m)
10%A 5% 1%
2 120 170 420
5 250 390 820
10 400 610 1350
15 510 780 >1500

A Deposit expressed as fraction of applied active.

Table 2. Example of small block separations in relation to possible 'contamination' from another up-wind block calculated using AGDISP with conditions given in the text.

In a multi-block experimental design, AGDISP should be used to ensure crossblock
contamination is minimized to an acceptable level. Potential cross-block
contamination should be reported.


Campbell, J. F. 2000. Experimental design: Statistical considerations and analysis. Pp.
39-76 in Field manual of techniques in invertebrate pathology, L. A. Lacey & H. K.
Kaya (Editors).

Hurlbert, S. H. 1984. Pseudoreplication and the design of ecological field experiments.
Ecological monographs 54: 187-211.

Mead, R. 1988. The design of experiments: statistical principles for practical
applications. Cambridge University Press, New York.

Milliken,G.A. and D.E. Johnson. 1984. Analysis of messy data, Volume 2:
Nonreplicated experiments. Van Nostrand Reinhold, New York.

Teske M, Bowers J, Rafferty J, Barry J 1993. FSCBG. An aerial spray dispersion model
for predicting the fate of released material behind aircraft. Environmental
Toxicology and Chemistry 12(3): 453-464.

Teske ME, Thistle HW 2004. Aerial application model extension into the far field.
Biosystems Engineering 89(1): 29-36.

Teske ME, Thistle HW, Ice GG 2003. Technical advances in modeling aerially applied
sprays. Transactions of the ASAE 46(4): 985-996.

Standards and Protocols

ASME 799 Practice for Determining Data Criteria and Processing for Liquid
Drop Size Analysis
ASME 1296 Standard Terminology Relating to Liquid Particle Statistics
ASME 1458 Test Method For Calibration Verification of Laser Diffraction
Particle Sizing Instruments Using Photomask Reticles
ASTM (Draft) Sampling sprays with laser diffraction instruments
SDTF-11A/5 Atomization Study Procedures - SDTF SOP
DIR2003-04 PMRA: Efficacy Guidelines for Plant Protection Products.