EPA-600/2-77-107g
July 1977
Environmental Protection Technology Series
SOURCE ASSESSMENT:
DEFOLIATION OF COTTON
State of the Art
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Researc h reports at the Office of Research and Development. U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned (o foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonsliale instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service. Springfield. Virginia 22161.
-------
EPA-600/2-77-107g
July 1977
SOURCE ASSESSMENT:
DEFOLIATION OF COTTON
State of the Art
by
J. A. Peters and T. R. Blackwood
Monsanto Research Corporation
1515 Nicholas Road
Dayton, Ohio 45407
Contract No. 68-02-1874
ROAP No. 21AXM-071
Program Element No. 1AB015
EPA Task Officer: David K. Oestreich
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
EPA-600/2-77-107g
July 1977
SOURCE ASSESSMENT:
DEFOLIATION OF COTTON
State of the Art
by
J. A. Peters and T. R. Blackwood
Monsanto Research Corporation
1515 Nicholas Road
Dayton, Ohio 45407
Contract No. 68-02-1874
ROAP No. 21AXM-071
Program Element No. 1AB015
EPA Task Officer: David K. Oestreich
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
PREFACE
The Industrial Environmental Research Laboratory (IERL) of
EPA has the responsibility for insuring that pollution con-
trol technology is available for stationary sources to meet
the requirements of the Clean Air Act, the Federal Water
Pollution Control Act and solid waste legislation. If con-
trol technology is unavailable, inadequate, uneconomical or
socially unacceptable, then financial support is provided
for the development of the needed control techniques for
industrial and extractive process industries. The Chemical
Processes Branch of the Industrial Processes Division of
IERL has the responsibility for investing tax dollars in
programs to develop control technology for a large number
(>500) of operations in the chemical industries.
Monsanto Research Corporation (MRC) has contracted with
EPA to investigate the environmental impact of various indus-
tries which represent sources of pollution in accordance with
EPA's responsibility as outlined above. Dr. Robert C. Binning
serves as MRC Program Manager in this overall program entitled,
"Source Assessment," which includes the investigation of sources
in each of four categories: combustion, organic materials,
inorganic materials, and open sources. Dr. Dale A. Denny of
the Industrial Processes Division at Research Triangle Park
serves as EPA Project Officer. Reports prepared in the Source
Assessment Program are of two types: Source Assessment
Documents and State-of-the-Art reports.
111
-------
Source Assessment Documents contain data on emissions from
specific industries. Such data are gathered from the litera-
ture, government agencies and cooperating companies. Sampling
and analysis are also performed by the contractor when the
available information does not adequately characterize the
source emissions. These documents contain all of the infor-
mation necessary for IERL to decide whether a need exists to
develop additional control technology for specific industries.
State-of-the-Art Reports include data on emissions from
specific industries which are also gathered from the litera-
ture, government agencies and cooperating companies. However,
no extensive sampling is conducted by the contractor for such
industries. Sources in this category are considered by EPA
to be of insufficient priority to warrant complete assessment
for control technology decision making. Therefore, results
from such studies are published as State-of-the-Art Reports
for potential utility by the government, industry, and others
having specific needs and interests.
This study was undertaken to provide information on air
emissions from the defoliation of cotton. In this project,
Mr. D,. K. Oestreich served as EPA Project Leader.
IV
-------
CONTENTS
Section
Preface
Figures
Tables
Symbols
I
II
III
IV
Introduction
Summary
Source Description
A. Characteristics of Cotton Defoliation
and Desiccation
B. Process Description
1. Ground Machinery
2. Aircraft
3. Nozzles
C. Factors Affecting Emissions
1. Spray Fluid Properties
2. Nozzles
3. Type and Operation of Equipment
4. Meteorological Conditions
D. Geographical Distribution
Emissions
A. Selected Pollutants
1. Folex and DBF
2. Sodium Chlorate
3. Arsenic Acid
4. Paraquat
B. Emission Factors
C. Definition of Representative Sources
D. Source Severity
1. Definition
2. Ground Level Concentration
3. Population Exposed
4. Total Air Emissions
Page
111
vii
viii
x
1
3
7
7
9
9
10
11
13
15
18
20
23
31
35
35
38
38
39
40
41
43
46
46
46
47
48
v
-------
CONTENTS (continued)
Section Page
V Control Technology 63
A. State of the Art 63
1. Fluid Additives 64
2. Nozzles and Atomizers 67
3. Equipment Modification 68
4. Meteorological Timing 69
B. Future Considerations 70
1. Foam Spray Systems 70
2. Microfoil® 73
3. Thermal Defoliation 75
VI Growth and Nature of the Industry 78
A. Present and Emerging Technology 78
B. Industry Production Trends 81
VII Appendixes 84
A. Derivation of Source Severity and 85
Input Data
B. Preliminary Air Sampling of Cotton 95
Desiccation
C. Method for Estimating TLV Values for 105
Compounds when None Exists
VIII Glossary of Terms 109
IX Conversion Factors and Metric Prefixes 112
X References 114
VI
-------
FIGURES
Figure Pag(
1 Hydraulic Pressure Nozzles 12
2 Drop Spectra for Four Nozzle Orientations 21
3 Cotton Harvested, 1969 32
4 Acreage Treated with Chemicals for 33
Defoliation or for Growth Control of Crops
or Thinning of Fruit, 1969
5 Component Parts of Foam Generators which 72
Mix Air and Liquid to Form Foam
6 Representation of the Microfoil 75
7 Schematic of 1970 Thermal Defoliator 76
8 U.S. Cotton Acreage, Yield, and Production 82
A-l Representative Field for Agricultural 89
Spraying
B-l Terminal Velocities of Spherical Particles 104
of Different Densities Settling in Air and
Water at 21°C under the Action of Gravity
VII
-------
TABLES
Table Page
1 Severity Factors and Population Exposed 6
to Pollutant for which S > 0.1
A
2 Horizontal Transport of Droplets in Light 16
Winds
3 Effect of Spray Pressure on Droplet Size 18
4 Droplet Size Comparison of Four Nozzle Types 19
5 Terminal Velocities of Particles in Air and 26
Number of Drops/Area
6 Time and Vertical Fall Distance for Pure 31
Water to Evaporate from D0 to Df at 25°C,
101.3 kPa
7 Cotton Acreage Harvested, Percent of U.S. 34
Total
8 Defoliants and Desiccants Used for Cotton 36
9 Formulation and Dilution of Major Harvest- 37
Aid Chemicals
10 Calculated Emission Factors from Published 42
Data for Drift from Agricultural Spraying
11 Calculated Emi&sion Factors from Preliminary 44
Field Sampling of Arsenic Acid Application
to Cotton
12 Emission Factors for Defoliation or 45
Desiccation of Cotton
13 Source Severity, Area, and Population 48
Exposed to Pollutants for Which x/F, ^_ 0.1
14a Cotton Acreage Harvested and Defoliated, 50
1971 (Metric units)
14b Cotton Acreage Harvested and Defoliated, 51
1971 (English units)
15 Quantities of Defoliants and Desiccants 52
(Active Ingredients) Used on Crops and
Acreage of Crops Treated, by Region, 1971
16 Cotton Acreage Harvested and Defoliated in 55
Texas, 1971
17 Agricultural Use of Defoliants and 60
Desiccants in Arizona, 1971
Vlll
-------
TABLES (continued)
Table Page
18 Emission Estimates for Cotton Defoliants 61
and Desiccants by State and Nationwide, 1971
19 Typical Drop Size Distribution, Cumulative 74
Percent by Volume Below Sizes Shown
20 U.S. Cotton Acreage, Yield, and Production, 82
1947-73
21 Changes in Use of Harvest-Aid Chemicals for 83
Cotton
B-l Arsenic Acid Spraying Data 99
B-2 Emission Rate Calculation Data 101
C-l Agricultural Chemicals with Published TLV's 106
IX
-------
SYMBOLS
Symbol Definition
a Constant
B Length of side of square field
B1 Intercept
b Exponent
C Effective transfer coefficient at instan-
taneous fall velocity, V
D Distance from center of representative
field to perimeter
d Drop diameter
Df Final diameter
D Dosage or concentration from line puff
(i.e., due to instantaneous line source)
D Initial diameter
o
e 2.72
F Hazard factor defined as the primary
ambient air quality standard for criteria
pollutants or a "corrected" threshold
limit value for noncriteria pollutants
(i.e., F = TLV • 8/24 • 1/100 for non-
criteria pollutants)
F Time-adjusted exposure factor related to
threshold limit value and including a
safety factor for general population
exposure (i.e., F = TLV • 1/100)
£\
h Effective emission height
K Diffusivity of water vapor in air at
ambient temperature
LD50 Acute oral dose for male rat
M Slope
N Number of drops of specific diameter, d
-------
SYMBOLS (continued)
Symbol Definition
n Number of spray passes or swaths made in
representative field
n1 Sample size (i.e., number of samples)
nmd Mean based on number and mass of drops
P Partial pressure of air
AP Vapor pressure gradient between surround-
ing air and droplet surface
Q Emission rate
Q_ Total amount of material emitted per
length from a line source
R2 Coefficient of correlation
S Source severity of emissions resulting
A
from agricultural field spraying
SD Standard error of B1
D
S., Standard error of M
M
S__v Standard error of estimate
x X
t Time to complete spraying a representa-
tive field including the time needed for
turning the spray equipment
T Time to spray one pass
TLV Threshold limit value
u Mean wind speed
vmd Mean based on volume (mass) median
diameter
V. Terminal velocity
X,x Arbitrary independent variable
x Distance from source
x. i independent variable
Time of exposure
*i *
xi
-------
SYMBOLS (continued)
Symbol Definition
X2 Temperature
Y,y Arbitrary dependent variable
a/2 . The a/2 percentage point of standard
normal distribution
z Roughness length; height at which air
° velocity near boundard layer reaches
zero
a 1.0 minus the confidence level
IT 3.14
a T Standard deviation of the distribution of
x pollutant material in the x direction in
a puff
a T Standard deviation of the distribution of
pollutant material in the z (vertical)
direction in a puff
X Ground level concentration of pollutant
emitted by a continuous elevated point
source
X^ Time-averaged ground level concentration
of pollutant at downwind perimeter of
representative field undergoing spraying
for defoliation or dessication
Q Emission rate
Xmav Time-averaged maximum ground level con-
m=iv centration of a pollutant emitted from a
continuous nearpoint source
Maximum ground level concentration of
pollutant emitted by a continuous ele-
vated point source
xn
-------
SECTION I
INTRODUCTION
Defoliation of cotton encompasses both defoliation and
desiccation as chemical harvest-aid practices which are used
to prepare the cotton crop for mechanical harvesting machines,
Because harvest-aid chemicals are sprayed as fine droplets
on the cotton, this practice constitutes a source of air
pollution in the form of fugitive aerosols. The objective
of this work was to assess the environmental impact of defo-
liation of cotton and to produce a reliable and timely Source
Assessment Document for use by the EPA in deciding on the
need for the development of additional control technology.
This document summarizes information relating to the emis-
sions from defoliation of cotton. The areas studied were:
(1) characteristics of emissions and factors affecting
emissions; (2) source sites; (3) state and nationwide mass
emissions; (4) effects of emissions on air quality and hazard
potential to local population; (5) current and future consid-
erations in pollution control technology; and (6) projected
growth and anticipated technological development of the
industry.
Emission factors were developed by preliminary field sampling
of one of the major harvest-aid chemicals (arsenic acid) dur-
ing application by a method that is characteristic of the
industry (ground rig application). Emission factors for the
major chemicals and application methods were assumed, based
1
-------
on analogy to data found in the literature. These emission
factors; were used to compile the estimated effects on air
quality. More complete and reliable data could be obtained
by further sampling and analysis of the following agricul-
tural practices: (1) aerial tributylphosphorotrithioate
(DBF) application; (2) aerial sodium chlorate application;
and (3) aerial paraquat application.
-------
SECTION II
SUMMARY
Cotton is defoliated or desiccated prior to harvest wherever
it is grown in the U.S. Defoliation is defined as the
process by which leaves are abscissed from the plant by the
action of topically applied chemical agents. Desiccation
by chemicals is the drying or rapid killing of the leaf
blades and petioles with the leaves remaining in a withered
state on the plant. Defoliants are used on the taller vari-
eties of cotton which are machine picked for lint and seed
cotton, while desiccants usually are used on short, storm-
proof cotton varieties of lower yield that are harvested by
mechanical stripper equipment.
The major cotton producing regions are located in the <
Mississippi River Valley extending from the top of Louisiana ^';
to the bootheel of Missouri, the Blacklands region of Texas,
and the High and Low Rolling Plains regions of Texas. The
top three cotton producing states in 1972, which together
contributed over 60% of the harvested acreage, were Texas
(39.4%), Mississippi (12.3%), and Arkansas (10.7%). A total
of 16 states were considered in this study out of a possible
18. The two excluded states comprised less than 0.5% of
annual harvested acreage.
Currently, almost 50% of total cotton acreage harvested is
pretreated with defoliants or desiccants, ranging from a low
-------
of 3% of New Mexico's acreage to a high of 85% of California's
acreage.
Cotton defoliants and desiccants are applied as water-based
sprays either by aircraft or by a ground machine. In both
cases, nozzles situated on a boom break up the liquid formu-
lations into spray droplets. The likelihood of spray droplets
drifting into the atmosphere from their point of emission is
primarily a function of the droplet diameters; in order to
be emitted to the atmosphere rather than being deposited on
target, the critical diameter of droplets has been proposed
to be of. the order of 100 pm. The small droplets drift.
The major defoliant chemicals used are sodium chlorate,
tributylphosphorotrithioite (Folex), and tributylphosphoro-
trithioate (DEF). The major desiccants are arsenic acid and
paraquat. The U.S. emissions of cotton defoliants in 1971
were 22.9 metric tons (25.2 tons) of DEF and Folex, and
33.0 metric tons (36.3 tons) of sodium chlorate. Total
emissions of cotton desiccants were 16.8 metric tons
(18.5 tons) of arsenic acid, and 1.39 metric tons (1.53 tons)
of paraquat. (All emission rates were based on estimated
usage figures and on some assumed emission factors.)
s These emissions occur from July to October, preceding by two
weeks the period of harvest in each cotton producing region.
The emission factors for each major harvest-aid chemical are
assumed to be 10 g/kg (20 Ib/ton) for sodium chlorate, DEF,
Folex, and paraquat, and 6.1 ± 2.9 g/kg (12.2 ± 5.7 Ib/ton)
at the 95% confidence level for arsenic acid.
The source severity, S , was defined to indicate the hazard
f\
potential of a representative emission source for the special
case of agricultural field spraying:
-------
where x is the time-averaged ground level concentration of
the chemical emitted at the downwind perimeter of a repre-
sentative field undergoing spraying for defoliation or
desiccation, and F is a time-adjusted exposure factor
£\
related to threshold limit value (TLV®) and also includes
a safety factor for general population exposure.
Four representative sources of harvest-aid chemical spray
application were defined. For sodium chlorate, the repre-
sentative source was a 0.70-km2 (173-acre) cotton farm
located in the Mississippi River Valley with an aerial appli-
cation rate of 0.56 g/m2 (5.0 Ib/acre) . The representative
source for DBF application was defined as a 0.70-km2 cotton
farm located in the Mississippi River Valley with an aerial
application rate of 0.17 g/m2 (1.5 Ib/acre) . The arsenic
acid representative source consisted of a 0.61-km2 (150-acre)
cotton farm located in the Blacklands of Texas with a ground
machine application rate of 0.49 g/m2 (4.4 Ib/acre). The
representative source for paraquat application was defined
as a 1.05-km2 (260-acre) cotton farm located in the High
Plains of Texas with an aerial application rate of 0.056 g/m2
(0.5 Ib/acre) .
The calculated source severity factors for the representative
sources of each of the major harvest-aid chemicals are given
in Table 1, accompanied by the affected population to a
severity of 0.1 or greater.
Control technology for aerial application of pesticides has
been implemented in the practice of techniques that are
effective in reducing chemical drift. Fluid additives that
increase the viscosity of the spray formulation and thus
-------
Table 1. SEVERITY FACTORS AND POPULATION EXPOSED
TO POLLUTANTS FOR WHICH SA >_ 0.1
Pollutant
Arsenic acid
Paraquat
Sodium chlorate
DEF
Representative
source
severity
0.69 ± 0.32
0.30
0.44
0.67
Exposed
population,
persons
6,134
322
754
2,517
decrease the number of fine (<100 ym) droplets have been
used. Nozzle design and orientation control the droplet
size spectrum. Future control technology considerations
include the use of foam spray systems to reduce overlapping,
multiple hypodermic needle nozzle systems, and the replace-
ment of chemical defoliation with thermal defoliation.
The cotton industry has been growing (7.5% per year) since
1967 when acreage harvested hit a modern day low point.
However, the growth trend is leveling off, and 1978 cotton
acreage is anticipated to be no more than that of 1972 due
to strong competition from foreign producers and from syn-
thetics. The growth factor for the industry (1978 emissions/
1972 emissions) is 1.
-------
SECTION III
SOURCE DESCRIPTION
A. CHARACTERISTICS OF COTTON DEFOLIATION AND DESICCATION
Artificial defoliation of cotton was first discovered by
researchers at the Pee Dee Agricultural Experiment Station
in South Carolina. In being applied to cotton as a side
dressing, some calcium cyanamide fertilizer accidently
drifted onto the cotton which was wet with dew, and it
caused the leaves to drop off. At harvest time some of the
fertilizer was purposely dusted on other cotton; it was
defoliated, also. Although it had always been known that
cotton sheds its leaves just after a frost, the Pee Dee
discovery marked the beginning of artificial defoliation.
By 1945, the increasing labor shortage and high cost of
conventional hand picking of cotton led to the introduction
of mechanical harvesters. Although efficient, these machines
collected bolls and foliage together, so that the lint was
stained with the sap from damaged leaves. A chemical that
would either destroy the leaves or cause premature leaf fall
but maintain the bolls unharmed had an obvious economic
value. Although in 1941 only a few fields of cotton were
defoliated experimentally, now about half the cotton in the
U.S. is sprayed with defoliants.1
^sborne, D. J. Defoliation and Defoliants. Nature.
219:564-567, August 10, 1968.
-------
Defoliation may be defined as the process by which leaves are
abscised from the plant. The process may be initiated by
drouth stress, low temperatures, or disease, or it may be
chemically induced by topically applied agents or by over-
fertilization. The chemicals used to initiate the process
are termed defoliants. The practice of desiccating cotton
plants with chemicals is often mistakenly called "defoli-
ation. ";>-
Desiccation, the drying or removal of moisture, is a term
used to describe the effect of harvest-aid chemicals on
cotton plants, which involves rapid killing of the leaf
blades and. petioles. The severe chemical injury also pre-
vents the formation of an abscission layer, and the leaves
do not detach from the stalks. The term "frozen" is
commonly used to describe the leaf condition. When the
blades of leaves are killed by chemical action with appre-
ciable injury to the petioles, leaf abscission does occur.
Under certain conditions, the dry leaf blades of frozen,
dead leaves are removed from the petioles by wind-induced
thrashing of the plant stem, giving the field a defoliated
appearance.2
Defoliation is especially advantageous in machine harvesting
and is used mostly where spindle picker harvesters are used.
Defoliation helps lodged plants to return to an erect
position; removes the leaves which can clog the spindles of
the picking machine, add trash, and stain the fiber;
accelerates the opening of mature bolls; and reduces boll
2Miller, C. S., E. D. Cook, J. L. Hubbard, J. S. Newman,
E. L. Thaxton, and L. H. Wilkes. Cotton Desiccation
Practices and Experimental Results in Texas. Texas
Agricultural Experiment Station. College Station, Texas.
Miscellaneous Publication No. MP-903. November 1968. 14 p.
8
-------
rot. Defoliation reduces populations of insects which feed
on leaves in late season; it has the immediate effect of
eliminating fiber damage by the honeydew of aphids and white
flies, and the more important long-range effect of greatly
reducing the number of overwintering insects, such as the
pink bollworm.3
Desiccants usually are used on stormproof or semistormproof
cotton varieties that are small and low in growth. Cotton
desiccation is standard practice in the high plains of Texas
and Oklahoma, where yields are relatively low and production
costs must be kept low. Most harvesting in these regions is
done by mechanical "strippers" - a much less expensive method
than spindle machine harvesting. At harvest time there is
usually very little moisture in the cotton and it responds
poorly to defoliants, green leaves often remaining on the
plants; farmers therefore prefer to desiccate and have
thoroughly dry leaves, since strippers remove leaves and
burrs with the seed cotton.3
B. PROCESS DESCRIPTION
Harvest-aid chemicals are applied to cotton as water-based
sprays either by aircraft or by a ground machine.
1. Ground Machinery
A complete sprayer unit is equipped with a power source
(engine or power take-off), pump, pressure gauge, pressure
regulator, tank, booms, pressure hoses, and nozzles. The
sprayer unit may be self propelled, tractor mounted, or pull
3Addicott, F. T., and R. S. Lynch. Defoliation and
Desiccation: Harvest-Aid Practices. In: Advances in
Agronomy, Vol. 9. 1957. p. 69-93.
-------
type. The pump, which is operated by the power source,
pumps the chemical formulation from the tank through the
s
pressure regulator, then through the hoses and boom and out
through the nozzles. Typically, a sprayer pump will force
about four times as much spray through the pressure regulator
as is discharged through the nozzles. The excess spray is
forced through a bypass line and discharged back into the
sprayer tank. This agitates the spray mixture, keeping it
well mixedo4
Three factors determine the amount of liquid a sprayer can
apply. These are: (1) ground speed of the sprayer unit,
typically from 1.3 m/s to 6.7 m/s (3 to 15 mph); (2) size and
number of nozzles used, usually one to five nozzles per crop
row; and (3) pressure at which the spray is applied, typically
140 kPa to 620 kPa (20 to 90 psi).
The mos-c popular type of ground sprayer used on cotton is the
High Clearance Tractor Sprayer, or Hi-Boy, a sprayer mounted
on an elevated tractor with wheel shields for crop protection.
2. Aircraft
Sprayer units mounted on aircraft are comprised of the same
elements as ground rig sprayers. Sprays are pumped out
through a wing-length boom on which hydraulic atomizing
nozzles are located. The power source for the pumps is
either an additional small engine on board or, more typi-
cally, a centrifugal pump that is wind driven by a small
propeller located beneath the aircraft. There are from
28 to 56 nozzles located on the boom, operating at 210 kPa
^Insecticidal Spraying of Field Crops With Ground Machinery,
Texas Agricultural Extension Service. College Station,
Texas. Bulletin No. L-486. August 1961.
10
-------
to 410 kPa (30 to 60 psi). Airplanes spray at speeds of
36 m/s to 54 m/s (80 to 120 mph), while helicopters maintain
speeds of about 9 m/s (20 mph). Swath widths of 10 m to
20 m are typical.
The release patterns from a fixed-wing plane and from a
helicopter are similar. Chemicals are released downward,
dispersed outward, drawn upward near the wing tip or rotor
tip, rotated in this zone, and then settle to the ground.
The vortex system — the rotation at the wing tip or rotor
tip — is a basic function of both types of equipment. A
strong central propeller wash that develops with fixed-wing
aircraft has the undesirable effect of skewing the wake to
one side of the aircraft's centerline. The helicopter
pattern is generally better than the fixed-wing aircraft
because this skewing is not a factor.5
3. Nozzles
Regardless of the rates and dosages used by aircraft and
ground equipment, both types of operations use essentially
the same techniques and devices for breaking up a liquid
formulation into a spray. The most frequently used devices
are the hydraulic pressure nozzles illustrated and identified
in Figure 1 according to the type of droplet pattern that
each produces:6
5Riley, J. A., and W. L. Giles. Agricultural Meteorology in
Relation to the Use of Pesticides. Agricultural Meteorology,
2_:225-245, 1965.
6Akesson, N. B., and W. E. Burgoyne. Spray Atomization,
Application Volume and Coverage. Proceedings and Papers
of the Thirty-Fifth Annual Conference of the California
Mosquito Control Association, Inc., and the American
Mosquito Control Association. February 1967. p. 139-144.
11
-------
B
CYLINDRICAL CONE HOLLOW CONE
CD
HOLLOW CONE FULL CONE
E
aAT FAN
ROOD ING PATTERN
Figure 1. Hydraulic pressure nozzles6
A-oylindrical jet, formed by liquid ejected from
a small circular orifice;
B-hollow cone, created by a small whirl plate ahead
of the orifice which gives the spray a tangential
spin, thus spreading and breaking up the liquid;
C-c,nother form of the hollow cone in which the
tangential spin is produced by an offset entrance
to the whirl chamber (frequently described as
"ncnclog nozzles");
D-full, or solid cone produced because a small hole
has been drilled in the center of the whirl plate
to fill in the normal hollow cone;
E-flat fan, wherein proper milling of the orifice
slot gives a long, narrow pattern; and
F-flooding pattern, formed by simple impaction of
the liquid against a sloping plane.
Each of these designs has been used for pesticide applica-
tions arid is adapted to a particular service primarily on
the basi.-s of the coarse or fine spray it produces. The
hollow cone (B) and flat fan (E) are the most commonly used
of the group. The greatest flexibility is available in the
hollow cone where different combinations of whirl plate and
disc orifice size can provide a wide range of spray particle
sizes.6
12
-------
C. FACTORS AFFECTING EMISSIONS
Contamination of the air from harvest-aid chemical applica-
tion can come about in any of three ways: (1) aerosol spray
suspended during application; (2) subsequent wind erosion of
contaminated soil; or (3) vaporization of the chemical from
the treated area. Wind erosion and vaporization are not
treated in this assessment; however, primary chemical spray
drift from a source or application site is investigated.
In analyzing treated plants and soil, researchers have long
been plagued by the fact that they can usually account for
only a fraction of the amount of a pesticide applied. It is
not unusual to find only 50% or less of the applied material
accounted for in the materials balance in the treated area
immediately after application. Most of the missing part is
dispersed in the air as fine sprays, or aerosols, and carried
to adjacent areas.
There are three zones of spray drift contamination. The
first, the target area where deposit takes place primarily
by ballistic fallout, includes the actual aircraft or ground-
rig swaths and the area about 70 meters downwind. The second
zone is the drift fallout zone, extending from about 70 meters
to over a kilometer downwind. This zone may receive some
fallout, but within 300 meters most of the material is air-
borne (aerosol-size droplets under 50 ym),7 and meteorological
factors dominate the deposit of residues. The general en-
vironmental area is the third zone. It continues from a
kilometer or so onward, and becomes the sink for material
7Yates, W. E., and N. B. Akesson. Reducing Pesticide
Chemical Drift. In: Pesticide Formulations. Van Valkenburg,
Wade, (ed.). New York, Marcel Dekker, Inc., 1973.
p. 275-341.
13
-------
transported in the atmosphere as very fine aerosol particles
of less than 10 ym to 15 ym. These may be deposited by
settling and impingement, but may also be carried aloft and
not returned to earth except by washout from precipitation
of some form.7
In the drift fallout zone, data have shown8 that the amount
of material still in the air is from 6 to 40 times that
which falls on the ground. This would likely be true at
further distances downwind; however, continued settling
would be expected and spreading by air diffusion would
rapidly reduce the air concentration. Other published data9
from controlled pesticide aerosol wave release experiments
revealed that the aerosol droplets that settle at distances
of over 1 km from the point of aerosol generation did not
exceed i>% for the large drops (>25 ym) , but reached about
90% for aerosol smaller than 15 ym.
The likeilihood of droplets drifting into the atmosphere from
their point of emission is primarily a function of the
droplet diameter or size. The critical diameter of droplets
for agricultural spraying has been proposed to be of the
order of 100 ym;10'11 all droplets smaller than that are apt
8Akesson, N. B., and W. E. Yates. Problems Relating to
Application of Agricultural Chemicals and Resulting Drift
Residues. Annual Review of Entomology. 9^:285-318, 1964.
9Kutsenogiy, K. P., V. I. Makarov, Y. P. Chankin, V. M.
Sakharov, and G. N. Zagulyayev. Study of the Physicochemical
Characteristics of Large Aerosol Waves. Institut
Eksperimental'nye Meteorologiya. 2_7:97-104, 1972.
10Maybank, J., and K. Yoshida. Delineation of Herbicide
Drift Hazards on the Canadian Prairies. Transactions of
the American Society of Agricultural Engineers.
12^759-762, 1969.
^Courshee, R. J. Investigations on Spray Drift. II. The
Occurrence of Drift. Journal of Agricultural Engineering
Research. 4_:229-241, 1959.
14
-------
to drift away from the target area. To investigate this
problem, it is necessary to specify and study the overall
size spectrum produced by a typical sprayer.
Drop size and frequency distribution information may be
presented in many ways. The simplest of these is the arith-
metic mean, or ENd/EN, where N is the number of drops having
a diameter d. This gives an arithmetic average which tends
to be weighted in favor of small drops. To weight the mean
drop diameter on the basis of volume, another mean can be
used, (ENd3/EN)*/3. More expressive means, those based on
number and mass of the drops, are termed number (nmd) and
volume (mass) median diameters (vmd). These medians are
defined as the values that divide the numbers or volumes of
the spray into two equal parts, or the 50% cumulative point.
The vmd is most commonly used and is often refered to as mass
median diameter (constant density of droplets). Still another
mean is used, particularly by fuel burner investigators, called
the Sauter mean diameter. It is an expression of volume-to-
surface relation, ENd3/ENd2.8
The many factors that influence the droplet size spectrum
formed and the movement of droplets discharged from an air-
craft or ground machine are discussed below. A rough idea
of the drift pattern of various droplet sizes can be obtained
from Table 2.
1. Spray Fluid Properties
Most agricultural spray nozzles produce a wide range of spray
drop sizes. In addition, selection of the fluid properties
can affect the drop size spectrum. The most important
physical properties related to droplet size are the surface
tension, viscosity, density, and vapor pressure.
15
-------
Table 2. HORIZONTAL TRANSPORT OF DROPLETS IN LIGHT WINDS8
Drop
diameter,
Drop type
Distance droplet would
be carried by a 1.34 m/s
(3 mph) wind while
falling 3 m (10 ft), m
400
150
100
50
20
10
Coarse aircraft spray
Medium aircraft spray
Fine aircraft spray
Air carrier spray
Fine spray and dusts
Usual dusts and
aerosols
Aerosols
2.59 (8.5 ft)
6.71 (22 ft)
14.63 (48 ft)
52.25 (178 ft)
338.3 (0.21 mi)
1,352 (0.84 mi)
33,800
(21 mi)
a. Surface Tension - The surface tension represents a
direct force that resists the formation of a new surface
area. The minimum energy required for atomization is equal
to the s'urface tension multiplied by the increased liquid
surface area. Thus, it may represent a predominant force
for certain types of atomization. The surface tension
commonly encountered in sprays ranges from 0.073 N/m
(73 dynes/cm) for water to as low as 0.020 N/m (20 dynes/cm)
for some petroleum distillates.
For most pure liquids the surface tension in contact with
air decreases with an increase in temperature and is inde-
pendent of the age of the surface. Since most agricultural
sprays are mixtures of surfactants, carrier, and active
ingredient, it must be noted that the surface tension of a
newly formed surface is close to the value for the bulk of
the liquid and with time reaches an equilibrium or static
surface tension that is normally reported as "surface
tension.1]
16
-------
The term "dynamic surface tension" is the value obtained
before equilibrium and is related to the age of the surface.
The dynamic surface tension for the age of the surface at
the time of disintegration should be used for prediction
of drop size. A decrease in the dynamic surface tension
increases the number of droplets available for drift.
b. Viscosity - Viscosity is one of the most important
liquid properties that can affect the drop size spectrum.
An increase in viscosity physically dampens the natural wave
formations. This generally delays disintegration and
increases droplet size. The viscosity of most spray solutions
is relatively low, ranging from 0.001 Pa-s (0.01 poise) for
water to 0.01 Pa-s (0.1 poise) for some weed oils. The
viscosity of simple (or Newtonian) liquids is independent of
the shear rate and generally decreases with an increase in
temperature. However, for a complex (or non-Newtonian) fluid,
and most spray formulations are complex, the viscosity is a
function of the shear rate. It is this particular parameter
to which many developments in drift control are addressed
(see Section V).7
c. Density - The spray formulation density has little
effect on the atomization due to the small range that is
normally encountered in commercial spray formulations. The
density can range from a low of 800 kg/m3 (0.8 g/ml) for an
oil carrier to 1200 kg/m3 (1.2 g/ml) for some technical
materials, but the bulk of agricultural formulations use
water, with a density of 1000 kg/m3 (1.0 g/ml).7
d. Vapor Pressure - For most agricultural spray systems
the vapor pressure has no effect on the initial droplet size
spectrum.7 However, the vapor pressure gradient between the
surrounding air and the drop surface has a direct effect
upon the rate of evaporation which consequently determines
17
-------
the size of a given drop with respect to time. The effect of
evaporation is discussed later in this section (Ill.C.4.d).
2. Nozzles
One of the most important means for controlling droplet size
is through selection of the type, design, operating pressure,
and orientation of the atomizer or nozzle.
a. Spray Pressure - Spray pressure controls the speed at
which the ejected liquid moves through the air. An increase
in pressure increases the speed and forms larger numbers of
small drops. Typical effects are shown in Table 3. Several
factors including spray pressure affect the momentum of the
airstream which accompanies the spray and tends to carry it
to the ground.11
Table 3. EFFECT OF SPRAY PRESSURE ON DROPLET SIZE11
Spray pressure,
kPa (psi)
68.9 (10)
110.3 (16)
420.6 (61)
Percentage of spray
volume that is <100 ym
Nozzle A
3
6
18
Nozzle B
1
3
18
Nozzle A is a swirl nozzle with swirl ports cut
into the plate containing the orifice.
'Nozzle B is a flat fan nozzle molded from a
ceramic material.
b. Type of Nozzle - The type of nozzle or atomization
system can affect the droplet spectrum when all other factors
18
-------
are equal. Laboratory tests have shown12 the differences in
droplet size characteristics. Under controlled conditions,
the percentage of spray volume in the driftable category,
less than 100 ym diameter, is: cone c* flat > flooding >
Raindrop® nozzle. Table 4 presents these laboratory test
results.
Table 4. DROPLET SIZE COMPARISON OF FOUR NOZZLE TYPES12
Spray
pattern
Flooding
Flat
Cone
Raindrop
(cone)
Pressure,
kPa (psi)
275.8 (40)
275.8 (40)
275.8 (40)
275.8 (40)
Flow,
kg/min (gal/min)
1.13 (0.30)
1.13 (0.30)
1.10 (0.29)
1.10 (0.29)
Volume
median
diameter,
ym
210
202
195
410
% of Spray
volume
under
100 ym
13.0
15.5
15.9
0.8
c. Orientation of Nozzle - Further changes in the atomiza-
tion of sprays can be obtained by altering the discharge angle
of the nozzle in relation to the spray machine's airstream.
With ground machines, an increased wind speed deflects the
spray more quickly and deflects larger drops, also. If the
spray is aimed partly in the direction of the wind, instead
of vertically, it becomes drift more readily. On the other
hand, if the spray is projected cross-wind at an angle to the
vertical it is not inclined with the wind, but it takes
longer to reach the ground; the wind then has more time to
act upon it and deflect it. The end effect is similar — it
is more likely to become drift than a similar drop aimed
downward in the vertical direction.11 Preliminary field
12Ware, G. W., W. P. Cahill, and B. J. Estesen. Pesticide
Drift: Aerial Applications Comparing Conventional Flooding
vs. Raindrop® Nozzles. Journal of Economic Entomology.
68(3) :329-330, 1974.
19
-------
sampling has shown that nozzles directed upward emit three
times as much spray drift as nozzles directed downward (see
Appendix B).
An aircraft atomization system is more complex than that of
a ground machine, involving, first, the hydraulic ejection
of the liquid under pressure through an orifice and, secondly,
the effect of the slipstream's air velocity, which has a
specific relation to the final atomization.8 Laboratory
tests have been conducted13 to determine the effects of
orienting the nozzle in four positions relative to the air-
stream: into the airstream, vertically downward, horizontal
with the flow, and at an angle of 0.785 rad(45°) to the air-
stream. The effects of these orientations on the drop size
spectrum, shown in Figure 2, indicate that orientation
horizontally with the flow gives the largest drops. The
State of California requires jet nozzles to be directed back
or with the slipstream for aerial application of injurious
herbicides in specified hazardous areas.7
3. Type and Operation of Equipment
The type of application equipment (aircraft or ground machine)
and the way in which it is operated can affect both the drop-
let size spectrum produced and the amount of spray available
for drift.
a. Aircraft vs. Ground Machine - The emphasis in drift
control work has been on the aircraft applicator rather than
the ground rig because the principal problem areas have been
more frequently associated with large-scale pesticide
13Coutts, H. H., and W. E. Yates. Analysis of Spray Droplet
Distributions from Agricultural Aircraft. Transactions of
the American Society of Agricultural Engineers.
11(1):25-27, 1968.
20
-------
o
an
700
600
500
400
300
200
100
70
Q
£ 50
_i
Q_
O
2 30
D6/46® 275 kPa UOpsi)
AIR @ 160.9km /hr( 100 mph )
2.8% OIL IN EMULSION
I
I
I I I I I I I II I
0.01 0.1 1 10 20 30 40 50 60 70 80 90 95
CUMULATIVE PERCENTAGE OF VOLUME
98 99
Figure 2. Drop spectra for four nozzle orientations13
Courtesy of H. H. Coutts, W. E. Yates and
the American Society of Agricultural Engineers.
operations involving aircraft. It has been demonstrated14
that large ground sprayers, particularly those using air car-
rier means such as a mist blower, do produce a drift hazard
equal to or greater than that produced by aircraft. However,
greater control exists over a ground rig and its lower dis-
charge rate has less drift potential than that of an aircraft.
Field research to compare the spray drifts from simultaneous
applications by a high clearance, self-propelled ground
sprayer and by a "standard" airplane sprayer showed that at
llf.Ware, G. W. , E. J. Apple, W. P. Cahill, P. D. Gerhardt,
and K. R. Frost. Pesticide Drift. II. Mist Blower vs.
Aerial Application of Sprays. Journal of Economic
Entomology. ^2_(4) : 844-846 , August 1969.
21
-------
all distances downwind the aerial application resulted in 4
to 5 times as much drift as the ground sprayer created for
both evening and morning treatments.15
The major factor affecting increased drift emissions due to
aerial application is the pattern of release into the air
wake created by the wing in flight. This wake carries the
material out toward the wing tips, then drops it in a swath
of about wingspan width. The vortex patterns develop into
two distinct vortices at each wing tip and a strong central
propeller wash. This vortex system is common for both
fixed-wing and helicopter equipment. Altering the wing tips
with spoil plates or other devices does not stop the vortex
from developing.7 Also, at a forward speed above 6.7 m/s to
11.2 m/s (15 mph to 25 mph) the helicopter does not develop
any greater downwash than does a fixed-wing aircraft; only
when hovering does a helicopter develop a large downwash.
The wake that any aircraft develops is principally a function
of the total weight of the craft and its load; the amount of
drag is a function of wing design being affected by all ex-
ternal equipment such as spreaders, propeller pump drives,
and booms. The lighter and aerodynamically "cleaner" the
aircraft is, the less turbulence there will be in the wake.
Field research has also shown that the high-shear turbulence
on the aircraft wake has more effect on atomizing the liquid
spray than has the viscosity in reducing this atomization.8
b. Height of Emission - The release height is an important
element to be considered in confining spray to the target
area. Although an increase in height is sometimes used to
I5Ware, G. W., B. J. Estesen, W. P. Cahill, P. D. Gerhardt,
and K. R. Frost. Pesticide Drift. II. Mist Blower vs.
Aerial Application of Sprays. Journal of Economic
Entomology. 62^4) .-840-843, August 1969.
22
-------
increase the swath width by allowing the wind to carry the
material downwind, conversely the elevation should be mini-
mized to reduce the drift hazard. Applications of injurious
herbicides in California must be released at an elevation
lower than 3.05 m (10 ft) for aerial applications.7
For ground sprayers, the situation is similar in principle;
the wind through the spray is that which results from the
combination of wind over the ground and the travel speed of
the tractor. It is this wind that determines whether or not
the small drops are winnowed out of the spray.11
c. Number of Swaths - The cumulative effect of successive
swaths in an area will affect the amount of material emitted
to the atmosphere and deposited as downwind drift. An in-
crease in the number of swaths from one to five increases
the ground deposit and the airborne concentration at the
downwind edge of the target area by about twice, but a
further increase in the number of swaths to 40 results in
only a slight further increase. At 100 m downwind the swaths
are still not additive, the hazard from 40 swaths being only
8 to 10 times that of one; at 1,000 m the hazard from 40
swaths is about 30 times that of one swath, while it can be
inferred that at 10,000 m the swaths would be nearly
additive.16
4. Meteorological Conditions
The airborne drift of agricultural sprays is a direct result
of the transport of the droplets by atmospheric movement.
Some of the major meteorological parameters that affect
16Yeo, D., N. B. Akesson, and H. H. Coutts. Drift of Toxic
Chemicals Released from a Low-Flying Aircraft. Nature.
ljn:131-132, January 10, 1959.
23
-------
drift are; wind speed, air temperature, humidity, and tur-
bulent mixing. The diffusion, transport, and deposition
characteristics of the wide range of droplet sizes present
are very complex, and the fundamental relationships for
predicting drift concentrations are not fully established.
a. Gravitational Forces - The gravitational force on a
droplet is one of the most significant factors on which
attention must be focused in order to understand airborne
drift. Drift studies begun in 1947 set a pattern that is
followed today. The researcher stated the principle:17
"Underlying all problems of field application of toxic
materials is the rate of settling of particles suspended in
the air."
The movement of a particle in air is a function of the
resultant of the gravitational and aerodynamic drag forces.
The gravitational force acts straight downward and is simply
the volume of the particle multiplied by the difference
between particle density and air density. The aerodynamic
force on a rigid particle is related to the particle's air
velocity, to its size and shape, and to the density and
viscosity of the air. In addition, for liquid particles the
surface tension and viscosity of the liquid may also affect
the drag force.
Whenever the forces are unbalanced, the particle will accel-
erate ir; the direction of the resultant force at a rate
defined by force = mass x acceleration. Thus a particle
falling from rest into still air will accelerate until the
gravitational force is counterbalanced by the drag force,
17Brooks, F. A. The Drifting of Poisonous Dusts Applied by
Airplanes and Land Rigs. Agricultural Engineering.
_28(6) :233-239, June 1947.
24
-------
and approach a constant terminal velocity, V.. It should be
emphasized that for water drops falling in air, particles less
than 100 pm in size will approach their terminal velocity in
less than 25 mm (1 inch). The distance required to achieve
95% of the terminal velocity increases to approximately 0.6 m
for a 500 ym particle and 5 m for a 2,000 ym particle.7
The terminal velocity for liquid droplets may vary from the
rigid sphere terminal velocity due to deformation of the
particle as well as circulation within the droplet. The
terminal velocity of water drops falling in air has been
accurately determined and the results indicate that for
drops below 80 ym the terminal velocity approaches that cal-
culated by Stokes Law.7 Table 5 illustrates the terminal
velocities of water drops as well as rigid spheres.
To minimize drift, the droplets should be large. However,
for a given application rate the number of droplets available
varies inversely with the cube of the mass median diameter.
Table 5 also illustrates the theoretical number of uniformly
sized drops per square area of flat surface for a 9.3 g/m2
(10 gal/acre) application. This hypothetical case was included
to illustrate the relative effect of droplet size on the
coverage and distribution aspects. A plant canopy is three
dimensional and the surface area of the plant that requires
coverage is many times larger than the surface ground area it
occupies.
Although the theoretical number of droplets continues to
increase with a reduction in size, the settling velocity
decreases and the resultant deposition at the desired location
may reach a peak and then drop off rapidly with a further reduc-
tion in droplet size. Aerodynamic catch also plays a part in
the deposit of small droplets which below 25 ym increasingly
tend to be directed around an object rather than impacting.7
25
-------
Table 5. TERMINAL VELOCITIES OF PARTICLES IN AIR AND NUMBER OF DROPS/AREA7
Diameter,
urn
1
10
50
100
200
300
400
500
1,000
2,000
3,000
4,000
5,000
Rigid sphere
Specific gravity = 0.8
V
m/s (ft/sec)
0.000027 (0.000088)
0.002 (0.008)
0.06 (0.20)
0.21 (0.68)
0.6 (1.9)
1.0 (3.2)
1.4 (4.6)
1.7 (5.6)
3.4 (11.0)
5.8 (19.0)
7.6 (25.0)
9.1 (30.0)
10.4 , (34.0)
Specific gravity =1.0
V
m/s (ft/sec)
0.000034 (0.00011)
0.003 (0.01)
0.07 (0.25)
0.26 (0.85)
0.7 (2.4)
1.2 (3.9)
1.6 (5.3)
2.1 (6.8)
4.1 (13.3)
6.7 (22.0)
8.8 (29.0)
10.4 (34.0)
11.6 ' (38.0)
Specific gravity =2.5
V
m/s (ft/sec)
0.000085 (0.00028)
0.008 (0.025)
0.2 (0.63)
0.5 (1.8)
1.4 (4.6)
2.3 (7.5)
3.0 (10.0)
3.8 (12.5)
7.0 (23.0)
11.3 (37.0)
14.0 (46.0)
16.5 (54.0)
18.3 (60.0)
Water droplet
Specific gravity = 1.0
V
m/s (ft/sec)
0.00003 (0.0001)
0.003 (0.01)
0.07 (0.25)
0.27 (0.89)
0.7 (2.4)
1.2 (3.8)
1.6 (5.3)
2.1 (6.8)
4.0 (13.2)
6.4 (21.0)
7.9 (26.0)
8.8 (29.0)
9.1 (30.0)
No. of drops based on
9.3 g/m2,
drops/m2 (drops/in2)
1.78 x 1013 (1.15 x 1010)
1.78 x 1010 (1.15 x 107)
1.43 x 108 (9.22 x 101*)
1.79 x 107 (1.15 x lO1*)
2.23 X 106 (1.44 X 103)
6.6 x 10s (4.27 x 102)
2.8 x 10s (1.80 x 102)
1.4 x 10s (9.2 x 101)
1.8 x 10" (1.15 x 101)
2.2 x 103 (1.4)
6.6 x 102 (4.3 x 10"1)
2.8 x 102 (1.8 x 10"1)
1.4 x 102 (9.2 x 10~2)
to
Reprinted from Pesticide Formulations, p. 275-341, by Marcel Dekker, Inc., 1973.
-------
b. Wind Speed - Wind speed is of importance in determining
transport distances and can provide an estimate of movement
under stable atmospheric conditions. Table 2 illustrates
the theoretical horizontal transport at nonturbulent con-
ditions for various size droplets falling at terminal
velocity. The table serves only as a guide to show the
effect of droplet size and points out the dramatic increase
in drift distance for droplets below 100 ym. The table is
based on no evaporation and no turbulence as well as uniform
wind velocity. However, in air movement near the boundary
layer the velocity decreases with a decrease in height until
it reaches zero at a height referred to as z0, a value called
the roughness length. The wind velocity profile varies
with surface roughness and atmospheric stability.7
c. Turbulence and Atmospheric Stability - Turbulence is
related to the roughness of the ground surface, the tempera-
ture gradient with height, and the wind velocity gradient
with height. Turbulence near the ground is partially induced
by the surface roughness, which is dependent on the size of
and distance between protruding elements. Vertical and hori-
zontal eddies are mechanically produced as the air streams
over and around the protruding elements. In addition,
mechanical turbulence is induced by the gradient of wind
velocity as it produces wind shear. The velocity gradient
is generally greater near the ground, increases with wind
speed for a given height, and is affected by the surface
roughness. The temperature gradient is important since it
represents the energy available for producing or depressing
eddies by buoyancy forces.7
The temperature profiles near the ground change diurnally.
At midday a superadiabatic condition may exist near the
ground because of high solar radiation. During early
morning or late afternoon a strong inversion may exist.
27
-------
During the morning heating period a mixed layer may exist
near the ground with an inversion layer persisting above.7
An irrigated crop will modify these temperature and stability
conditions because some of the incoming solar radiation is
utilized in evapotranspiration from the crops so that less is
available for heating the air. Thus the air over an irrigated
crop will never be as unstable as that over dry land. It is
even possible to get stable conditions over an irrigated crop
several hours before sunset. This is even more significant
if the irrigated field is located immediately downwind from
a large: dry area so that hot air is being carried above and
across the colder field.18
Field studies have shown19 that there is a progressive
decrease in downwind drift residues (and, presumably, airborne
concentrations) with a decrease in stability. Turbulent or
unstable conditions cause the spray effluent to swirl downward
and reach the ground near the source. Inversion or stable
conditions, which permit long periods of horizontal diffusion,
allow the effluent to spread over a wide area. The fact that
stability is favorable for stack disposal, but not for pesti-
cide spraying, is a result of height of disposal and particle
size. Stack particle concentration is reduced by a wide area
diffusion that is not possible in a low height pesticide
distribution.
18Scotton, J. W. Atmospheric Transport of Pesticide
Aerosols. U.S. Department of Health, Education, and
Welfare, Public Health Service. Washington. PB 228 612.
July 1965. 30 p.
19Yates, W. E., N. B. Akesson, and H. H. Coutts. Drift
Hazards Related to Ultra-Low-Volume and Diluted Sprays
Applied by Agricultural Aircraft. Transactions of the
American Society of Agricultural Engineers.
10(5):628-632,638, 1967.
28
-------
d. Evaporation - All spraying equipment produces a spectrum
of droplet sizes, with the further complication that evapora-
tion of a water carrier will occur unless atmospheric humidity
is quite high. Since small droplets fall more slowly than
larger ones, evaporation of the carrier serves to increase
the size range of the droplets over their range when emitted
into the air.20 Water is the most frequently used carrier
because of availability, low cost, and freedom from phytotoxic
effects. Vapor pressure is the prevailing factor controlling
evaporation, but it is not easily evaluated, particularly
when complex mixtures of emulsions and solutions are used in
the spray formulation.
The fraction of droplets that are subject to drift is set
roughly as that portion of the droplet spectrum below 100 ym.
While this diameter may be considered unduly large for drift,
in a typical spray mixture 95% of the liquid is water, which
will quickly evaporate and reduce a 100-ym droplet to only
40 ym in about 15 seconds.10 Thus, within a few hundred
meters of an aircraft, the airborne fraction of the spray in
a drift "cloud" will be reduced to a volume equivalent to the
relatively nonvolatile fraction. The reduced droplet size
produces a lower settling rate that causes a greater portion
of the drift "cloud" to be dispersed and carried out of the
target area.
The driving force of evaporation can be expressed as the
difference between the vapor pressure at the droplet surface
and that in the surrounding air. The rate of change of
20Pooler, F. Atmospheric Transport and Dispersion of
Pesticides. (Presented at the Symposium on Guidelines
for Environmental Studies of Pesticides. 162nd National
Meeting, American Chemical Society. Washington.
September 1971.) 20 p.
29
-------
diameter (d) of a single drop in a large volume of surround-
ing air with time (t) can be expressed as:
at
where AP = vapor pressure gradient between the surrounding
air and the droplet surface
P = partial pressure of air
K = diffusivity of water vapor in air at the ambient
temperature
C.. = effective transfer coefficient at instantaneous
fall velocity V
The equation shows that the rate of change in diameter is
inversely proportional to the drop diameter at zero relative
velocity. Thus evaporation would change the diameter of a
small crop at a faster rate than that of a larger drop.7
Curves of the drop diameter as a function of time for water
drops falling in air of different humidities have been
presented in the literature.7 Table 6 illustrates the time
required for various size droplets to reduce to 10% of
original volume, and the vertical distances they would fall.
In thin study the evaporation rate was based on the assumption
that the drops were falling at a terminal velocity which
varied with evaporation, and the instantaneous velocity was
based on Stokes Law; thus, results are limited to drops less
than 100 ym. The table also represents a minimum time and
distance since the data are based on an evaporation rate for
a single pure water drop in a large atmosphere. The evapora-
tion from the emission of a large number of drops in a spray
would increase the partial vapor pressure in the surrounding
air and increase the drying period.7 The effect of an
increase in relative humidity from 30% to 70% is also shown
in Table 6.
30
-------
Table 6. TIME AND VERTICAL FALL DISTANCE FOR PURE WATER TO EVAPORATE
FROM D0 TO Df AT 25°C, 101.3 kPa7
Initial
diameter
D0, ym
100
80
60
40
Final diameter
equivalent to
10% of initial
volume D , ym
46
37
28
19
30% Relative humidity
Time, s
4.2
2.8
1.7
0.8
Vertical
distance, m
0.76
0.24
<0.15
<0.15
70% Relative humidity
Time , s
9.2
6.3
3.8
1.8
Vertical
distance, m
1.62
0.67
0.23
<0.15
AP = 2.3 kPa (0.68 in. Hg).
'AP = 1.0 kPa (0.29 in. Hg).
Reprinted from Pesticide Formulations, p. 275-341,
by Marcel Dekker, Inc., 1973.
D.
GEOGRAPHICAL DISTRIBUTION
Cotton is defoliated or desiccated prior to harvest wherever
it is grown in the U.S. The major concentrations of cotton
producing regions are located in the Mississippi River Valley
from the bootheel of Missouri to the top of Louisiana, the
Blacklands region of Texas running roughly from Austin to
Paris, and the High and Low Rolling Plains regions of Texas
situated in and just below the panhandle. Other, smaller
regions are the Lower Rio Grande Valley of Texas, the Gulf
Coast (around Corpus Christi) of Texas, the San Joachin
Valley in California, and a disperse belt in the Deep South
below the Appalachian Mountains.
Figure 321 illustrates the geographical distribution of cotton
harvested. In conjunction with this figure, Figure 4 shows
the distribution of all crop acreage treated with chemicals
for defoliation, growth control, or thinning of fruit. The
2 Census of Agriculture, 1969. Volume V, Special Reports.
Part 15, Graphic Summary. Washington, U.S. Bureau of the
Census, 1973.
31
-------
u>
UNITED STATES
TOTAL
46,524km2
(11,496,320 ACRES)
i
Figure 3. Cotton harvested, 196921
-------
1 DOT=4 km
(1,000 ACRES)
UNITED
STATES
TOTAL
2
23,395kni
\(5, 780,991 ACRES)
Figure 4. Acreage treated with chemicals for defoliation or for growth
control of crops or thinning of fruit, 196921
-------
other crops chemically defoliated are potatoes, canning
tomatoes, and species of legumes grown for seed,1 but it can
be readily seen from these two figures that the usage of
defoliation chemicals closely follows the areas where cotton
is harvested. One notable exception is the upper regions of
the High and Low Rolling Plains (panhandles) of Texas and
Oklahoma where the cotton matures just before the first
freeze of autumn. Growers there wait for the freeze to
desiccate the cotton naturally.
Texas ;Ls the major cotton producing state, harvesting about
40% of the U.S. cotton acreage. Table 7 shows the percent
of U.S. total cotton acreage harvested for the major producing
states. The Mississippi River Valley states of Mississippi,
Arkansas, Louisiana, Tennessee, and Missouri produce about
35% of the total cotton crop.
Table 7. COTTON ACREAGE HARVESTED, PERCENT OF U.S. TOTAL
State
Texas
Mississippi
Arkansas
California
Alabama
Louisiana
Oklahoma
Tennessee
Georgia
South Carolina
Arizona
Missouri
TOTAL
1970
43.9
10.7
9.6
5.9
4.8
4.0
4.0
3.5
3.4
2.6
2.5
2.2
97.1
1971
41.3
11.6
9.9
6.5
4.9
4.4
3.5
3.7
3.4
2.8
2.5
2.7
97.0
1972
39.4
12.3
10.7
6.5
4.4
5.1
3.9
3.7
3.6
2.6
2.4
3.1
97.8
34
-------
SECTION IV
EMISSIONS
A. SELECTED POLLUTANTS
The emissions from cotton defoliation or desiccation consist
entirely of fugitive aerosols of the herbicide used. Table 8
presents the chemicals used and their respective toxicities
and TLV's. Sodium chlorate, DBF, and Folex are most commonly
used to defoliate cotton; arsenic acid and paraquat are most
commonly used to desiccate cotton.22"26 Table 9 shows the
rates of application and dilution data for the major chemicals,
All are diluted with water, and perhaps a very small amount
of surfactant or sticking agent is added to the formulation.
22Akesson, Dr. N. B. Department of Agricultural Engineering,
University of California-Davis. Personal communication,
March 1975.
23Metzer, Dr. R. B. Texas Agricultural Extension Service,
College Station, Texas. Personal communication,
February 1975.
24Ware, Dr. G. W. Department of Entomology, University of
Arizona, Tucson, Arizona. Personal communication,
January 1975.
25Miller, Dr. C. S. Department of Plant Sciences, Texas
A&M University, College Station, Texas. Personal
communication, January 1975.
26Mullins, Dr. J. A. Tennessee Agricultural Extension
Service, Jackson, Tennessee. Personal communication,
January 1975.
35
-------
Table 8. DEFOLIANTS AND DESICCANTS USED FOR COTTON
GJ
en
Most cormon name
Defoliants
DBF
Folex
Sodium chlorate
Pentachlorophenol
Endothall
Calcium cyanamide
Magnesium chlorate
Ammonia
Desiccants
Arsenic acid
Paraquat
Cacodylic acid
Sodium cacodylate
Potassium azide
Chemical name
s , s , s-Tributylphosphorotrithioate
s ,s f s-Tributylphosphorotrithioite
Chloric acid, sodium salt
Pentachlorophenol
7-Oxabicyclo (2.2.1) heptane-
2,3-dicarboxylic acid
Calcium cyanamide
Chloric acid, magnesium salt
Ammonia
Orthoarsenic acid
1,1' -Dimethyl-4 , 4 ' -bipyridinium salt
Hydroxydimethyl arsine oxide
Methanearsonic 'acid
Potassium azide
Other names
Butiphos, Chemagro 1776,
Chemagro B-1776,
DBF Defoliant, De-Green,
Fos-Fall A, E-Z-off-D,
Ortho Phosphate Defoliant
Merphos, Deleaf Defoliant,
Easy Off-D,
Mobil Cotton Defoliant
Chlorax, De-Pol-Ate,
Drop-Leaf, Fall, MBC,
Monobor- chlorate.
Shed-A-Leaf , Tumbleaf
PCP, Ded-Leaf, Dowicide 7
Accelerate, Des-I-Cate
AERO Cyanamid
Desiccant L-10, Zotox,
Sinergized H-10, Hi-Xield
Gramoxone ,
Aerial Gramoxone ,
Paraquat CL
Silvisar 510, Rad-E-Cate,
Ansar 138
Phytar 560, Bollseye
Oral-rat LD5(),
mg/kg
ISO27
910 27
1.2Q028
2727
8027
3927
5,250(LDLO)27
_b
_b
5727
1.35027
3.20027
b
TLV,
mg/m^
0.963
3.9a
4.83
0.529
0.69
0.529
153
1829
0.529
0.529
5.23
10.23
b
Comments
Major use
Major use
Major use, mixed
with sodium
metaborates or
magnesium chloride
Old, minor use
Minor use, sodium
salt used
Old, once
principally used
Minor use
Old
Major use
Major use, chloride
and bismethyl
sulfate salts
Minor use
Minor use
Minor use
Estimate, see Appendix C.
b
Not available.
27The Toxic Substances List, 1974 Edition. U.S. Department of Health, Education, and Welfare. Rockville, Maryland.
HEW Publication No. (NIOSH) 74-134. June 1974. 904 p.
2S1969 Farm Chemicals Handbook. Willoughby, Ohio, Meister Publishing Co., 1968. p. D158.
29TLVs® Threshold Limit Values for Chemical Substances and Physical Agents in the Workroom Environment with
Intended Changes for 1975. American Conference of Governmental Industrial Hygienists. Cincinnati. 1975. 97 p.
-------
Table 9. FORMULATION AND DILUTION OF MAJOR HARVEST-AID CHEMICALS
30
Chemical
name
Sodium
chlorate
DBF
Folex
Arsenic
• acid
Paraquat
Percent
principal
formulation
18.2 to 28.0
70.5
71.2
75.0
29.1
Suggested
rate of
application
1.14 x 10~5 to .4. 59 x 10~5 m
(0.75 to 3 gal/acre)
2.5 x 10~6 to 3.8 x 10~6 m
(0.16 to 0.25 gal/acre)
2.5 x 10"6 to 3.8 x 10~6 m
(0.16 to 0.25 gal/acre)
3.8 x 10~6 to 5.7 x 10~5 m
(0.25 to 0,375 gal/acre)
1.9 x 10~6 to 3.8 x 10~6 m
(0.125 to 0.25 gal/acre)
Maximum
registered
rate
9.2 g/m2
(5 Ib/acre)
2.8 g/m2
(1.5 Ib/acre)
2.8 g/m2
(1.5 Ib/acre)
8.1 g/m2
(4.4 Ib/acre)
0.9 g/m2
(0.5 Ib/acre)
Dilution data
Application
by airplane
7.65 x 10~5 to 15.3 x 10~5 m
(5 to 10 gal/acre)
7.65 x 10~5 to 15.3 x 10~5 m
• (5 to 10 gal/acre)
7.65 x 10~5 to 15.3 x 10~5 m
(5 to 10 gal/acre)
7.65 x 10~5 to 15.3 x 10~5 m
(5 to 10 gal/acre)
Not applicable
Application
by ground machine
1.53 x 10-1* to 3.06 x 10"1* m
(10 to 20 gal/acre)
1.53 x 10'1* to 3.06 x 10"1* m
(10 to 20 gal/acre)
1.53 x 10-1* to 3.06 x KT1* m
(10 to 20 gal/acre)
1.53 x lO"1* to 2.30 x 10"1* m
(10 to 15 gal/acre)
1.53 x 1Q-1* to 4.59 x lO"1* m
(10 to 30 gal/acre)
U)
301969 Cotton Defoliation Guide. Texas Agricultural Extension Service.
College Station, Texas. Bulletin No. L-145.
-------
Calcium cyanamide is excluded as a major cotton defoliant,
since Lt is believed now to be used primarily as a fertilizer
rather than an herbicide. The compound is not produced
domestically, but imports have declined from a high of
1.48 x 105 metric tons (1.63 x 105 tons) in 1946, when it
enjoyed major use as a cotton defoliant, to 6.8 x 103 metric
tons (7.5 x 103 tons) in 1970.31
1. Folex and DBF
Folex and DEF function exclusively as defoliants. Chemically
they are closely related and can be made from the same raw
materials; butyl mercaptan and phosphorus trichloride react
to form the phosphorotrithioite (Folex). This is then air
oxidized to produce the phosphorotrithioate (DEF). When Folex
is used as a cotton defoliant it probably is converted to DEF
in the atmosphere. In chemical residue analysis, Folex oxi-
dizes to DEF upon standing in dilute solution, especially in
acetone, and is often seen as DEF if the analysis is carried
through an extraction and clean-up procedure at residue levels
where air oxidation would cause conversion.32 No data are
published on further degradation products, and no quantita-
tive data were found on toxic properties of DEF or Folex.
2. Sodium Chlorate
Sodium chlorate (NaClOs) is freely soluble in water and
highly toxic to most plants, hence it is a nonselective
31Strickland, J., and T. Blue. Environmental Indicators for
Pesticides. Stanford Research Institute, Council on
Environmental Quality Contract EQC 217. Menlo Park,
California. PB 210 666. April 1972. p. 38.
32FDA Pesticide Analytical Manual. Vol. II. Pesticide Reg.
Sec. 120,272. U.S. Department of Health, Education, and
Welfare. November 1973.
38
-------
herbicide. Chlorate defoliants are usually sold in the form
of dry crystals, to be dissolved in water and applied as a
spray. Although it is apparently safe to handle magnesium
chlorate, the more popular sodium chlorate is a powerful
oxidizing agent and may cause spontaneous combustion of
organic matter.33 For commercial use, sodium chlorate is
mixed with fire suppressors, usually sodium borates or
magnesium chloride; these mixtures have proved quite safe.3
The compound is leached from the soil rather rapidly, and
appears to be slowly broken down by soil microorganisms, so
that in humid areas the herbicidal effect is not permanent.33
Sodium chlorate is severely irritating to mucous membranes.
No data are available on its acute inhalation toxicity, nor
on its chronic toxicity, and no residue tolerances have been
set. No subacute or chronic hazards to human health have
been attributed to the use of sodium chlorate as an
herbicide.31t
3. Arsenic Acid
Arsenic acid functions exclusively as a desiccant. Chemically
it is known as orthoarsenic acid (HsAsOiJ and is sold as a
75% formulation. It is corrosive to metal and is not applied
by airplane for this reason.
When arsenical compounds are present in the air, arsenic may
be absorbed by inhalation, ingestion, or absorption through
33Weed Killers. In: Kirk-Othmer Encyclopedia of Chemical
Technology, Second Edition. Vol. 22. New York, John
Wiley & Sons, Inc., 1969. p. 19.
3ltvon Riimker, R. , E. W. Lawless, and A. F. Meiners.
Production, Distribution, Use and Environmental Impact
Potential of Selected Pesticides. Midwest Research
Institute, Council on Environmental Quality. Contract
EQC-311. Kansas City, Missouri. March 1974. p. 256.
39
-------
the skin. The airborne arsenic frequently causes irritation
of the skin and mucous membranes, absorption taking place
most readily on moist surfaces such as folds in the skin or
mucous membranes. Thus, dermatitis, mild bronchitis, and
nasal irritation are common symptoms of arsenic poisoning.
With more severe exposure, perforation of the nasal septum
takes place.35
A residue tolerance of 4 ppm as As2C>3 equivalent on cotton-
seed has been set.30
4 . Paraquat
Paraquat is the accepted common name for a formulation of
l,l'-dimethyl-4,4 ' -bipyridinium ion, a quarternary ammonium
compound. The concentration of the active ingredient is
expressed as the amount of bipyridinium cation per gallon
and is formulated to contain 240 kg of the cation per cubic
meter (2 Ib/gal) . Anions included in paraquat formulations
are the chloride (Cl) and the bis methyl sulfate (MS) .
However, all application rates are expressed in terms of the
active cation. The material is readily soluble in water,
nonvolatile, and nonflammable. The concentrated solution is
corrosive to mild steel, tin plate, galvanized iron, and
aluminum. Paraquat is formulated with a corrosion inhibitor;
however, the dilute solution is still corrosive to galvanized
iron.2
The chemical can be degraded by ultraviolet light to methyl
quarternary isonicotinic acid and methyl amine hydrochloride. 36
35Sullivan, R. J. Preliminary Air Pollution Survey of Arsenic
and Its Compounds, A Literature Review. Litton Systems,
Inc., HEW Contract PH 22-68-25. October 1969. p. 2.
36Slade, P. Photochemical Degradation of Paraquat. Nature.
207^:515-516, 1965.
40
-------
A very small oral dose of the concentrate by ingestion or
inhalation may produce irreversible lung fibrosis.37 Damage
to the lung is characterized initially by edema and hemorrhage,
and at later stages by fibrosis. Except when extremely large
amounts are taken, signs of pulmonary damage are not usually
seen for several days after ingestion.38
A residue tolerance of 0.5 ppm on cottonseed has been set.
30
B. EMISSION FACTORS
Investigations of spray drift from agricultural application
of pesticides have been reported by several authors. Concern
has been mostly centered on off-target deposits of chemicals
rather than the remaining airborne fraction; however, some
researchers supplemented their drift deposit collection
stations downwind with air sampling devices. Table 10 is a
compilation of the information gathered from those articles
containing data that can be used to calculate emission
factors.15'39-1*1
37Staiff, D. C., S. W. Comer, J. F. Armstrong, and H. R. Wolfe.
Exposure to the Herbicide, Paraquat. Bulletin of Environ-
mental Contamination and Toxicology. 1JJ3):334-340, 1975.
38Rose, M. S. The Search for an Effective Treatment of
Paraquat Poisoning. Chemistry.and Industry (London).
1975(10):413-415, May 17, 1975.
39Argauer, R. J., H. C. Mason, C. Corley, A. H. Higgins,
J. N. Sauls, and L. A. Liljedahl. Drift of Water-Diluted
and Undiluted Formulations of Malathion and Azinphosmethyl
Applied by Airplane. Journal of Economic Entomology.
61^(4) :1015-1020, August 1968.
^OWare, G. W., B. J. Estesen, W. P. Cahill, P. D. Gerhardt,
and K. R. Frost. Pesticide Drift. III. Drift Reduction
with Spray Thickeners. Journal of Economic Entomology.
6J3(4) :1314-1316, August 1970.
^Ware, G. W., B. J. Estesen, W. P. Cahill, and K. R. Frost.
Pesticide Drift. V. Vertical Drift from Aerial Spray
Applicaitons. Journal of Economic Entomology.
(55_(2) :590-592, April 1972.
41
-------
Table 10. CALCULATED EMISSION FACTORS FROM PUBLISHED DATA
FOR DRIFT FROM AGRICULTURAL SPRAYING
Pesticide
chemical
Azinphosmethyl
Az inphosmethyl
Azinphosmethyl
Az inphosmethyl
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Methoxychlor
Application
equipment
Airplane
Airplane
Airplane
Airplane
Hi-Boy
Hi-Boy
Hi-Boy
Hi-Boy
Airplane
Airplane
Airplane
Airplane
Airplane
Airplane
Airplane
Airplane
Airplane
Airplane
Airplane
Airplane
Nozzle
type
Hollow cone, down
Hollow cone, down
Hollow cone, down
Hollow cone, down
Tee-Jet
Tee-Jet
Tee-Jet
Tee-Jet
Hollow cone
Hollow cone
Hollow cone
Hollow cone
Diaphram
Diaphram
Diaphram
Diaphram
Flood tip
Flood tip
Flood tip
Flood tip
Application
ra te ,
mg/m2
56
56
56
56
168
168
202
202
224
224
224
224
224
224
224
224
52
52
52
52
Downwind
distance ,
m
60
60
600
600
50
100
50
100
50
100
50
100
131
332
735
1,539
25
50
100
200
Emission
factor,
g/kg
118.539
93.139
31.839
28. O39
28. O15
29. O15
17. O1 5
11. O15
59. O15
25. O15
14. O15
12. O1*
110. O40
143. O40
35. O1*0
101. 0"°
19. O^1
7.041
78. O1*1
17. 5* !
0.785 rad (45°) tilt down.
-------
The emission factors shown in Table 10 range over an order
of magnitude, mostly due to the variability in application
systems and meteorological conditions. In general, the
emission factors within experiments tend to be lower as the
downwind distance of the air samplers is increased, due to
settling of the spray aerosols.
Preliminary field sampling of arsenic acid application to
cotton was conducted and emission factors calculated. Table
11 summarizes the arsenic acid emission data. Appendix B
contains the details and methods of calculation used to
prepare Table 11.
The mean emission factor for arsenic acid application
(6.1 ± 2.9 g/kg) was eight times lower than the mean emission
factor estimated from literature data (48.8 g/kg). This is
attributable to the use of a ground machine rather than an
airplane and to the low volatility of arsenic acid. Emission
factors for application of sodium chlorate, DBF (or Folex),
and paraquat have been assumed to be 10 g/kg which is slightly
higher than arsenic acid. Table 12 summarizes the emission
factors for defoliation/desiccation of cotton which were used
in calculations of ground level concentration, mass emissions,
and affected population in Section IV.D.
C. DEFINITION OF REPRESENTATIVE SOURCES
Due to the heterogeneous nature of cotton defoliation and
desiccation, four representative sources are defined for use
in determining the source severity which is described in
Section IV.D and Appendix A.
43
-------
Table 11. CALCULATED EMISSION FACTORS FROM PRELIMINARY FIELD SAMPLING
OF ARSENIC ACID APPLICATION TO COTTON
Sample
number
A-1R
A-2R
A-1L
B-1R
B-4R
B-2L
B-3L
C-1R
C-2R
C-3R
C-4R
C-3L
Application
equipment
Hi-Boy
Hi-Boy
Hi-Boy
Hi-Boy
Hi -Boy
Hi-Boy
Hi-Boy
Hi-Boy
Hi-Boy
Hi-Boy
Hi-Boy
Hi-Boy
Nozzle
type
Hollow cone, up
Hollow cone, up
Hollow cone, up
Hollow cone, up
Hollow cone, up
Hollow cone, up
Hollow cone, up
Spinner cone, down
Spinner cone, down
Spinner cone, down
Spinner cone, down
Spinner cone, down
Application
rate,
mg/m2
1,052
1,052
1,052
1,052
1,052
1,052
1,052
1,052
1,052
1,052
1,052
1,052
Downwind
distance,
ra
300
250
250
50
250
150
250
200
25
400
25
25
Average =
Emission
factor,
g/kg
11.4
16.2
5.7
0.9
6.8
6.8
10.0
1.9
2.3
4.3
4.2
2.7
6.1 + 2.9
(95% confi-
dence level)
tfe.
-------
Table 12. EMISSION FACTORS FOR DEFOLIATION OR DESICCATION
OF COTTON
Pollutant
Sodium chlorate
DBF
Arsenic acid
Paraquat
Emission
g/kg
10. O3
10. Oa
6.1 ± 2.9
10. Oa
factor,
(Ib/ton)
(20.0)
(20.0)
(12.2 ± 5.7)
(20.0)
Assumed.
The representative source for sodium chlorate application
consists of a cotton farm of 0.70 km2 (173 acres), located
in the Mississippi River Delta region, with an aerial appli-
cation rate of 0.56 g/m2 (5.0 Ib/acre) and an emission factor
of 10 g/kg. The representative source for DBF application
consists of a cotton farm of 0.70 km2 located in the
Mississippi River Delta region, with an aerial application
rate of 0.17 g/m2 (1.5 Ib/acre) and an emission factor of
10 g/kg. The representative source for arsenic acid appli-
cation consists of a cotton farm of 0.61 km2 (150 acres)
located in the Blacklands region of Texas, with a ground
machine application rate of 0.49 g/m2 (4.4 Ib/acre) and an
emission factor of 6.1 g/kg. The representative source for
paraquat application consists of a cotton farm of 1.05 km2
(260 acres) located in the Panhandle region of Texas, with
an aerial application rate of 0.056 g/m2 (0.5 Ib/acre) and
an emission factor of 10 g/kg.
The following assumptions were included to characterize the
spraying conditions: (1) the cotton field is square; (2) the
spray swath is perpendicular to the wind direction; (3) the
effective height of emission is negligible; (4) U.S. average
meteorological conditions prevail; and (5) the time of
exposure to emissions is taken to be the time necessary to
45
-------
spray the complete field plus the time needed to turn the
spray equipment.
D. SOURCE SEVERITY
1. Definition
To obta.in an indication of the hazard potential of the
emissions from agricultural spray applications, the source
severity, SA, for the special case of agricultural field
spraying was defined (Equation 1) as:
c _ *"
A ~ *A
where X is the time-averaged ground level concentration
during spraying at the downwind field perimeter from a
representative source (see Section IV.C), and F is a
./*
threshold limit value (TLV) for noncriteria pollutants with
a safety factor of 100 applied to the TLV. No correction is
applied for exposure time since it is 8 hours or less for the
representative source. This source severity factor represents
the ratio of time-averaged maximum ground level exposure to
the hazard level of exposure for a particular pollutant.
2. Ground Level Concentration
The time-averaged ground level concentration, X, of the
pollutant resulting from agricultural spray applications was
estimated by Gaussian plume dispersion techniques (see
Appendix A). The maximum concentration to which a population
may be exposed is assumed to be located at a field's peri-
meter downwind from the source. The following equation was
used for the calculation of x";
,1/2
X = f -
Try tuozl
46
-------
where n = number of spray swaths made in the representative
field (dimensionless)
QL = emissions per length for a single spray pass, g/m
t = time to complete spraying representative field
including turning time, s
u = mean wind speed, m/s (assumed to be U.S. average,
4.5 m/s)
a ... = standard deviation of the distribution of
z pollutant material in the vertical direction for
a puff (neutral stability assumed,
a T = 0.15 D0-70), m
zi
D = distance from center of representative field to
perimeter, m
TT = 3.14
Substituting Equation 3 into Equation 1 and including
F- = TLV • 1/100 yielded the following equation for the
source severity:
119 • n • QT
S. = - ^— (4)
t • TLV • D ° • 7 °
The nature of this type of source precludes the inclusion of
a source severity distribution since the model predicts sever-
ities which approach infinity for small field sizes. The
smallest cotton field size (finite) is unknown and of only
academic interest.
3. Population Exposed
To obtain a quantitative evaluation of the maximum population
invluenced by a high pollutant concentration due to emissions
from spray applications in a typical cotton field, the area
exposed to the time-averaged ground level concentration, x»
for which x/Fa - 0-1 was obtained by determining the area
jfi
within the isopleth for x-42 The number of persons within
the exposed area was then calculated by using the proper
population density.
42Turner, D. B. Workbook of Atmospheric Dispersion Estimates.
U.S. Department of Health, Education, and Welfare. Cincinnati,
Public Health Service Publication No. 999-AP-26. May 1970,
65 p.
47
-------
The representative population density used in the calculation
of affected population was the average of the state popula-
tion densities for Mississippi, Louisiana, and Arkansas for
the Mississippi River Delta region. For the Blacklands and
Panhandle regions of Texas, the average of the population
densities of the counties listed in Appendix A was used.
For each of the major cotton defoliant and desiccant chemicals
with an emission source severity, S , greater than or equal
to 0.1, the area and population exposed are shown in Table 13.
The source severity for arsenic acid application by ground
machine was 0.69 ± 0.32, the highest value among the major
defoliants and desiccants. Also, the emission factor for
arsenic: acid was derived from field sampling of normal prac-
tice, actual situations. Source severity factors for paraquat,
sodium chlorate, and DEF applications were 0.30, 0.44, and
0.67, respectively. It should be emphasized that these
calculations were based on the assumed emission factors men-
tioned in Section IV.B.
Table 13.
SOURCE SEVERITY, AREA, AND POPULATION EXPOSED TO POLLUTANTS
FOR WHICH
Pollutant
Arsenic acid
Paraquat
Sodium chlorate
DEF
Source
severity
0.69 ± 0.32
0.30
0.44
0.67
Affected
area, km2
X/FA ^o.i
116.75
17.59
37.48
125.10
Population
density,
persons/km2
52.5
18.3
20.1
20.1
Maximum
exposed
population ,
persons
6,134
322
754
2,517
4.
Total Air Emissions
The contribution of cotton defoliation and desiccation to
statewide and nationwide air emissions was estimated from
48
-------
the statewide cotton acreage that was defoliated, multiplied
by the percent usage by area for each of the four major
chemicals, multiplied by the application rate of each chemical
for each state, multiplied by the emission factor for each
chemical.
Published data on agricultural usage of pesticides are
incomplete, frequently incongruous, and at times actually
misleading. Data from several sources had to be collated in
order to arrive at the estimates of total air emissions. No
measure of accuracy can be attached to these estimates.
Table 14 presents the reported cotton acreage harvested and
defoliated for each of the cotton producing states. Metric
units are used in Table 14a while English units are used in
Table 14b. The area of cotton defoliated in each state was
estimated from these data and appears in the right-hand
column. Another literature source yielded the quantities
of defoliants and desiccants used on crops (mostly cotton)
and acreage of crops treated by region, as shown in Table 15.
From these data the rate of application by region of the U.S.
and the percent usage by area for each of the major harvest-
aid chemicals (excluding paraquat) were estimated for each
state.
More comprehensive data for Texas, the largest cotton pro-
ducing state, were obtained and are formatted in Table 16.
Desiccation by arsenic acid occurs in Texas in the Rolling
Plains, Central Basin, and Grand Prairies regions, while
paraquat desiccation was assumed to occur strictly in the
High Plains region. The ratio of sodium chlorate to DBF
usage by area was taken from Table 15 for the regions of
Texas which defoliate.
49
-------
Table I4a. COTTON ACREAGE HARVESTED AND DEFOLIATED, 1971
(Metric units)
State
Alabama
Arizona
Arkansas
California
Florida
Georgia
Kentucky
Louisiana
Mississippi
Missouri
New Mexico
North Carolina
Oklahoma
South Carolina
Tennessee
Texas
TOTALS
No . farms
reporting1* 3
2,376
724
4,424
2,142
16
900
15
2,389
4,957
1,287
745
328
828
786
1,818
13,440
37,175
Reported
crop area,1*3
km2
1,378
743
3,383
2,223
16
656
9
1,390
3,960
793
362
212
439
629
779
12,153
29,125
Total crop
area,1"1
km2
2,258
1,155
4,613
3,000
38
1,558
17
2,023
5,362
1,267
609
708
1,603
1,295
1,720
19,164
46,390
% Area
reported
61.0
64.3
73.3
74.1
42.7
42.1
52.9
68.7
73.8
62.6
59.5
30.0
27.4
48.6
45.3
63.4
62.8
Reported area
defoliated,"3
km2
805
498
2,405
1,887
15
461
4
878
3,260
311
11
164
38
479
277
2,333
13,825
% Area
defoliated
58.4
67.0
71.1
84.9
90.3
70.2
41.1
63.2
82.3
39.2
2.9
77.4
8.6
76.2
35.6
19.2
47.5
Estimated area
defoliated,
km2
1,319
774
3,280
3,916
34
1,094
7
1,279
4,413
497
18
548
138
987
612
3,679
22,035
Part 3, Cotton.
**3Census of Agriculture, 1971. Volume V, Special Reports.
Washington, U.S. Bureau of the Census, 1973.
^Agricultural Statistics 1973. Washington, U.S. Department of Agriculture,
1973. 617 p.
-------
Table 14b.
COTTON ACREAGE HARVESTED AND DEFOLIATED, 1971
(English units)
State
Alabama
Arizona
Arkansas
California
Florida
Georgia
Kentucky
Louisiana
Mississippi
Missouri
New Mexico
North Carolina
Oklahoma
South Carolina
Tennessee
Texas
TOTALS
No . farms
reporting1* 3
2,376
724
4,424
2,142
16
900
15
2,389
4,957
1,287
745
328
828
786
1,818
13,440
37,175
Reported
crop area,43
acres
340,491
183,573
835,920
549,308
3,969
162,120
2,273
343,521
978,444
195,939
89,564
52,439
108,383
155,517
192,437
3,002,967
7,196,865
Total crop
area,1*1*
acres
558,000
285,400
1,140,000
741,600
9,300
385,000
4,300
500,000
1,325,000
313,000
150,600
175,000
396,000
320,000
425,000
4,735,400
11,463,200
% Area
reported
61.0
64.3
73.3
74.1
42.7
42.1
52.9
68.7
73.8
62.6
59.5
30.0
27.4
48.6
45.3
63.4
62.8
Reported area
defoliated, lt3
acres
198,914
123,005
594,346
466,199
3,584
113,837
934
217,061
805,627
76,833
2,611
40,567
9,336
118,435
68,413
576,403
3,416,105
% Area
defoliated
58.4
67.0
71.1
84.9
90.3
70.2
41.1
63.2
82.3
39.2
2.9
77.4
8.6
76.2
35.6
19.2
47.5
Estimated area
defoliated,
acres
325,900
191,200
810,500
629,600
8,400
270,300
1,800
316,000
1,090,500
122,700
4,400
135,500
34,100
243,800
151,300
909,200
5,245,200
-------
Table 15. QUANTITIES OF DEFOLIANTS AND DESICCANTS (ACTIVE INGREDIENTS) USED ON CROPS
AND ACREAGE OF CROPS TREATED, BY REGION, 1971
Region
(states with cotton)
Corn Belt (Missouri)
Appalachian
(Kentucky, Tennessee,
North Carolina)
Southeast (Alabama,
Georgia, South
Carolina, Florida)
Delta States
(Arkansas, Louisiana,
Mississippi)
Southern Plains
(Texas, Oklahoma)
Mountain States
(Arizona, New Mexico)
Pacific States
(California)
Arsenic acid
Quantity,45
1,000 kg (1,000 Ib)
21 (46)
2,710 (5,975)
24 (52)
Area, k5
km2 (1,000 acres)
93 (23)
3,557 (879)
125 (31)
Rate,
g/m2 ( Ib/acre )
0.22 (2)
0.76 (6.8)
0.19 (1.7)
% Usage
by area3
0
0
2.7
0
54.7
48.4
0
Ul
to
Estimated.
Note: Blanks indicate data not reported.
-------
Table 15 (continued). QUANTITIES OF DEFOLIANTS AND DESICCANTS (ACTIVE INGREDIENTS)
USED ON CROPS AND ACREAGE OF CROPS TREATED, BY REGION, 1971
Region
(states with cotton)
Corn Belt (Missouri)
Appalachian
(Kentucky, Tennessee,
North Carolina)
Southeast (Alabama ,
Georgia, South
Carolina, Florida)
Delta States
(Arkansas, Louisiana,
Mississippi)
Southern Plains
(Texas, Oklahoma)
Mountain States
(Arizona, New Mexico)
Pacific States
(California)
DBF and Folex
Quantity,1*5
1,000 kg (1,000 Ib)
38
73
592 (1
1,362 (3
140
24
62
(84)
(162)
,306)
,003)
(308)
(51)
(136)
km2
360
704
3,379
8,324
1,643
125
320
Area,45
(1,000 acres)
(89)
(174)
(835)
(2,057)
(406)
(31)
(79)
Rate,
g/m2 (Ib/acre)
0.10 (0.9)
0.10 (0.9)
0.18 (1.6)
0.17 (1.5)
0.9 (0.8)
0.18 (1.6)
0.19 (1.7)
% Usage
by area3
91.8
89.7
97.3
90.0
25.3
48.4
13.8
Estimated.
-------
Table 15 (continued). QUANTITIES OF DEFOLIANTS AND DESICCANTS (ACTIVE INGREDIENTS)
USED ON CROPS AND ACREAGE OF CROPS TREATED, BY REGION, 1971
Region
(states with cotton)
Corn Belt (Missouri)
Appalachian
(Kentucky, Tennessee,
North Carolina)
Southeast (Alabama,
Georgia , South
Carolina, Florida)
Delta States
(Arkansas, Louisiana,
Mississippi)
Southern Plains
(Texas, Oklahoma)
Mountain States
(Arizona, New Mexico)
Pacific States
(California)
Chlorates and borates
Quantity,1*5
1,000 kg (1,000 Ib)
118
7
130
401
I
2,333
(260)
(16)
(287)
(884)
(2)
(5,145)
Area,45
km2 (1,000 acres)
32 (8)
81 (20)
927 (229)
1,303 (322)
8 (2)
2,003 (495)
Rate,
g/m2 (Ib/acre)
0.56 (5.0)a
0.09 (0.8)
0.15 (1.3)
0.30 (2.7)
0.11 (1.0)
1.17 (10.4)
% Usage
by areab
8.2
10.3
0
10.0
20.0
3.2
86.2
Ul
Assumed.
Estimated.
Note: Blanks indicate data not reported.
^Andrilenas, P. A. Farmers' Use of Pesticides in 1971 — Quantities. U.S. Department of
Agriculture. Washington. Agricultural Economic Report No. 252. July 1974. 56 p.
-------
Table 16. COTTON ACREAGE HARVESTED AND
DEFOLIATED IN TEXAS, 1971
High Plains region
Counties
Andrews
Armstrong
Bailey
Briscoe
Carson
Castro
Cochran
Crosby
Dawson
Deaf Smith
Floyd
Gaines
Glasscock
Gray
Hale
Hansford
Hockley
Howard
Lamb
Lubbock
Lynn
Martin
Midland
Moore
Farmer
Randall
Swisher
Harvested
km2
21.9
5.3
231.1
104.4
178.1
315.3
515.2
872.5
27.9
377.6
564.5
59.9
7.3
619.2
777.8
296.6
667.7
920.7
768.5
408.7
91.9
174.8
4.9
148.9
area,46
(acres)
(5,400)
(1,300)
(57,100)
(25,800)
(44,000)
(77,900)
(127,300)
(215,600)
(6,900)
(93,300)
(139,500)
(14,800)
(1,800)
(153,000)
(192,200)
(73,300)
(165,000)
(227,500)
(189,900)
(101,000)
(22,700)
(43,200)
(1,200)
(36,800)
Note: Blanks indicate less than 2.0 km2 (500 acres)
55
-------
Table 16 (continued). COTTON ACREAGE HARVESTED AND
DEFOLIATED IN TEXAS, 1971
High Plains region (continued)
Counties
Terry
Yoakum
Total Harvested Area
Total Defoliated Area
Percent Defoliated
Harvested
km2
639.0
223.8
9,023.4
2,255.7
area,1* 6
(acres)
(157,900)
(55,300)
(2,229,700)
(557,400)
25% (largely desiccated)23
Rio Grande Plain region
Counties
Atascosa
Cameron
Dimmit
Duval
Frio
Hidalgo
Jim Wells
La Salle
Live Oak
Maverick
Starr
Webb
Willacy
Zapata
Z aval a
Total Harvested Area
Total Defoliated Area
Percent Defoliated
Harvested
km2
2.4
414.4
4.5
371.5
24.3
8.7
1.6
9.7
225.4
13.4
1,075.9
753.1
area,1*6
(acres)
(600)
(102,400)
(1,100)
(91,800)
(6,000)
(2,150)
(400)
(2,400)
(55,700)
(3,300)
(265,850)
(186,100)
70% (mostly defoliated)23
Note: Blanks indicate less than 2.0 km2 (500 acres).
56
-------
Table 16 (continued). COTTON ACREAGE HARVESTED AND
DEFOLIATED IN TEXAS, 1971
Trans-Pecos region
Counties
Culberson
Ector
El Paso
Hudspeth
Jeff Davis
Pecos
Presidio
Reeves
Total Harvested Area
Total Defoliated Area
Percent Defoliated
Harvested
km2
11.7
48.6
29.7
25.9
5.1
103.2
224.2
0%23
area,"6
(acres)
(2,900)
(12,000)
(7,350)
(6,400)
(1,250)
(25,500)
(55,400)
Coast Prairie region
Counties
Brazoria
Calhoun
Fort Bend
Harris
Jackson
Liberty
Matagorda
Victoria
Wharton
Total Harvested Area
Total Defoliated Area
Percent Defoliated
Harvested
km2
35.6
6.5
194.3
6.1
25.9
6.1
34.0
13.0
213.3
534.6
267.3
area,1*6
(acres)
(8,800)
(1,600)
(48,000)
(1,500)
(6,400)
(1,500)
(8,400)
(3,200)
(52,700)
(132,100)
(66,050)
50% (mostly defoliated) 23
Note: Blanks indicate less than 2.0 km2 (500 acres).
57
-------
Table 16 (continued). COTTON ACREAGE HARVESTED AND
DEFOLIATED IN TEXAS, 1971
Rolling Plains and Central Basin
Counties
Archer
Baylor
Bordon
Callahan
Childress
Coleman
Collingsworth
Cottle
Dickens
Don ley
Fisher
Foard
Gcirza
Hcill
Ho.rdeman
Ha.skell
Jones
Kent
King
Knox
Mitchell
Motley
Nolan
Runnels
Scurry
Stonewall
Taylor
Wheeler
Harvested
km2
5.3
53.8
81.3
13.0
170.0
19.4
155.0
176.8
132.3
72.4
244.0
35.6
156.6
319.3
40.1
408.7
315.7
46.5
32.8
153.0
198.3
104.0
152.2
195.5
204.4
69.2
48.6
49.8
region
area,46
(acres)
(1,300)
(13,300)
(20,100)
(3,200)
(42,000)
(4,800)
(38,300)
(43,700)
(32,700)
(17,900)
(60,300)
(8,800)
(38,700)
(78,900)
(9,900)
(101,000)
(78,000)
(11,500)
(8,100)
(37,800)
(49,000)
(25,700)
(37,600)
(48,300)
(50,500)
(17,100)
(12,000)
(12,300)
58
-------
Table 16 (continued). COTTON ACREAGE HARVESTED AND
DEFOLIATED IN TEXAS, 1971
Rolling Plains and Central Basin region (continued)
Counties
Wichita
Wilbarger
Total Harvested Area
Total Defoliated Area
Percent Defoliated
Harvested area,46
km2
21.9
119.8
3,795.2
2,467.0
65% (mostly
(acres)
(5,400)
(29,600)
(937,800)
(609,600)
desiccated) 23
Grand Prairies region
Counties
Bell
Bosque
Comanche
Cooke
Coryell
Denton
Hamilton
Hill
Johnson
Tarrant
Williamson
Total Harvested Area
Total Defoliated Area
Percent Defoliated
Harvested area,46
km2
171.2
10.9
14.6
27.9
42.9
16.2
347.6
108.1
24.7
271.1
1,035.2
1,011.5
95% to 100% (all
Note: Blanks indicate less than 2.0 km2 (500
(acres)
(42,300)
(2,700)
(3,600)
(6,900)
(10,600)
(4,000)
(85,900)
(26,700)
(6,100)
(67,000)
(255,800)
(249,950)
desiccated) 2 3
acres) .
tt6Texas Cotton Review, 1973-74. Natural Fibers Economic
Research. University of Texas at Austin. Research
Report No. 104 (PB 235 388). July 1974. 143 p.
59
-------
For Arizona, very specific agricultural usage data were
obtained and emission estimates were made by simply multiply-
ing the quantities used by the appropriate emission factor
for each chemical; Table 17 presents these fine data.
Table 17. AGRICULTURAL USE OF DEFOLIANTS AND
DESICCANTS IN ARIZONA, 197I47
Material
Arsenic acid
DBF and Folex
Chlorates and borates
Paraquat
27.
49.
744.
12.
kg
9
4
9
6
Quantity
X
X
X
X
10
10
10
10
3
3
3
3
used
(
i
(1
61.
(108.
(1,642.
(
27.
b1
5
9
4
7
X
X
X
X
10
10
10
10
3)
3)
3)
3)
Courtesy of G. W. Ware, C. H. Kreader, L. Moore, Progressive
Agriculture, and the University of Arizona.
Applications of harvest-aid chemicals were assumed to occur
only once, although multiple applications are known to occur,
and mixtures of the chemicals are sometimes applied (e.g.,
small amounts of paraquat added to DBF formulations). Only
California (0.8%), Mississippi (0.5%), and Texas (2.2%) have
farms that reported1*3 treating crops with defoliants three
or more times. Since it was impossible to obtain data on
the use of mixtures, such use was taken to be negligible.
Emission estimates for cotton defoliants and desiccants by
states and for the U.S. are shown in Table 18. Texas was
the largest contributor to arsenic acid emissions (96.1%)
since its use is predominant there; Arkansas (21»7%) and
Mississippi (29.2%) were the largest contributors to national
47Ware, G. W., C. H. Kreader, and L. Moore. Agricultural
Use of Pesticides in Arizona. Progressive Agriculture.
University of Arizona. Tucson. July-August 1974.
p. 12-13, 16.
60
-------
Table 18. EMISSION ESTIMATES FOR COTTON DEFOLIANTS AND DESICCANTS
BY STATE AND NATIONWIDE, 1971
State
Alabama
Arizona
Arkansas
California
Florida
Georgia
Kentucky
Louisiana
Mississippi
Missouri
New Mexico
North Carolina
Oklahoma
South Carolina
Tennessee
Texas
TOTALS
Arsenic
kg
49
170
40
10
352
37
16,171 (
16,829 (
acid,
(lb)
(107)
(375)
(89)
(22)
(776)
(81)
35,656)
37,108)
DBF or
kg
2,301
494
4,963
670
59
1,908
6
1,935
6,677
459
15
496
31
1,721
554
574
22,865
Folex,
(lb)
(5,074)
(1,089)
(10,943)
(1,477)
(131)
(4,208)
(14)
(4,266)
(14,723)
(1,013)
(34)
(1,094)
(69)
(3,795)
(1,221)
(1,266)
(50,417)
Sodium
kg
5,272
478
25,597
186
643
229
51
83
57
405
33,000
chlorate ,
(lb)
(11,624)
(1,054)
(56,441)
(411)
(1,417)
(505)
(112)
(184)
(125)
(892)
(72,765)
Paraquat,
kg (lb)
126 (277)
1,264 (2,787)
1,390 (3,064)
CTi
Note: Blanks indicate values are negligible.
-------
DEF emissions; California had the largest amount of sodium
chlorate emissions (77.6%); and nearly all of the paraquat
used for cotton desiccation was used in Texas.
That the accuracy of the emission estimates is questionable
can be demonstrated by the following anomaly in defoliant
usage delta. The estimated defoliant usage for Washington,
Bolivar, and Sunflower counties in Mississippi (which contain
34% of the state's cotton acreage) has been reported;34 the
us£ of defoliants in those countries was 76% to 96% sodium
chlorate, the remainder was DEF. In Table 15 the estimated
defoliant use was 90% DEF and the remainder was sodium
chlorate, for Mississippi and other states. Clearly, the
choice of which data to believe can affect estimates greatly.
62
-------
SECTION V
CONTROL TECHNOLOGY
A. STATE OF THE ART
In general, drift hazard can be reduced by increasing the
droplet size of agricultural sprays, but if coverage is also
an important factor, it may then be necessary to increase
the total volume applied per unit area. Major efforts to
reduce drift hazard by reducing the number of fine drops for
a given application have utilized one of the following
approaches:
• Production of a more uniform droplet size — attempts
have been made to improve the uniformity by altering
the liquid properties as well as by changing nozzle
design and operating conditions. This would allow the
mean diameter to be maintained; although a perfectly
uniform spray may not be desired, a major reduction in
the number of fine drops would reduce the drift and
would probably improve coverage efficacy, resulting in
a lower application rate.
• Removal of the fine droplets — this approach utilizes
present types of atomization equipment and liquids but
attempts to remove the fine drops by coalescence or by
physical forces.
• Increase in the drop size spectrum — use of larger
drop size spectra generally results in a reduction in
the number of fine drops that may drift. In this case
the total applied volume may need to be increased to
maintain satisfactory coverage.7
Various methods of controlling drop size, proper timing of
application, and modification of equipment are practices
which can reduce drift hazards; they are discussed below.
63
-------
1. Fluid Additives
One possible avenue for reducing drift from spray applica-
tions is the use of adjuvants that alter the physical proper-
ties of the fluid. Some physical properties can affect the
basic atomization process and thereby reduce the number of
fine drops (less than 100 ym). Several commercial adjuvants
or formulations have been introduced that have a marked
effect on the viscosity, surface tension, and/or viscoelastic
properties of the fluid. However, few data are available on
the effectiveness of different adjuvant properties for re-
ducing drift under various field conditions.k8
One method of increasing viscosity is the use of a water-in-
oil or "inverted" emulsion. Such inverts have been shown to
reduce drift under many conditions for insecticidal sprays,
but they may be limited to use with phenoxy-acid herbicides
where good coverage is not necessary. They also have the
disadvantages of being unstable, of shifting rather than
narrowing the drop spectrum, and of increasing the phytotox-
icity of the emulsion. Economically, they compare favorably
with other spray thickeners available, but present more of a
logistics problem.1*9
There are many materials that increase the apparent viscosity
of sprays and, hence, reduce drift when properly added to the
mixture. Some of the materials that have recently been
48Yatesl, W. E., N. B. Akesson, and D. Bayer. Effects of
Spray Adjuvants on Drift Hazards. (Paper No. 74-1008,
presented at the 1974 Annual Meeting, American Society
of Agricultural Engineers. Stillwater, Oklahoma.
June 23-26, 1974.) 26 p.
£*9Butler, B. J., N. B. Akesson, and W. E. Yates. Use of
Spray Adjuvants to Reduce Drift. Transactions of the
American Society of Agricultural Engineers.
12^(2) :182-186, 1969.
64
-------
introduced are: Dacagin/ a mixture of natural carbohydrates50
(Diamond Shamrock); Norbak, a crosslinked polyacrylate50
(Dow); Vistik, a hydroxyethyl-cellulose50 (Hercules);
Nalcotrol, a polyvinyl polymer48 (Nalco Chemical); and
Cab-0-Sil, a submicroscopic fumed silica (Cabot Corp.).
These can all be added to water-based sprays, and have been
used almost exclusively with herbicides.49
These materials require care in mixing, and it is known that
Dacagin, Norbak, and Vistik lose viscosity with increasing
concentrations of salt. This means that changes in water
source, pesticide-water ratios, and pesticide types will
change viscosities. Therefore, consistency checks in the
field are desirable, with the amount of adjuvant used being
changed accordingly. The mixtures require time to increase
in viscosity, so waiting times of at least 20 minutes before
application are necessary, and longer in cooler weather.49
Water-based sprays are normally quite similar to water in
viscosity, and are Newtonian in their reaction to the in-
creasing shear rates encountered during passage through a
spraying system. The materials mentioned above are non-
Newtonian and pseudoplastic in their behavior. This tendency
to decreasing viscosity with increasing shear rate requires
the use of a highly viscous liquid in the spray tank, in
order that the liquid emitting from the nozzle will be a few
times more viscous than water.
In a typical spray system, shear rates are usually less than
50 s"1 in the tank, 500 to 1,000 s"1 in the lines, and
10,000 to 200,000 s"1 at the nozzle. The range in the
50Kanellopoulos, A. G. Additives in Herbicide Formulations.
Chemistry and Industry (London). 1974(9);951-955,
December 7, 1974.
65
-------
amount of adjuvant used, then, is limited on one end by the
ability of the system to move the highly viscous fluid from
the tank, and on the other by the need for high viscosities
at the nozzle to eliminate fine droplets.49
Other problems include the fact that the high air shear on
aircraft operated above 27 m/s (60 mph), even when nozzles
are directed with the slipstream, causes breakup of the
large drops: on ground equipment this is not a problem.
Recirculating the liquid through pumps can cause reduced
function of the viscous effect, also.51
Field tests have shown that, in addition to reducing drift,
Dacagi.n acts as a "sticker" material when added to a defoliant
spray of DBF and applied aerially. Tests with varying amounts
of adjuvant resulted in an 8% to 20% heavier leaf drop.52
Other field experiments were conducted with several adjuvants
to determine their effects on drift. In one comparison,
Dacagin, Cab-0-Sil, and blackstrap molasses were added to an
aerial pesticide spray.** ° All three decreased the downwind
drift; Cab-0-Sil was the most effective. Other researchers
compared fallout and air samples downwind from aerial appli-
cations of sprays containing an oil-water emulsion, a mixture
with Nalcotrol, and a mixture containing an experimental
hydroxyethyl cellulose buffer system called HEC/B (probably
similar to Vistik) for three types of atomization.^8 When
compared to the standard oil-water emulsion application, each
of the thickening adjuvants tested reduced the amount of
drift collected by the air samplers at all downwind stations.
51Akesson, N. B., W. E. Yates, and R. E. Cowden. What's
Happening in Aerial Application Research. Unpublished
paper. Agricultural Engineering Department, University
of California at Davis. 1975.
52Dacagin Speeds Cotton Defoliation. Agricultural Chemicals.
October 1969. p. 83.
66
-------
2. Nozzles and Atomizers
Low velocity jets have been introduced as a means of producing
a minimum number of fine drops. One approach to satisfactory
distribution with jets on a tractor-mounted boom sprayer in-
corporated jets directed backward, spaced at 63.5-mm (2.5-in.)
intervals, operated at 13.8 kPa to 27.6 kPa (2 psi to 4 psi),
and vibrated laterally at approximately 540 cycles/min.
Following the same approach, but without vibrating the boom,
a multijet nozzle was designed with an electrical drive unit
for each nozzle that produced a rotary oscillation of
0.436 rad (25°) at 4,000 cycles/min, to be operated with a
41.4-kPa (6-psi) spray pressure.7
When using one of the various thickened sprays for reducing
drift, the flat fan nozzle has a particular advantage because
the discharge coefficient remains nearly constant over a wide
range of viscosities.7
Field tests with a new type of nozzle, the Raindrop (Delavan
Manufacturing Co.), indicated that the spray measured as
downwind drift was reduced by approximately one-half that
resulting from applications with flooding, flat, or cone
nozzles under identical operating conditions.12
Conclusions drawn from the results of another drift study
conducted with a ground machine in the field were that
lowering the nozzle height decreased downwind drift deposits
and lowering the nozzle pressure decreased the spray loss.53
53Goering, C. E., and B. J. Butler. Paired Field Studies of
Herbicide Drift. (Paper No. 73-1575, presented at 1973
Winter Meeting, American Society of Agricultural Engineers,
Chicago. December 11-14, 1973.) 22 p.
67
-------
3. Equipment Modification
Ideally, the use of ground equipment instead of aircraft for
spray applications would reduce drift for cotton defoliation,
especially since ground operated air-carrier systems are not
needed., However, use of ground equipment is usually prohib-
ited by field conditions such as recent irrigation, height
or maturity of crop, lodging, broadcast or narrow-row
planting, and shortage of trained labor. It thus becomes
necesseiry to rely on aerial application. The utilization of
nozzles1, that minimize fine drops and the orientation of the
nozzle backwards into the airstream help achieve greater
on-tarcet deposits and reduced drift.
In studies of the air wake pattern from low-flying aircraft
it was observed that the fine spray droplets in the vicinity
of the wing tip were lifted high into the air to be carried
by whatever winds or thermal lifts existed. Because the
wing generates the wake, it was found that placing the boom
away from the wing reduced the movement of droplets to wing
vortices. 5l*
For ground machines, less drift can be achieved by using a
spectrum of large droplets and high volume application, and
by confining the spray closer to the target area. Hoods or
shields have been introduced to further reduce the drift for
specific hazardous applications. A simple deflector to con-
fine the trajectories11 and the use of an inflatable rubber
boom cover showed that drift was not eliminated, but under
strong winds, 4.0 m/s to 7.6 m/s (9 mph to 17 mph), it was
reduced by 53% to 89%.7
54Schultz, H. B., N. B. Akesson, W. E. Yates, and K. H,
Ingrebretsen. Drift of 2,4-D Applied by Plane.
California Agriculture., 10 (8) ;4-5,14, August 1956.
68
-------
4. Meteorological Timing
Loss of a defoliant chemical by drift is, among other things,
a loss of application efficiency. Applications that are
poorly timed or carelessly made during the "rush of the
season" are most likely to result in higher drift losses
which, in turn, generate the need for higher and more fre-
quent dosages than would be necessary under more efficient
methods.12 Here, the care of the applicator in observing
some simple meteorological parameters can minimize drift
hazards.
Since off-target drift deposits are greatest under stable,
inversion conditions, it has been recommended by researchers
that unstable meteorological conditions be chosen as often
as possible for aerial spray applications. Field studies
have demonstrated that the greatest on-target deposits are
achieved in the early morning, followed by midafternoon, then
early evening. The same studies showed that drift from
morning and afternoon applications was less than that from
evening application.55
The likelihood of good spraying winds (low velocity) is
greater in the early morning hours than in the evening; when
this situation is combined with the generally cooler and
more humid air conditions at this time of day, which lead to
reduced droplet evaporation and hence a reduction in the
potential drift fraction, the advantages of morning spraying
become obvious.* °
55Ware, G. W., B. J. Estesen, W. P. Cahill, and K. R. Frost,
Pesticide Drift. VI. Target and Drift Deposits vs. Time
of Applications. Journal of Economic Entomology.
65(4):1170-1172, August 1972.
69
-------
B. FUTURE CONSIDERATIONS
Much research has been directed toward reducing the drift
potential of herbicides; with sprays this has been concentrated
on increasing the coarseness of particles in the droplet
spectrum, controlling distribution of droplet sizes, and
increasing carrier viscosity. Additives which are pseudo-
plastic in behavior and invert emulsions have been studied,
and two opposing principles have been encountered. Droplets
coarse enough to reduce drift potential are highly desirable;
however, generation of very large drops may reduce herbicide
effectiveness by unevenly distributing the spray across
leaves. The optimum relationship would be the application
of unirform droplets of adequate coarseness to reduce spray
drift without reducing herbicide or defoliant effectiveness.
Presented below are some new developments which may achieve
increased effectiveness and reduce losses to the atmosphere
simultaneously.
1. Foam Spray Systems
Foam additives are the newest addition to the viscosity-
changing and sticking agent materials. Foam is a mixture of
liquids (adjuvant-aqueous phase) and gas (usually air) with
physical properties different from those of the original
constituents. Foams have potential for agricultural use as
evaporation suppressants, frost protection agents, soil
amendments, and pesticide carriers. They have been considered
as potential drift reduction agents because they can control
droplet coarseness.5 6
56Scifres, C. J., H. G. McCall, and D. W. Fryear. Foam
Systems as Herbicide Carriers for Range Improvement.
Texas Agricultural Experiment Station. Miscellaneous
. Publication 1974. 18 p.
70
-------
Present research indicates that foams have most of the
desirable characteristics and at least one of the undesirable
characteristics of thickeners. If drift reduction is to be
attained spray particles must be made large; small drops must
not be formed individually or allowed to disengage from the
larger foam clusters. When a foam is generated, it produces
a size range of bubbles which are held together by the
foaming agent in clusters of globules. As long as the foam
agent holds these together, little drift loss can occur.
But if the cluster should shed the small bubbles, as is
possible during aircraft application, drift losses can occur.
Because varying amounts of foam agent in a given mixture
produce varying degrees of bubble size, liquid content, and
stickiness or tenacity, further research is needed to
specifically evaluate the use of these three parameters as
potentials for drift reduction. If the amount of liquid or
air in each bubble is controllable, it could be adjusted to
control drop density and total amount of liquid applied.
Thus, the problem of having large drops with poor coverage
might be resolved by making the drops hollow, and the liquid
volume could then be reduced.57
Various types and designs of foam generators have been intro-
duced. Although these generators vary in construction,
several components are common to all: a nozzle body, an
orifice, and a chamber for foam formation (Figure 5). Some
generator types employ a detachable nozzle tip, whereas the
delivery port is constructed as part of the nozzle in others.
The greatest variation among generators is in the design of
the foam generator body. All are hollow tubes with a number
57Akesson, N. B., S. E. Wilce, and W. E. Yates. Confining
Aerial Applications to Treated Fields — A Realistic Goal
Agrichemical Age. December 1971. p. 11-14.
71
-------
STRAINER
ORIFICE
ADAPTER
NOZZLE
H
FOAM GENERATOR BODY
NOZZLE NUT
STANDARD MATERIALS-BRASS
Figure 5. Component parts of foam generators which mix
air and liquid to form foam56
of air inlet ports on the side; however, there are many
differences in size, number, location, and arrangement of
the ports and in size of the generator bodies. Plastic,
brass, and aluminum are used for construction.56
Experinents with foaming in a cotton defoliation program
resulted in improved visibility of the spray swath and
cleaner equipment but did not change the effectiveness of
defolicition from that provided by the conventional spray
method. Drift reduction was not studied. However, the
ability to see the areas where spray has been applied (there-
by reducing overlapping), better coverage of the plants from
the action of the spray adjuvant, and cleaner equipment
because' of the flushing action of the foaming agent can all
contribute to reduced total applications and more constant
droplet size ranges, which will reduce drift emissions.58
58Threadgill, E. D., and R. F. Colwick, Ground Applications
of Cotton Defoliants with Air Aspirating Nozzles. In:
Proceedings of the 27th Annual Beltwide Cotton Defoliation
and Physiology Conference. Phoenix. January 9-10, 1973.
p. 35.
72
-------
Field studies on the drift reduction effectiveness of foam
systems were conducted for both aerial and ground equipment.
Results gave little indication of reduced drift potential
for aerial sprays, but ground equipment runs indicated great
potential. It was postulated that air injected into the
foam generator changed particle densities to the extent that
flotation occurred, and potential for displacement increased.56
2. Microfoil®
Mechanically induced, constant droplet size sprays are
possible using the commercially available multiple hypodermic
needle Microfoil boom (Amchem Corp.) shown in Figure 6.59
The unit contains 3,120 capillary tubes on a 7.9-m (26-ft)
boom. Presently two sizes are available, 0.33 mm (0.013 in.)
and 0.71 mm (0.028 in.) in inside diameter, that are operated
at 14 kPa (2 psi) or less.7 This is as close to total drift
control as is presently possible with an estimated 98% to
99% recovery of spray in the applied swath. However, the
Microfoil cannot be used on aircraft at speeds greater than
26.8 m/s (60 mph), which limits it to helicopter operation
or ground rig use.57
Table 19 shows the drop sizes produced by a variety of atom-
izers. These range from completely airborne aerosols with a
volume median diameter (vmd) of 11 ym to the Microfoil which
produces about 99% of the drops in the 900-ym size with only
about 0.001% below 220 ym.
59Brazelton, R. W. Control of Chemical Drift. University
of California, Agricultural Extension. Bulletin No.
OSA #n5. July 1971. 2 p.
73
-------
Table 19. TYPICAL DROP SIZE DISTRIBUTION, CUMULATIVE PERCENT BY VOLUME BELOW SIZES SHOWN51
Drop size,
urn
1-5
5-10
11-um vmd
10-15
15-20
20-40
40-60
60-80
86-ym vmd
80-100
100-120
120-140
130-um vmd
140-180
180-200
200-220
220-240
240-260
260-280
280-300
278-ym vmd
300-350
350-400
400-450
460-ijm vmd
450-500
500-600
600-700
700-800
900-iim vmd
800-1000
vmd
Nozzle type
Fi n*>
aerosols
5
45
50
77
97
100
11 ym
Cold fogger,
34.5 kPa
(5 psi) air
and liquid
Co=-»e
aerosols
0.1
0.4
2.0
12
35
50
59
100
86 ym
2-Fluid,
206.8 kPa
(30 psi) air,
34.5 kPa
(5 psi) liquid
Fine sprays
0.1
2
5
15.8
50
81
100
130 ym
Spinner, in
40-45 m/s
(90-100 mph)
airstream
1
Medium sprays
0.1
2
6
17
46
50
92
100
278 urn
65015 Fan,
275.8 kPa
(40 psi) liquid
pressure at
1.57 rad (90°)
back, 40-45 m/s
airstream
Coarse sprays
0.01
0.1
0.4
3
7
14
24
36
46
50
55
74
88
96
100
460 ym
D6-46 Cone,
344.7 kPa
(50 psi) liquid
pressure ,
40-45 m/s
airstream
YcLy ouai. ac
sprays
0.001
0.1
5
15
25
50
100
900 urn
D6 Jet back.
275.8 kPa
(40 psi) liquid
pressure ,
40-45 m/s
airstream
Microfoil
0.001
0.01
0.1
1
98.88
100
Under 27 m/s
(60 mph) with
airstream
aThe vmd or volume median diameter is that size of drop which divides the total volume of drops found exactly in half; that
is, 50% of the volume is in drops above that size and 50% are below the vmd size. The size is measured in microns (urn).
-------
3930-1
Figure 6. Representation of the Microfoil used on
helicopters at 26.8 m/s airspeed or less
producing 800-ym to 1,000-ym drops59
3.
Thermal Defoliation
A completely different concept for defoliating or desiccating
cotton has been undergoing research in Oklahoma, where agri-
cultural engineers have been designing a machine to thermally
defoliate cotton. An acceptable machine for applying heat
to field crops has been developed and has gone through several
improvements and changes.
A two-row machine provides controlled airflow rates to main-
tain close control over the application temperatures of 422°K
to 588°K (300°F to 600°F) caused by fueling LP gas. The unit
consists essentially of two 2.7-m (9-ft) tandem units with a
0.3-m (1-ft) space between them, resulting in an overall heat
75
-------
unit 5.7 m (19 ft) in length, as shown in Figure 7. Spring-
loaded doors are positioned in front and behind the 5.7-m
unit or oven to enclose the heated air. The unit is attached
to a HL-tractor propelling unit.60
-5.7m
-2.7m-
FAN INLET
0.3
m
V
HOT AIR
RETURN
GAS BURNER
o
DIRECT I ON OF TRAVEL
-2.7m-
FAN INLET
XLP GAS BURNER
-HOT AIR
RETURN
Figure 7. Schematic of 1970 thermal defoliator60
Basic relationships to cause defoliation have been developed.
If a certain exposure time of the plants to a particular
temperature resulted in defoliation, increasing either the
exposure time or the temperature would be a more severe
treatment resulting in desiccation of the cotton instead of
defoliation. Three years' data were used to develop formulas
related to leaf drop and leaf kill:60
Leaf drop (%) = -19.29 +
+ O.llx2 - 0.01x^2 (5)
Leaf kill (%) = -29.96 + 13.81XJ + 0.14x2 - 0.01x^2 (6)
where
time of exposure = length of defoliator times
forward speed, s (1.2 <_ Xj <_ 5.5)
temperature, °F (200 <_ x2 <_ 700)
(nonmetric units used by Reference 60)
60Batchelder, D. G., J. G. Porterfield, and G. McLaughlin.
Thermal Defoliation of Cotton. In: Proceedings of the
25th Annual Beltwide Cotton Defoliation and Physiology
Conference. Atlanta. January 12-13, 1971. p. 36-37.
76
-------
A thermal defoliator owned by the Natural Gas Processors
Association was tested during the 1969 harvest season in the
lower Rio Grande Valley of Texas in July, continued in
Mississippi in September, and finished in Oklahoma in October.
At all locations, comparisons were made between chemical and
thermal defoliation, and cotton was subjected to fiber
analysis in addition to grade, staple, and micronaire. There
was a slightly higher net lint value for the thermally defol-
iated cotton which indicated essentially no difference in
fiber quality in favor of either chemical or thermal defolia-
tion.60
Cost of the fuel to cause defoliation was calculated to be
approximately $405/km2 ($1.64/acre). This figure does not
include any machinery costs, and is based upon an LP gas
estimate of $31.70/m3 (12<:/gal) , a January 1971 quote for
fuel bought in 1,000-gal quantities. A cost of $1.64/acre
to cause thermal defoliation was believed to be competitive
with chemicals for defoliation.60
In summary, thermal defoliation is believed to offer several
advantages for cotton as compared to chemical defoliation:
(1) costs are competitive, (2) thermal defoliation is positive
and is not affected by subsequent weather (no secondary
applications), (3) new and regrowth leaves are particularly
sensitive to thermal application, (4) thermal defoliation
does not result in changes in fiber properties if properly
applied, and (5) thermal defoliation does not result in any
residue or drift problems.60
77
-------
SECTION VI
GROWTH AND NATURE OF THE INDUSTRY
A. PRESENT AND EMERGING TECHNOLOGY
Any changes in technology in cotton defoliation and desicca-
tion, other than a major equipment technology transfer such
as introduction of thermal defoliators, will occur in the
usage of new and different harvest-aid chemicals.
The chemical industry will be required to establish toler-
ances, through feeding and toxicology studies, for all of the
currently available harvest-aid chemicals whose tolerances
are unknown. If this is not done, the cotton industry will
have only three chemicals available for this type of use:
Folex for defoliation, and arsenic acid and paraquat for
desiccation.61
The usa of cacodylic acid and sodium cacodylate (Bollseye)
has bean increasing in recent years as a substitute for
arsenic acid.2tf'25 The cacodylates are arsenic-based com-
pounds similar in structure to arsenic acid but with LD50's
(oral, male rat) ten times higher; they are thus much less
toxic. Usage of cacodylic acid (dimethylarsenic acid) in
Arizona, which has the most complete pesticide usage figures,
6 Cotton Growers Spent $60 Million for Herbicide in '66,
Shaw Tells Cotton Mech Conference. Farm Chemicals.
L3CK2) .-80-82, February 1967.
78
-------
has risen from 3.17 metric tons and 1.86 metric tons in 1970
and 1971, respectively, to 10.9 metric tons and 31.6 metric
tons in 1972 and 1973, respectively. Meanwhile, arsenic
acid use declined 29.8 metric tons from 1970 to 1973.k7
New patented chemicals for cotton harvest-aid utilization
include cis-2,3,5,5,5-pentachloro-4-keto-2-pentanoic acid,62
3-amino-3-carboxypropylmethylsulfoximine salts,63 derivatives
of dialkyl arsinic acids, OAs(R)(R1)(OR2), where R and R1 are
Ci-4 alkyl and R2 is H, NH^, Na, etc.,64 and substituted
triphenyl phosphates and phosphites.65
The discovery of the plant hormone abscisin II has been
heralded as a step forward. In cotton, abscisin II causes
leaf or flower shed. It might be usable as a biological
defoliant that would be effective in all weather conditions
and on all stages of plant maturity. Unfortunately, it took
about 225 kg (500 Ib) of cotton bolls to isolate and crystal-
lize a minute amount of abscisin II (9 mg).61'66
62Erby, W. A., W. E. Erner, J. S. Skaptason, and R. A.
Walde. Defoliation and Desiccation of Cotton with
cis-2,3,5,5,5-Pentachloro-4-keto-2-pentanoic Acid.
U.S. Patent 3,472,004 (to Air Products and Chemicals,
Inc.), October 14, 1969.
63Walworth, B. L. 3-Amino-3-carboxypropylmethylsulfoximine
Salts as Nonselective Water-Soluble Defoliants. U.S.
Patent 3,323,895 (to American Cyanamid Co.), June 6, 1967,
6ltNeuville, M. L., and R. B. Carroll. Cacodylic Acid
Plant Defoliants. U.S. Patent 3,378,364 (to Ansul Co.),
April 16, 1968.
65Hensel, J., and D. W. Gier. Defoliating and Desiccating
Plants with Substituted Triphenyl Phosphates and
Phosphites. U.S. Patent 3,416,911 (to Chemagro Corp.),
December 17, 1968.
66A Natural Defoliant. Agricultural Research. 14(5);11,
November 1965.
79
-------
One of the chief problems of defoliation is the waiting
period of 5 to 14 days after application for completion of
the defoliation action. Thus, the management aspects of
conventional defoliation and harvesting procedures are made
difficult by unpredictable weather conditions between the
time the chemical is applied and the time of harvest. The
effects of defoliants, in certain cases, can be totally
offset by rapid production of new-growth leaves if rains
occur between application and harvest.67
Another possible chemical change could be the introduction
of wilting agents, such as neodecanoic acid, instead of
defoliants. A new system or technique has been investigated
in which the mechanical picking was done while the leaves
were in a chemically wilted condition and still attached to
the plant. This system was given the name, "wilt-harvest."67
The wilted condition is produced by a wiltant — a chemical
that Cciuses rapid wilting of the leaf blades within a few
hours of application. The action on the blades is similar
to that of a desiccant; however, the leaf petioles are not
injured by the wiltant. Therefore, the leaves will defoliate
in the same manner as with conventional chemical defoliants,
given enough time.6 7
The principal objective of the new approach is to provide
more precise control over the harvesting operations during
the early part of the season. The development of a success-
ful wiltant will provide a means by which the producers can
utilize the more accurate short-range weather forecasts and
67Miller, C. S., L. H. Wilkes, E. L. Thaxton, and J. L.
Hubbard. Cotton Wilt-Harvest and Wiltant Defoliation
Effectiveness in Texas. Texas Agricultural Experiment
Station. Miscellaneous Publication No. MP-1010.
October 1971. 12 p.
80
-------
harvest within a matter of hours (3 to 48) after chemical
treatment. It is predicted that the maximum benefit will be
achieved where smaller acreages are treated in advance of
the pickers to take advantage of harvesting while the leaves
are in the proper condition.67
The defoliant action of a 30% formulation of neodecanoic
acid compared favorably to that of other commercial defoliants,
which indicates that a conventional defoliation picking may
be made in case the wilt-harvest picking is not properly
timed.67
B. INDUSTRY PRODUCTION TRENDS
Little, if any, growth is forecast for the amount of area to
be planted in cotton in the near future (to 1978). This is
due to strong competition from foreign growers and from
synthetic fiber producers. Figure 868 and Table 20 illustrate
the historical variability of U.S. cotton acreage harvested,
which is returning from the disastrous years of 1966-67 to
a more stable position.
The extent of harvest-aid chemical use has remained constant
since becoming widespread in 1960, and is illustrated in
Table 21. Emissions are proportional to acreage harvested
and extent of chemical use, so emissions from agricultural
spraying of cotton defoliants and desiccants are expected
to remain constant (1972 to 1978).
681973 Handbook of Agricultural Charts. Washington, U.S
Department of Agriculture, October 1973.
81
-------
80
60
40
1957
'60
'63
'66
YEAR BEG INNING AUGUST 1
* ESTIMATED
Figure 8. U.S. cotton acreage, yield, and production61
Table 20. U.S. COTTON ACREAGE, YIELD, AND PRODUCTION, 1947-7368
Year
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
Harvested
acreage,
1,000 (1,000
km2 acres)
86.3
92.7
111.0
72.2
109.1
104.9
98.5
77.9
68.5
63.2
54.9
48.0
61.2
62.0
(21,330)
(22,911)
(27,439)
(17,843)
(26,949)
(25,921)
(24,341)
(19,251)
(16,928)
(15,615)
(13,558)
(11,849)
(15,117)
(15,309)
Year
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973d
Harvested
acreage,
1,000 (1,000
km2 acres)
63.3
63.0
57.5
56.9
55.1
38.7
32.4
41.1
44.7
45.1
46.4
5-2.5
50.2
(15,634)
(15,569)
(14,212)
(14,057)
(13,615)
(9,552)
(7,997)
(10,160)
(11,055)
(11,155)
(11,471)
(12,984)
(12,406)
Preliminary. August 1 estimate,
82
-------
Table 21. CHANGES IN USE OF HARVEST-AID CHEMICALS
FOR COTTON
Year
1952
1955
1958
1960
1964
1971
Total area treated,
km2 (1,000 acres)
10,161 (2,510.8)69
11,946 (2,951.9)69
19,012 (4,697.9)69
29,046 (7,177.3)69
17,191 (4,248.0)70
22,642 (5,595.0)1*5
Total area harvested,68
km2 (1,000 acres)
104,900 (25,921)
66,584 (16,453)
47,814 (11,815)
61,735 (15,255)
56,887 (14,057)
46,422 (11,471)
Percent
treated
9.7
17.9
39.8
47.0
30.2
48.8
69Saunders, J. M. , and H. R. Cams. The Usage of Harvest-
Aid Chemicals, 1952-1960. In: Proceedings of the 16th
Annual Beltwide Cotton Defoliation and Physiology
Conference. Memphis. January 9-10, 1962. p. 8-12.
70Eichers, T., P. A. Andrilenas, R. Jenkins, and A. Fox.
Quantities of Pesticides Used by Farmers in 1964. U.S.
Department of Agriculture. Washington. Agricultural
Economic Report No. 131. January 1968. 37 p.
83
-------
SECTION VII
APPENDIXES
A Derivation of Source Severity and Input Data
B Preliminary Air Sampling of Cotton Desiccation
C Method for Estimating TLV Values for Compounds
When None Exists
84
-------
APPENDIX A
DERIVATION OF SOURCE SEVERITY AND INPUT DATA
1. DEFINITION OF SOURCE SEVERITY
The behavior of emission "plumes" from agricultural spraying
operations is different from that of plumes from elevated
stacks in the following respects: (1) the emissions from
a spray run are a ground level line source rather than an
elevated point source; and (2) the emissions are instan-
taneous and intermittent rather than continuous . Because of
these differences, the source severity, S, used to indicate
the hazard potential of an emission source cannot be used
"as is" for comparison purposes in this case. The source
severity, S, is defined as:
S = (A-l)
where 7 is the time-averaged maximum ground level con-
IT13.X
centration of each pollutant emitted from a continuous
near-point source, and F is the primary ambient air quality
standard for criteria pollutants and is a "corrected" thres-
hold limit value (i.e., TLV • 8/24 • 1/100) for noncriteria
pollutants.
85
-------
An alternative source severity to be used for the special
case of agricultural field spraying shall be defined
(Equation 1) as:
s = xl
A FA
where ;< is the average ground level concentration during
spraying at the field perimeter (where maximum exposure to
a population exists), and F is a "corrected" threshold
£\
limit value (i.e., TIM • 1/100).
The source severity as given in Equation A-l was developed
on the basis of the ratio of the dose of the pollutant
deliveired to a population relative to some potentially
hazardous dose for a specific time of interest.71 For
criteria pollutants the potentially hazardous dose is the
primary ambient air quality standard times the appropriate
averaging time. The dose delivered is, then, the concentra-
tion maximum times the same averaging time. Application of
chemicals to a field crop occurs in a time period of 8 hours
or less and dose from airborne drift from the application is
consequently 8 hours or less.
The hazard factor, F, in Equation A-l was used for noncriteria
pollutants to compensate for the fact that TLV's were estab-
lished for an 8-hr/day, 5-day work week exposure, and that
the general population is a higher risk group than healthy
workers. Hence, the multiplication by 8/24 corrects for
continuous exposure and the multiplication by 1/100 is a
safety factor. In the alternative hazard factor, F , there
71Eimutis, E. C. Source Assessment: Prioritization of
Stationary Air Pollution Sources, Model Description.
Monsanto Research Corporation. Dayton. Report No.
MRC-DA-508. Environmental Protection Agency,
EPA-600/2-76-032a. February 1976. 77 p.
86
-------
is no need to correct for continuous exposure since it does
not apply to this case, but the safety factor of 1/100 is
retained because it does apply to risk to the general
population.
Determining the ground level concentration, x"/ during
spraying requires the use of a dispersion model. The source
severity, S, used the Gaussian plume equation for maximum
ground level concentration as emitted by a continuous,
elevated point source:72
*max =
where Q = emission rate
u = wind speed
h = effective emission height
IT = 3.14
e = 2.72
For the case of an agricultural spraying operation the
"plume" emitted can be assumed to be an instantaneous line
source. The Gaussian dispersion model which describes this
is: 72
X = — exp -4IZ=2H\ exp -±HH (A-3)
rt IT rr rt *- \ J I rr II •*• I y 1 rf II
where QT is the total amount of material emitted per unit
. Jj
length from a line source, and a ... and a ... are the standard
deviations of the distribution of material in a puff in the
x- and z-directions, respectively. The above equation des-
cribes emissions from only one spraying pass with respect to
72Meteorology and Atomic Energy 1969. Slade, D. H. (ed.)
U.S. Atomic Energy Commission. (NTIS TID-24190).
July 1968. 445 p.
87
-------
time. The maximum ground level concentration at distance x
occurs at the second at which T = x/u, which causes the
exponential term containing time, T, to be unity.
The problem in using Equation A-3 is that in actual spraying
numerous passes are made, each one at a different distance
from the receptor at field's edge. A simpler form of the
same model uses the exposure or dosage from an instantaneous
line source:72
exp-iMi- (A-4)
where DT is the dosage from the line puff, or concentration,
LI
multiplied by time» If the time of emitting is known or
estimated, then the average concentration during that time
can be computed. The standard deviation of the distribution
of material in the vertical direction can be estimated from
power law functions of downwind distance. For neutral
atmospheric stability:72
azl = 0.15x°'70 (A-5)
An idealized representation of the method used to calculate
X is shown in Figure A-lo A square field with sides of
length B is aligned orthogonally to the wind. The spraying
swath is in the center of the field perpendicular to the
•D
wind direction and. at a distance D (=-j) from the receptor,
or affected population. The following simplifying assumptions
are made:
• The average distance of swaths to receptor is taken to
be D = B/2, or, every swath is made in the center of
the field.
• The emission from n swaths of QL each is taken to be
n QL- This is emitted from one swath at distance D
from the receptor.
88
-------
The total time, t, of dosage from emissions is taken
to be the total spraying time plus the time needed
for turning the spray equipment.
U.S. average meteorological conditions prevail
(u = 4.47 m/s, neutral stability).
The effective height of emission is negligible (h = 0)
SPRAY SWATH I ^
RECEPTOR
Figure A-l. Representative field for agricultural spraying
The total dosage due to this instantaneous line source,
is then:
1/2 nQ
(A-6)
Substituting for u and a
,1/2
= (!)
nQT
(4.47) (0.15)D°-70
(A-7)
or
x =
1.19 • n • Q
t • D°-70
(A-8)
Substituting the above value for x and the factor TLV • 1/100
for F into Equation 1 gives Equation 4:
f\
119 • n • QT
A t • TLV • D°-70
where n = number of passes or swaths made in the
representative field
QT = emissions in mass/length for a single spray pass
LI
89
-------
t = time to complete spraying representative field
including turning time
TLV = threshold limit value of material being sprayed
D = distance from center of field to field boundary
2. DEFINITION OF REPRESENTATIVE SOURCES
There are three major cotton growing regions in the U.S.,
each with different climate, soil type, and cotton varieties.
Figure 3 shows that these regions can be identified as the
Delta belt (encompassing the "Mississippi River valley in
Arkansas, Mississippi, and Louisiana), the Blacklands belt
in Texas (running roughly from Austin to Paris), and the
High Plains - Low Rolling Plains belt in Texas (located in
and below the panhandle).
Approximately 12,000 km2 (3 million acres) of cotton are
grown and defoliated in the Delta belt, where sodium chlorate
and DBF or Folex (tributylphosphorotrithioates) are most
commonly used.34 The Blacklands have about 4,000 km2
(1 million acres) of cotton which is desiccated, using a
ground rig, with arsenic acid.25 The major part of Texas
cotton acreage is on the High Plains centering at Lubbock.
Depending on the weather, most of this 12,000 km2 is not
treated for harvest but killed by frost. Approximately
4,000 km2 plus or minus 4,000 km2 are normally desiccated by
aircraft using mainly paraquat.25
From these three cotton producing regions, four representative
sources were defined: (1) paraquat application by aircraft
in the High Plains of Texas, (2) arsenic acid application by
ground rig in the Blacklands of Texas, (3) sodium chlorate
application by aircraft in the Delta, and (4) DEF application
by aircraft in the Delta.
90
-------
Representative sizes of cotton farms were determined by
analyzing the latest agricultural census data.43 Total
reported acres for the states of Arkansas, Louisiana,
Mississippi, and Tennessee divided by the number of farms
reporting yields an average-size cotton farm of 0.70 km2
(173 acres). The acreage of farms reporting represents
about 69% of total acreage harvested for these states. The
size of the representative cotton farm in the Blacklands area
of Texas was determined by dividing the number of farms into
the reported cotton acres harvested for the following
counties: Bowie, Collin, Hunt, Delta, Dallas, Kaufman,
Johnson, Ellis, Hill, Navarro, McClennan, Bell, Falls,
Williamson, Milam, Robertson, and Burleson. The representa-
tive cotton farm for this region is 0.61 km2 (150 acres).
For the High Plains the following counties were included:
Collingsworth, Farmer, Castro, Swisher, Briscoe, Hall,
Childress, Bailey, Lamb, Floyd, Motley, Cottle, Hardeman,
Wilbarger, Cochran, Hockley, Lubbock, Crosby, Dickens,
Yoakum, Terry, Lynn, Garza, Haskell, Gaines, Dawson, Scurry,
Fisher, Jones, Martin, Howard, Mitchell, Nolan, and Midland.
The representative cotton farm for this region is 1.05 km2
(260 acres).
3. INPUT DATA FOR CALCULATION OF S
The average ground level concentration, x/ of pollutants
resulting from agricultural pesticide application is
described in Section IV.D.2 of this document. The method
for estimating TLV's for pollutants which have no established
TLV is described in Appendix C. Input data used to calculate
the source severity, S , for the four representative cotton
£\
farms are provided below, based on:
- = / 2N1/2 QL
x \tr/ ua__t
zl
91
-------
a. Paraquat - Aerial Application
Representative field = 260 acres = 1,052,194 m2
Assume application swath = 50 ft = 15.24 m
Assume application speed = 80 mph = 35.76 m/s
Area application rate = (15.24)(35.76) = 545 m2/s
Time of spraying = (1,052,194)7(545) = 1,931 s
t = total application time (add 200% for turning) = 5,793 s
Application rate30 = 0.5 Ib/acre = 0.056 g/m2
Number of swaths = /I,052,194/15.24 = 67
D = distance to receptor = /I,052,194/2 = 513 m
TLV of paraquat29 = 0.0005 g/m3
*
Q = (application rate)(swath)(1.0%)a = 8.53 x 10~3 g/m
119 • n • QT
S =
FA t • TLV • D°*70
Therefore, SA - (H9) (67) (8. 53 x 10"3) . 0>3fl
A (5,793) (0.0005) (513°-70)
b. Arsenic Acid - Ground Rig Application
Representative field = 150 acres = 607,035 m2
Assume application swath = 32.8 ft = 10 m
Assume application speed = 12 mph = 5.36 m/s
Area application rate = (10)(5.36) = 53.6 m2/s
Time of spraying = (607,035)/(53.6) = 11,325 s
t = total application time (add 10% for turning) = 12,458 s
Application rate30 =4.4 Ib/acre = 0.49 g/m2
1.0% is the amount assumed lost as airborne drift.
92
-------
Number of swaths = /607,035/10 =78
D = distance to receptor = /607, 035/2 = 390 m
TLV of arsenic acid = TLV of inorganic arsenic29 = 0.0005 g/m3
Amount lost as airborne drift (from Appendix B) = 0.61 ± 0.29%
QL = (application rate) (swath) (0. 61 ± 0.29%)
= 3.0 x 10~2 ± 1.4 x 10~2 g/m
Therefore, S (119) (78? (3.0 x IP"2)
(12,458) (0.0005) (390°-70)
c. Sodium Chlorate - Aerial Application
Representative field = 173 acres = 700,114 m2
Assume application swath = 50 ft = 15.24 m
Assume application speed = 80 mph = 35.76 m/s
Area application rate = (15.24) (35.76) = 545 m2/s
Time of spraying = (700, 114) / (545) = 1,285 s
t = total application time (add 200% for turning) = 3,855 s
Application rate30 =5.0 Ib/acre = 0.56 g/m2
Number of swaths = /700,114/15.24 = 55
D = distance to receptor = /700,114/2 = 418 m
TLV of sodium chlorate = 0.0198(LD50)°•77k mg/m3
Acute oral-rat LD50 sodium chlorate28 = 1,200 mg/kg
TLV = 0.0048 g/m3
Q_ = (application rate)(swath)(1.0%) = 8.53 x 10~2 g/m
Li
Therefore, S, = (119)(55)(8.53 « 10"') . 0.44
A (3,855) (0.0048) (418°-70)
d. DBF - Aerial Application
Representative field = 173 acres = 700,114 m2
Assume application swath = 50 ft = 15.24 m
93
-------
Assume application speed =80 mph = 35.76 m/s
Area application rate = (15.24)(35.76) = 545 m2/s
Time of spraying = (700,114)/(545) = 1,285 s
t = total application time (add 200%) = 3,855 s
Application rate30 = 1.5 Ib/acre = 0.17 g/m2
Number of swaths = /700,114/15.24 = 55
D = distance to receptor = /700,114/2 = 418 m
TLV of DBF = 0.0198(LD50)°-771* mg/m3
LD50 acute oral-rat for DBF28 = 150 mg/kg
TLV = 0.00096 g/m3
QL = (application rate)(swath)(1.0%) = 2.59 x 10~2 g/m
Therefore, S, = U19) (55) (2.59 x IP"2) = Q>67
(3,855) (0.00096) (418°-70)
94
-------
APPENDIX B
PRELIMINARY AIR SAMPLING OF COTTON DESICCATION3
Atmospheric sampling of airborne drift losses due to arsenic
acid spraying for cotton desiccation was conducted in the
Blacklands area of Texas in late summer of 1975. The purpose
of this preliminary sampling effort was to quantify the drift
losses, within an order of magnitude, because no prior data
existed. Three cotton fields were sampled, each at a
different farm, one run per day.
1. MATERIALS AND METHODS
a. Field Descriptions
Field A consisted of two patches of cotton separated by a
grass strip. The west (first sprayed) patch was about
0.263 km2 (65 acres) and the east (second sprayed) was
0.445 km2 (or 110 acres) in size. Air samplers were located
in the grass strip and at the north end of the east patch.
Wind prevailed from the southeast.
Field B was 0.065 km2 (16 acres), ^200 m x 325 m, with air
samplers located at the west end. Wind was from the southeast,
Metric or nonmetric units are shown for some calculations
in this Appendix, depending on the.type of units that were
used for the particular data during the preliminary sampling;
metric units are provided for calculated results; units used
for calculation of drift are immaterial since these results
are reported as a percent.
95
-------
Field C was a 0.16-km2 (40-acre) patch of cotton, ^500 m x 325 m,
with samplers located at the northeast corner, one set on the
north edge and the other on the east edge. Wind prevailed '<•
from the south with easterly gusts.
b. Application Equipment
At fields A and B, two John Deere Hi-Boys, each equipped
with seven 6.35-mm (1/4-in.) KCL SS5 hollow cone nozzles
facing upwards on a boom, applied 0.7 m3/km2 (3/4 gal/acre)
of Desiccant L-10 (orthoarsenic acid, I^AsO^, 75% by weight)
mixed with a small amount (0.25%) of surfactant and diluted
(93%) with water. Total spray was about 9.05 m3/km2 (9.68
gal/acre), with a swath width of 9.14 m (30 ft) at a height
of 1.5 m to 1.8 m (5 ft to 6 ft) and a speed of 5.4 m/s (12
mph). Approximately 95 swaths were made during application
at field A and approximately 32 swaths were made at field B.
For fieild C, one John Deere Hi-Boy, equipped with 28 D-4-45
spinner hollow cone nozzles facing down on a boom, applied
0.7 m3/km2 (3/4 gal/acre) of Sinergized H-10 (orthoarsenic
acid, HsAsO^, 75% by weight) mixed (95%) with water at a
height of 1.5 m to 1.8 m (5 ft to 6 ft) and a speed of 5.4 m/s
(12 mph). Total spray was about 13.6 m3/km2 (14.59 gal/acre);
no surfactant was added. Swath width was 13.7 m (45 ft), and
approximately 22 swaths were made.
c. Meteorological Conditions
Application at field A was begun at 1:43 pm, August 30, 1975,
and completed at 4:15 pm. During this time, wind was from
the southeast, gusty, 0.89 m/s to 8.9 m/s (2 mph to 20 mph)
for the first hour and 0.89 m/s to 3.6 m/s (2 mph to 8 mph)
for the remainder of the afternoon. Temperature was 35°C
(95°F) and atmospheric stability was class C throughout the
sampling period.
96
-------
At field B, application began at 9:10 am, August 31, 1975,
and was completed at 9:30 am. Wind speed was 0 m/s to 1.8 m/s
(0 mph to 4 mph) from the southeast, temperature was 29°C
(85°F), and atmospheric stability was class B.
Field C application started at 2:23 pro, September 1, 1975,
and terminated at 5:00 pm. Wind speed was 0 m/s to 3.1 m/s
(0 mph to 7 mph) throughout the afternoon from the south with
easterly gusts. Temperature was 38°C (100°F) and stability
was class C.
d. Experimental Design
Two off-target collection stations were located downwind
from the spraying operations. Six air samplers were operated
at each of the stations, two of which were connected in series
and operated continuously to determine collection efficiency
and total dosage. The remaining four air samplers could be
remotely controlled by radio to sample either sequential drift
losses or drift from varying distances to the ground sprayers.
All samplers were operated at a height of 1.2 m (4 ft).
e. Air Samplers
At each collection station, six Smith-Greenburg impingers
were powered by two Cast rotary vane lubricated vacuum pumps
(Model 0522-V3-G18D). One pump operated the continuous pair
of impingers connected in series, and the other pump operated
one of the other four samplers through radio-controlled sole-
noid valves. The samplers (impingers) were connected to their
respective vacuum sources by heavy wall tubing (9.5 mm or
3/8-in. ID). Air flow was measured at the beginning and end
of sampling with a water manometer calibrated to the respec-
tive manifold orifices. Collection medium in each of the
impingers was 1.5 x IQ~k m3 (150 ml) of 0.1N NaOH solution.
97
-------
Flow rates were 3.1 x KT1* m3/s (0.65 cfm) to 5.0 m3/s
(1.05 cfm). Time of operation for each sampler was recorded.
Power for the pumps was provided by a 4-kW portable generator;
power for the radio transmitter and receiver was delivered
by individual batteries.
At the completion of each sampling period the collection
from each impinger was placed in a labeled, wide-mouth pint
jar and sealed. Samples were returned to the lab and
analyzed within 3 weeks.
f. Sample Analysis
All samples were analyzed by colorimetric measurement at
5,350 angstroms (535 nm) of the complex formed by the reac-
tion of arsine (generated from the arsenic acid) with silver
diethyldithiocarbamate on a Perkin-Elmer Model 111 UV-VIS
spectrophotometer. Minimum detection limit was 0.2 yg per
sample.
2. RESULTS AND DISCUSSION
The raw data from air sampling at three arsenic acid spraying
operations are summarized in Table B-l. Samples No. 5 and
No. 6 were from the total dosage and collection efficiency
samplers, in all cases. Calculations of airborne concentra-
tions of arsenic (As) were performed as follows:
As collected (ug) (scf) «/m3 m-l 1
Flow (scfm) x time (min) ' 0.028 (m3) ' yg/m vc ;
Collection efficiency of the air samplers was 82.53% ± 12.15
(95% confidence level). Efficiency was defined as:
Effi- _ (concentration No. 5) (B-2)
ciency (concentration No. 5 + concentration No. 6)
Calculated concentrations were then divided by the efficiency
to estimate true air concentration.
98
-------
Table B-l. ARSENIC ACID SPRAYING DATA
Sample
No.
A-1R
A-2R
A-5R
A-6R
A-1L
A-5L
B-1R
B-4R
B-5R
B-6R
B-2L
B-3L
B-5L
C-1R
C-2R
C-3R
C-4R
C-5R
C-6R
C-3L
C-5L
Weight
collected,
pg As
0.7
0.7
0.8
0.2
0.3
<0.2
0.2
1.4
3.4
0.6
0.6
0.5
0.6
0.6
1.2
0.7
2.2
4.2
0.8
1.1
<0.6
Orifice
2
2
4
4
3
1
3
3
1
1
2
2
4
3
3
3
3
4
4
• 1
2
AP
in. H2O
4.8
5.4
3.0
3.0
4.5
4.3
5.3
4.2
3.0
3.0
6.6
4.4
4.3
5.15
4.95
3.55
4.8
5.3
5.3
3.95
4.95
kPa
1.19
1.34
0.75
0.75
1.12
1:07
1.32
1.04
0.75
0.75
1.64
1.09
1.07
1.23
1.23
0.88
1.19
1.32
1.32
0.98
1.23
Flow
scfm
0.82
0.87
0.65
0.65
0.97
0.82
1.05
0.95
0.68
0.68
0.95
0.77
0.78
1.03
1.02
0.87
1.01
0.88
0.88
0.78
0.82
10-" m3/s
3.87
4.11
3.07
3.07
4.58
3.87
4.95
4.48
3.21
3.21
4.48
3.63
3.68
4.86
4.81
4.11
4.77
4.15
4.15
3.68
3.87
Time
min
17
17
177
177
21
153
13
32
52
52
9
23
52
6
2
61
3.4
143
143
3.4
143
s
1,020
1,020
10,620
10,620
1,260
9,180
780
1,920
3,120
3,120
540
1,380'
3,120
360
120
3,660
204
8,580
8,580
204
8,580 .
Concentration ,
pg As/m3
1.8
1.7
0.2
0.06
0.5
<0.06
0.5
1.6
3.4
0.6
2.5
1.0
0.5
3.5
21.0
0.5
22.9
1.2
0.2
14.8
<0.2
VD
-------
Table B-2 summarizes the data used to estimate emission rates
from the spraying operation, or the portion emitted to the
air to be available for drift. The atmospheric diffusion
model which best represents a moving, spraying source is the
instantaneous infinite crosswind line source:72
(B-3)
°L (* )
where D- = dose, or concentration, times time exposed,
g-s/m3
Q = emission rate, g/m
a ... = standard deviation of the distribution of
material in the cloud in the vertical direc-
tion, m
u = average wind speed, m/s
h = height of emission, m
The vertical diffusion coefficient, a _, for an instantaneous
(as opposed to continuous) source varies with downwind dis-
tance and is approximated by the power law functions:72
a T = 0.53x°'73 Unstable atmosphere (classes A and B) (B-4)
Z JL
a = 3.15x°-70 Neutral atmosphere (classes C and D) (B-5)
Z _L
a = O.OSx0-61 Unstable atmosphere (classes E and F) (B-6)
Z -L
The emission rate, QT, is given in terms of milligrams per
LI
meter of the application swath for the instantaneous line
source,, An example calculation to estimate drift from Q
was performed as follows:
Given: Emission per length, mg/m
Swath width, m
Application rate, gal/acre
Composition of application, wt %
100
-------
Table B-2. EMISSION RATE CALCULATION DATA
Sample
No.
A-1R
A-2R
A-1L
B-1R
B-4R
B-2L
B-3L
C-1R
C-2R
C-3R
C-4R
C-3L
Concentration ,
pg H3AsOi,/m3
2.2
2.1
0.6
0.6
1.9
3.0
1.2
4.2
25.4
0.6
27.7
17.9
Time,
s
1,020
1,020
1,260
780
1,960
540
1,380
360
120
3,660
204
204
Distance,
m
300
250
250
50
250
150
250
200
25
400
25
25
Stability
class
C
C
C
B
B
B
B
C
C
C
C
C
°zl'
m
8.13
7.16
7.16
9.21
20.55
20.55
29.84
6.12
1.43
9.94
1.43
1.43
Wind
speed ,
m/s
2.7
4.5
4.5
0.9
0.9
0.9
0.9
1.3
1.3
1.3
1.3
1.3
QL'
mg K3AsO^/m
120.3
170.0
60.0
9.4
71.3
71.3
42.7
30.1
35.7
68.6
66.2
42.8
g I^AsO^acre
48.7
68.8
24.3
3.8
28.9
28.9
42.7
8.1
9.6
18.5
17.8
11.5
kg H3AsOi,/km2
12.03
17.00
6.00
0.94
7.14
7.14
10.55
2.00
2.37
4.57
4.40
2.84
Drift,
%
1.14
1.62
0.57
0.09
0.68
0.68
1.00
0.19
0.23
0.43
0.42
0.27
Average (at 95% confidence level) 0.61
± 0.29
-------
H3AsOit specific gravity = 2.073 (density = 16.667 Ib/gal
or 1.997 kg/m3)
Active ingredient HsAsOit = 75%
Application rate = °'75 ?al . 16'66? lb . i^-SL .
** acre gal lb
= 4,256 g
(or 1.05 Mg H3As0lt/km2)
Field A swath = 10 m
4, 046.9 m2 ? swath _ 404.7 m » swath
acre " 10 m acre
Emission rate = 63'46 m? As . 404'7 m .
T-^ _
m acre 1,000 mg
= 25.7 g As/acre (or 6.35 kg As/km2)
Formula weight I^AsO^ ,., Qt.
__ _ 131. 3D _ -i
Formula weight As 74.92 -1--
1>895 g
25 7 a As
Emission rate = Z3j/ g AS
acre g As
= 48.7 g HaAsO^/acre (or 12.03 kg
Emission rate
Application rate
48.7 g
4,256 g
x 100% = % drift
x 100% = 1.1% drift
Average drift for field A was 1.1% of acid applied, as
measured as airborne losses, while for fields B and C the
losses were 1.0% and 0.3%, respectively. Off-target deposits
might show drift values much higher than 1%, but they were
not measured in this study. In fields A and B spray nozzles
73Handbook of Chemistry and Physics, 33rd Edition. Cleveland,
Chemical Rubber Publishing Co., 1951. p. 1651.
102
-------
were directed upward, which may have contributed to drift
losses being three times those of field C, where nozzles
were directed downward. Wind speed was greater at field A
than at field B; however, drift losses do not demonstrate
the effect of greater wind speed. Greater vertical mixing
of the atmosphere (class B) at field B may have negated the
wind speed effect.
Comparison of the drift calculations with data taken from
other researchers15/39'1*1 and treated in the same manner as
above shows that the magnitude of airborne drift computed is
reasonable. Azinphosmethyl insecticide applications39 show
drift losses of 2% to 12%, and methoxychlor drift15'1*1 is
1% to 8%. Arsenic acid is relatively nonvolatile, thus
drift would be expected to be minimal.
The maximum size of the aerosol droplets collected by the
impinger samplers can be estimated. For example, in a wind
of 1.3 m/s it will take 19 s for an aerosol droplet to reach
a sampler located 25 m downwind. If the droplet must fall
2 m in those 19 s then its terminal settling velocity is es-
timated to be 0.091 m/s (0.3 ft/s) . Figure B-l71* shows that
the droplet can be no larger than 60 ym in diameter. Similarly,
at 400 m downwind, the largest droplet collected by the sampler
will be 15 ym. Drift values in Table B-2 indicate that no
correlation exists with distance. Droplets evaporate rapidly
while traveling (settling) in air, particularly if they are
comprised mostly of water, and this appears to be the case
here. The same drift of arsenic is seen regardless of dis-
tance; however, droplet size is larger at closer distances.
This occurs because the initial droplet emitted is a water-
arsenic acid solution (about 95:5) and with increasing
distance more of the water portion of the droplet evaporates.
71*Chemical Engineers' Handbook, 4th Edition. Perry, J. H.
(ed.). New York, McGraw-Hill, Inc., 1969. p. 5-62.
103
-------
EQUIVALENT STANDARD
TYLER SCREEN MESH
THEORETICAL SCREEN MESH
30.5
3.05
3.05X10"1
E
*~ 3.05xlO~2
o
LU
O
^ 3. 05 x 10
t—
i—
LU
/ 1
'iA
2
> ^
y
( /
4-1-
. A
7 '
i!
3 _ "1
i
T./
Y
/
I
txj JS
0
K. S 8s
- * T7
/
/
-///
/ /
' ^
//
nWcSj
' /
tf\ ^S
I "3
1 //
, / /
i / >
/ //
///
~f .
t f /
, /
/ /
* //
//
1 '' f
( -
t
*
/
ft
y/
^
N
/
7
/
}
I
1
1
t
'f
/,'
-4 "4 t
r~*-
j/ f t
/ y
(I
K
7
r ~)
/ ,
4>
^ y •
//
_ 7
~]_:
/
TEMP
.UID "F
IR 70
WATER 70
10
sgs
.--/
/ /
'//
/ //y
'//
Hi/ '
do /
/ /
i /
'$./ L
1'^
i ^--i
1*XT
g
5?
^
T^
,/*
' ^
/
LJ
/
/
/
/
'
/
V-
Irf
i /*^>
' 77
<
RATURE
°C
21 0
21
NO
TR
SF
9 85
f
/ f
y ^
— 2 _
y
7^
/ /
7
/ ^
7
2 *
> ^
^ X
' *O
-------
APPENDIX C
METHOD FOR ESTIMATING TLV VALUES FOR COMPOUNDS
WHEN NONE EXISTS
In assessing the hazard potential associated with the appli-
cation of agricultural chemicals and subsequent airborne
losses and drift, it was found that no TLV value had been
assigned by the American Conference of Governmental Industrial
Hygienists (ACGIH) for many of the agricultural pesticides.
The TLV of air pollutants is utilized as an integral part
of the methods of emissions characterization in the source
severity criteria.
Thirty agricultural chemicals selected from the booklet
published by the ACGIH containing TLV values29 are shown in
Table C-l. Seven of these chemicals are herbicides, one is
a fungicide, and 22 are insecticides; no distinction was made
between inhalation and skin TLV. The most common toxicity
value published for chemical substances is the acute oral
LDso dose for male rats. These LDsg values were tabulated
with the TLV's and curve-fitting was attempted to correlate
LDso with TLV in the hope of obtaining a relationship whereby
compounds of unknown TLV could be assigned functional TLV's
for use in calculating the criteria described earlier. The
results of the best curve-fit are presented below.
The best APL regression fit was found using an equation of
the type:
105
-------
Table C-l. AGRICULTURAL CHEMICALS WITH PUBLISHED TLV'S
ic29
Substance (primary use)
Abate (insecticide)
Aldrin (insecticide)
Allyl alcohol (herbicide)
Ammate (herbicide)
Arsenic acid (herbicide)
Carbaryl (Sevin®) (insecticide)
Chlordane (insecticide)
Toxapheme (insecticide)
2,4-D (herbicide)
DDT (insecticide)
DDVP (insecticide)
Demeton (insecticide)
Diazinon (insecticide)
Dibrom (insecticide)
Dieldrin (insecticide)
Dinitro-o-cresol (insecticide)
Diquat (herbicide)
Endrin (insecticide)
EPN (insecticide)
Heptachlor (insecticide)
Malathion (insecticide)
Methoxychlor (insecticide)
Methylparathion (insecticide)
Paraquat (herbicide)
Parathion (insecticide)
Phosdrin (insecticide)
Ronnel (insecticide)
2,4,5-T (herbicide)
TEPP (insecticide)
Thiram (fungicide)
TLV,
mg/m3
10
0.25
3
10
0.5
5
0.5
0.5
10
1
1
O.J.
0.1
3
0.25
0.2
0.5
0.1
0.5
0.5
10
10
0.2
0.5
Ool
0.1
10
10
0.05
5
LD50, mg/kg
(acute oral
rat dose)
2,000
55
95
3,900
48
500
570
69
1,200
113
56
9
134
430
60
50
300
5
50
90
1,375
5,000
25
145
15
7
1,740
500
1.2
860
106
-------
y = axb (C-i)
Logarithmic transformation of Equation C-l yields:
In y = In a + b In x (C-2)
Equation C-2 can be further transformed to resemble the
familiar straight-line slope-intercept equation form:
Y = MX + B1 (C-3)
if Y = In y, B1 = In a, M = b, and X = In x. The indicators
of goodness-of-fit for this regression show that R2 = 0.7951
and the F-value = 108.6.
The fitted values for the slope-intercept form were:
B1 = -3.921
M = 0.774
Standard errors were computed and resulted in:
S = 0.07426 = standard error of M (slope)
SY „ = 0.821 = standard error of estimate
SD = 0.3936 = standard error of B1(intercept)
JD
S_, had to be calculated separately where
13
v
E(Transformed x.)2
R X • Y w
* n1E(Transformed x.- mean transformed x.)2
Using the above calculated values, 95% confidence intervals
were obtained about the slope and intercept of the equation
y = ax*3:
107
-------
(A) Slope
b (or M) ± Z , SM gives the upper and lower limits
of the confidence interval. For n1 = 30 and a = 0.05
(95% confidence level), Z-2 = 1.96; the confidence
interval is then 0.774 ± (1.96) (0.7426) or (0.6285 <_
slope <_ 0.9195) at the 95% level. The slope confi-
dence interval is the same in transformed space as in
the original space.
(B) Intercept
In transformed space, the 95% confidence interval would
be B1 ± Z , S_.; but in the original space, we have
a/2 B
——=-y <_ intercept <_ anti In a I anti ln(Z
[anti ln(Za/2 SB)J
r> m QR9
which is 2 1629 - intercePt 1 (0. 01982) (2.1629)
or (0.00916 <_ intercept <_ 0.04287) at the 95% level.
In the Y = MX + B1 equation form, the 95% confidence limits
for B1 are ±19.7% of B1, and for M are ±18.8% of M. In
original space using the exponential equation form y = ax ,
the limits for b are the same as those for M, but the confi-
dence limits for "a" become +216.5% and -46.3%. Dividing
the maximum value by the minimum value for the 95% confidence
interval yields 4.68 for "a" and 1.46 for b.
The final form of the regressed equation relating LDs0 to
TLV, given the original (LD5Q, TLV) pairs, is:
TLV = 0.0198 (LD50) °'7714 (C-4)
where LD50 = acute oral dose, mg/kg, for male rat
TLV = threshold limit value, mg/m3
108
-------
SECTION VIII
GLOSSARY OF TERMS
ABSCISSION - The process by which a leaf or other part is
separated from a plant.
ACTIVE INGREDIENT - A substance contained in a formulation
which will by itself act in the same manner and for the same
purposes as the directions provide for the formulation as a
whole.
ADJUVANT - An ingredient which, when added to a formulation,
aids the action of the toxicant.
ATMOSPHERIC STABILITY CLASS - Categories used to describe the
turbulent structure and wind speed of the atmosphere.
ATOMIZATION - The process of reducing a liquid to a fine
spray.
BOLL - The pod of a plant, especially of cotton.
BRONCHITIS - An inflammation, acute or chronic, of the mucous
lining of the bronchial tubes.
CARRIER - An inert material added to a technical poison,, to
facilitate later dilution to field strength in simple
blending equipment.
DBF - A defoliant, tributylphosphorotrithioate.
DEFOLIATION - Accelerated leaf abscission.
DERMATITIS - Inflammation of the skin.
DESICCATION - Accelerated drying of plant or plant part.
EDEMA - An abnormal accumulation of fluid in cells, tissues,
or cavities of the body, resulting in swelling.
109
-------
FIBROSIS - An abnormal increase in the amount of fibrous
connective tissue in an organ, part, or tissue.
FOLEX -• A defoliant, tributylphosphorotrithioite.
HEMORRHAGE - The escape of blood from its vessels; especially,
heavy bleeding.
HERBICIDE - A chemical intended for killing plants or inter-
rupting their normal growth.
LDso - Abbreviation of median lethal dose which indicates
the amount of toxicant necessary to effect a 50% kill of
the pest being tested.
MICRONAIRE - A measure of cotton lint fineness.
NASAL SEPTUM - The part of the nose which separates the
nostrils.
NEWTONIAN FLUID - A fluid in which there is a linear relation
between the shear stress and the rate of shear.
NON-NEWTONIAN FLUID - A fluid in which the relation between
the she:ar stress and the rate of shear is not linear.
PARAQUAT - Common name for compounds containing the cation
l:l'-dimethyl-4,4'-bipyridylium; a herbicide for coarse
grasses! and a major desiccant for cotton.
PESTICIDE - Substance or mixture of substances intended for
preventing, destroying, repelling, or mitigating any insects,
rodents, nematodes, fungi, or weeds, or any other forms of
life declared to be pests.
PETIOLE! - The stalk to which a leaf is attached.
PHYTOTOXICITY - The state of being poisonous to plants.
PULMONARY DAMAGE - Damage to the lung or lung-like organs.
ROUGHNESS LENGTH - A term which expresses the effect of
varying ground surface roughness on the wind velocity
profile close to ground level.
SHEAR RATE - The rate at which material (water) is fragmented
by friction and tearing forces when the material is sprayed.
t
SLIPSTREAM - The current of air thrust backward by the
spinning propeller of an aircraft.
STAPLE - A particular length and degree of fineness of
cotton fibers.
110
-------
STOKES1 LAW - An equation used to calculate the drag force
between a particle and surrounding fluid in relative motion.
SUPERADIABATIC - An atmospheric condition in which the lapse
rate (9T/3Z) is less than -1°C/100 meters.
SURFACTANT - A substance that reduces the interfacial tension
of two boundary lines.
SWATH - The space or width covered by one pass of a moving
device.
Ill
-------
SECTION IX
CONVERSION FACTORS AND METRIC PREFIXES75
To convert from
angstrom
degree Celsius (°C)
degree Kalvin (°K)
gram (g)
gram/kilogram (g/kg)
gram/meter2 (g/m2)
kilogram (kg)
kilogram (kg)
kilogram/meter3 (kg/m3)
kilometer/hour (km/hr)
kilometer2 (km2)
meter (m)
meter (m)
meter/second (m/s)
meter2 (ni2)
meter3 (iri )
metric ten
Newtons/rreter (N/m)
pascal (Pa)
pascal (Pa)
pascal-second (Pa-s)
radian (rad)
CONVERSION FACTORS
t£
meter
degree Fahrenheit
degree Celsius
pound-mass
pound/ton
pound/acre
pound-mass (Ib mass
avoirdupois)
ton (short, 2,000 Ib mass)
Ib mass/foot3
miles/hr
acre
foot
mile
miles/hr
acre
foot3
pound
dynes/centimeter
inch of Hg (60°F)
pound-force/inch2 (psi)
poise
degree (angle)
Multiply by
1.000 x 10~10
toF = 1.8 t0(, + 32
toc = tOR - 273.15
2.205 x 10~3
1.999
8.928
2.204
1.102
6.243
6.215
2.470
3.281
6.215
2.237
2.470
3.531
2.205
1.000
2.961
1.450
1.000
5.730
x 10~3
x 10~2
x 10"1
x 102
x 10-**
x 10'1*
x 101
x 103
x 103
x 10-"
x 10-4
x 101
x 101
Metric Practice Guide. American Society for Testing and Materials.
Philadelphia. ASTM Designation: E 380-74. November 1974. 34 p.
112
-------
PREFIXES
Multiplication
Prefix Symbol Factor Example
kilo k 103 1 kPa = 1 x 103 paschal
milli m 10~3 1 mg = 1 x 10~3 gram
micro y 10~6 1 ym = 1 x 10~6 meter
nano n 10~9 1 nm = 1 x 10~9 meter
113
-------
SECTION X
REFERENCES
1. Osborne, D. J. Defoliation and Defoliants. Nature.
21^:564-567, August 10, 1968.
2. Miller, C. S., E. D. Cook, J. L. Hubbard, J. S. Newman,
E. L. Thaxton, and L. H. Wilkes. Cotton Desiccation
Practices and Experimental Results in Texas. Texas
Agricultural Experiment Station. College Station, Texas.
Miscellaneous Publication No. MP-903. November 1968.
14 p.
3. Addicott, F. T., and R. S. Lynch. Defoliation and
Desiccation: Harvest-Aid Practices. In: Advances in
Agronomy, Vol. 9. 1957. p. 69-93.
4. Insecticidal Spraying of Field Crops With Ground
Machinery. Texas Agricultural Extension Service.
College Station, Texas. Bulletin No. L-486. August 1961,
5. Riley, J. A., and W. L. Giles. Agricultural Meteorology
in Relation to the Use of Pesticides. Agricultural
Meteorology. 2^:225-245, 1965.
6. Akesson, N. B., and W. E. Burgoyne. Spray Atomization,
Application Volume and Coverage. Proceedings and Papers
of the Thirty-Fifth Annual Conference of the California
Mosquito Control Association, Inc., and the American
Mosquito Control Association. February 1967. p. 139-144,
7. Yates, W. E., and N. B. Akesson. Reducing Pesticide
Chemical Drift. In: Pesticide Formulations. Van
Valkenburg, Wade, (ed.). New York, Marcel Dekker, Inc.,
1973. p. 275-341.
8. Akesson, N. B., and W. E. Yates. Problems Relating
to Application of Agricultural Chemicals and Resulting
Drift Residues. Annual Review of Entomology. 9^:285-318,
1964.
114
-------
9. Kutsenogiy, K. P., V. I. Makarov, Y. F. Chankin, V. M.
Sakharov, and G. N. Zagulyayev. Study of the Physico-
chemical Characteristics of Large Aerosol Waves.
Institut Eksperimental'nye Meteorologiya. 27:97-104,
1972.
10. Maybank, J., and K. Yoshida. Delineation of Herbicide
Drift Hazards on the Canadian Prairies. Transactions
of the American Society of Agricultural Engineers.
12^:759-762, 1969.
11. Courshee, R. J. Investigations on Spray Drift. II.
The Occurrence of Drift. Journal of Agricultural
Engineering Research. £:229-241, 1959.
12. Ware, G. W., W. P. Cahill, and B. J. Estesen. Pesticide
Drift: Aerial Applications Comparing Conventional
Flooding vs. Raindrop® Nozzles. Journal of Economic
Entomology. 6£(3):329-330, 1974.
13. Coutts, H. H., and W. E. Yates. Analysis of Spray
Droplet Distributions from Agricultural Aircraft.
Transactions of the American Society of Agricultural
Engineers. 11 (1):25-27, 1968.
14. Ware, G. W., E. J. Apple, W. P. Cahill, P. D. Gerhardt,
and K. R. Frost. Pesticide Drift. II. Mist Blower vs.
Aerial Application of Sprays. Journal of Economic
Entomology. 6_2_(4) :844-846, August 1969.
o
15. Ware, G. W., B. J. Estesen, W. P. Cahill, P. D. Gerhardt,
and K. R. Frost. Pesticide Drift. I. High-Clearance
vs. Aerial Application of Sprays. Journal of Economic
Entomology. £2 (4):840-843, August 1969.
16. Yeo, D., N. B. Akesson, and H. H. Coutts. Drift of
Toxic Chemicals Released from a Low-Flying Aircraft.
Nature. I83_: 131-132, January 10, 1959.
17. Brooks, F. A. The Drifting of Poisonous Dusts Applied
by Airplanes and Land Rigs. Agricultural Engineering.
28^(6) :233-239, June 1947.
18. Scotton, J. W. Atmospheric Transport of Pesticide
Aerosols. U.S. Dept. of Health, Education, and Welfare,
Public Health Service. Washington. PB 228 612. July
1965. 30 p.
115
-------
19. Yates, W. E., N. B. Akesson, and H. H. Coutts. Drift
Hazards Related to Ultra-Low-Volume and Diluted Sprays
Applied by Agricultural Aircraft. Transactions of the
American Society of Agricultural Engineers.
10 (5) :628-632, 638, 1967.
20. Pooler, F. Atmospheric Transport and Dispersion of
Pesticides. (Presented at the Symposium on Guidelines
for Environmental Studies of Pesticides. 162nd National
Meeting, American Chemical Society. Washington.
September 1971.) 20 p.
21. Census of Agriculture, 1969. Volume V, Special Reports.
Part 15, Graphic Summary. Washington, U.S. Bureau of
tha Census, 1973.
22. Akesson, Dr. N. B. Department of Agricultural Engineer-
ing, University of California-Davis. Personal communi-
cation, March 1975.
23. Metzer, Dr. R. B. Texas Agricultural Extension Service,
College Station, Texas. Personal communication,
February 1975.
24. Ware, Dr. G. W. Department of Entomology, University
of Arizona, Tucson, Arizona. Personal communication,
January 1975.
25. Miller, Dr. C. S. Department of Plant Sciences, Texas
A & M University, College Station, Texas. Personal
coirmunication, January 1975.
26. Mullins, Dr. J. A. Tennessee Agricultural Extension
Service, Jackson, Tennessee. Personal communication,
January 1975.
27. The Toxic Substances List, 1974 Edition. U.S. Department
of Health, Education, and Welfare. Rockville, Maryland.
HEW Publication No. (NIOSH) 74-134. June 1974. 904 p.
28. 1969 Farm Chemicals Handbook. Willoughby, Ohio, Meister
Publishing Co., 1968. p. D158.
29. TLVts® Threshold Limit Values for Chemical Substances
and Physical Agents in the Workroom Environment with
Intended Changes for 1975. American Conference of
Governmental Industrial Hygienists. Cincinnati. 1975.
97 p.
116
-------
9. Kutsenogiy, K. P., V. I. Makarov, Y. F. Chankin, V. M.
Sakharov, and G. N. Zagulyayev. Study of the Physico-
chemical Characteristics of Large Aerosol Waves.
Institut Eksperimental'nye Meteorologiya. 27:97-104,
1972.
10. Maybank, J., and K. Yoshida. Delineation of Herbicide
Drift Hazards on the Canadian Prairies. Transactions
of the American Society of Agricultural Engineers.
1^:759-762, 1969.
11. Courshee, R. J. Investigations on Spray Drift. II.
The Occurrence of Drift. Journal of Agricultural
Engineering Research. £:229-241, 1959.
i
12. Ware, G. W., W. P. Cahill, and B. J. Estesen. Pesticide
Drift: Aerial Applications Comparing Conventional
Flooding vs. Raindrop® Nozzles. Journal of Economic
Entomology. 6£(3):329-330, 1974.
13. Coutts, H. H., and W. E. Yates. Analysis of Spray
Droplet Distributions from Agricultural Aircraft.
Transactions of the American Society of Agricultural
Engineers. Ll(l):25-27, 1968.
14. Ware, G. W., E. J. Apple, W. P. Cahill, P. D. Gerhardt,
and K. R. Frost. Pesticide Drift. II. Mist Blower vs.
Aerial Application of Sprays. Journal of Economic
Entomology. £2_(4) : 844-846, August 1969.
15. Ware, G. W., B. J. Estesen, W. P. Cahill, P. D. Gerhardt,
and K. R. Frost. Pesticide Drift. I. High-Clearance
vs. Aerial Application of Sprays. Journal of Economic
Entomology. 6^(4):840-843, August 1969.
16. Yeo, D., N. B. Akesson, and H. H. Coutts. Drift of
Toxic Chemicals Released from a Low-Flying Aircraft.
Nature. 183^:131-132, January 10, 1959.
17. Brooks, F. A. The Drifting of Poisonous Dusts Applied
by Airplanes and Land Rigs. Agricultural Engineering.
2_8(6) :233-239, June 1947.
18. Scotton, J. W. Atmospheric Transport of Pesticide
Aerosols. U.S. Dept. of Health, Education, and Welfare,
Public Health Service. Washington. PB 228 612. July
1965. 30 p.
115
-------
19. Yates, W. E., N. B. Akesson, and H. H. Coutts. Drift
Hazards Related to Ultra-Low-Volume and Diluted Sprays
Applied by Agricultural Aircraft. Transactions of the
American Society of Agricultural Engineers.
1CI(5) :628-632, 638, 1967.
20. Pooler, F. Atmospheric Transport and Dispersion of
Pesticides. (Presented at the Symposium on Guidelines
for Environmental Studies of Pesticides. 162nd National
Meeting, American Chemical Society. Washington.
September 1971.) 20 p.
21. Census of Agriculture, 1969. Volume V, Special Reports.
Part 15, Graphic Summary. Washington, U.'S. Bureau of
the Census, 1973.
22. Akesson, Dr. N. B. Department of Agricultural Engineer-
ing, University of California-Davis. Personal communi-
cation, March 1975.
23. Metzer, Dr. R. B. Texas Agricultural Extension Service,
College Station, Texas. Personal communication,
February 1975.
24. Ware, Dr. G. W. Department of Entomology, University
of Arizona, Tucson, Arizona. Personal communication,
January 1975.
25. Miller, Dr. C. S. Department of Plant Sciences, Texas
A & M University, College Station, Texas. Personal
communication, January 1975.
26. Mullins, Dr. J. A. Tennessee Agricultural Extension
Service, Jackson, Tennessee. Personal communication,
January 1975.
27. The Toxic Substances List, 1974 Edition. U.S. Department
of Health, Education, and Welfare. Rockville, Maryland.
HEW Publication No. (NIOSH) 74-134. June 1974. 904 p.
28. 1969 Farm Chemicals Handbook. Willoughby, Ohio, Meister
Publishing Co., 1968. p. D158.
29. TLVs® Threshold Limit Values for Chemical Substances
and Physical Agents in the Workroom Environment with
Intended Changes for 1975. American Conference of
Governmental Industrial Hygienists. Cincinnati. 1975.
97 p.
116
-------
30. 1969 Cotton Defoliation Guide. Texas Agricultural
Extension Service. College Station, Texas. Bulletin
No. L-145.
31. Strickland, J., and T. Blue. Environmental Indicators
for Pesticides. Stanford Research Institute, Council
on Environmental Quality Contract EQC 217. Menlo Park,
California. PB 210 666. April 1972. p. 38.
32. FDA Pesticide Analytical Manual. Vol. II. Pesticide
Reg. Sec. 120,272. U.S. Department of Health, Education,
and Welfare. November 1973.
33. Weed Killers. In: Kirk-Othmer Encyclopedia of Chemical
Technology, Second Edition. Vol. 22. New York, John
Wiley & Sons, Inc., 1969. p. 19.
34. von Riimker, R. , E. W. Lawless, and A. F. Meiners.
Production, Distribution, Use and Environmental Impact
Potential of Selected Pesticides. Midwest Research
Institute, Council on Environmental Quality. Contract
EQC-311. Kansas City, Missouri. March 1974. p. 256.
35. Sullivan, R. J. Preliminary Air Pollution Survey of
Arsenic and Its Compounds, A Literature Review. Litton
Systems, Inc., HEW Contract PH 22-68-25. October 1969.
p. 2.
36. Slade, P. Photochemical Degradation of Paraquat.
Nature. 20^:515-516, 1965.
37. Staiff, D. C., S. W. Comer, J. F. Armstrong, and H. R.
Wolfe. Exposure to the Herbicide, Paraquat. Bulletin
of Environmental Contamination and Toxicology.
14_(3) :334-340, 1975.
38. Rose, M. S. The Search for an Effective Treatment of
Paraquat Poisoning. Chemistry and Industry (London).
1975(10);413-415, May 17, 1975.
39. Argauer, R. J., H. C. Mason, C. Corley, A. H. Higgins,
J. N. Sauls, and L. A. Liljedahl. Drift of Water-
Diluted and Undiluted Formulations of Malathion and
Azinphosmethyl Applied by Airplane. Journal of Economic
Entomology. £1(4):1015-1020, August 1968.
40. Ware, G. W., B. J. Estesen, W. P. Cahill, P. D. Gerhardt,
and K. R. Frost. Pesticide Drift. III. Drift Reduc-
tion with Spray Thickeners. Journal of Economic
Entomology. £3(4) :1314-1316, August 1970.
117
-------
41. Ware, G. W., B. J. Estesen, W. P. Cahill, and K. R.
Frost. Pesticide Drift. V. Vertical Drift from
Aerial Spray Applications. Journal of Economic
Entomology. 65_(2) : 590-592, April 1972.
42. Turner, D. B. Workbook of Atmospheric Dispersion Esti-
mates. U.S. Department of Health, Education, and
Welfare. Cincinnati. Public Health Service Publication
No. 999-AP-26. May 1970, 65 p.
43. Census of Agriculture, 1971. Volume V, Special Reports.
Part 3, Cotton. Washington, U.S. Bureau of the Census,
1973.
44. Agricultural Statistics 1973. Washington, U.S. Depart-
ment of Agriculture, 1973. 617 p.
45. Andrilenas, P. A. Farmers' Use of Pesticides in 1971 —
Quantities. U.S. Department of Agriculture. Washington.
Agricultural Economic Report No. 252. July 1974. 56 p.
46. Texas Cotton Review, 1973-74. Natural Fibers Economic
Research. University of Texas at Austin. Research
Report No. 104 (PB 235 388). July 1974. 143 p.
47. Ware, G. W., C. H. Kreader, and L. Moore. Agricultural
Use of Pesticides in Arizona. Progressive Agriculture.
University of Arizona. Tucson. July-August 1974.
p. 12-13,16.
48. Yates, W. E., N. B. Akesson, and D. Bayer. Effects of
Spray Adjuvants on Drift Hazards. (Paper No. 74-1008,
presented at the 1974 Annual Meeting, American Society
of Agricultural Engineers. Stillwater, Oklahoma.
June 23-26, 1974.) 26 p.
49. Butler, B. J., N. B. Akesson, and W. E. Yates. Use of
Spray Adjuvants to Reduce Drift. Transactions of the
American Society of Agricultural Engineers.
12(2):182-186, 1969.
50. Kartellopoulos, A. G. Additives in Herbicide Formula-
tions. Chemistry and Industry (London). 1974(9);951-955,
December 7, 1974.
51. Akesson, N. B., W. E. Yates, and R. E. Cowden. What's
Happening in Aerial Application Research. Unpublished
paper. Agricultural Engineering Department, University
of California at Davis. 1975.
118
-------
52. Dacagin Speeds Cotton Defoliation. Agricultural
Chemicals. October 1969. p. 83.
53. Goering, C. E., and B. J. Butler. Paired Field Studies
of Herbicide Drift. (Paper No. 73-1575, presented at
1973 Winter Meeting, American Society of Agricultural
Engineers. Chicago. December 11-14, 1973.) 22 p.
54. Schultz, H. B., N. B. Akesson, W. E. Yates, and K. H.
Ingebretsen. Drift of 2,4-D Applied by Plane.
California Agriculture. 10^(8) :4-5,14, August 1956.
55. Ware, G. W., B. J. Estesen, W. P. Cahill, and K. R.
Frost. Pesticide Drift. VI. Target and Drift Deposits
vs. Time of Applications. Journal of Economic
Entomology. £5(4):1170-1172, August 1972.
56. Scifres, C. J., H. G. McCall, and D. W. Fryear. Foam
Systems as Herbicide Carriers for Range Improvement.
Texas Agricultural Experiment Station. Miscellaneous
Publication 1974. 18 p.
57. Akesson, N. B., S. E. Wilce, and W. E. Yates. Confining
Aerial Applications to Treated Fields — A Realistic
Goal. Agrichemical Age. December 1971. p. 11-14.
58. Threadgill, E. D., and R. F. Colwick. Ground Applica-
tions of Cotton Defoliants with Air Aspirating Nozzles.
In: Proceedings of the 27th Annual Beltwide Cotton
Defoliation and Physiology Conference. Phoenix.
January 9-10, 1973. p. 35.
59. Brazelton, R. W. Control of Chemical Drift. University
of California, Agricultural Extension. Bulletin No.
OSA #n5. July 1971. 2 p.
60. Batchelder, D. G., J. G. Porterfield, and G. McLaughlin.
Thermal Defoliation of Cotton. In: Proceedings of the
25th Annual Beltwide Cotton Defoliation and Physiology
Conference. Atlanta. January 12-13, 1971. p. 36-37.
61. Cotton Growers Spent $60 Million for Herbicide in '66,
Shaw Tells Cotton Mech Conference. Farm Chemicals.
13£(2) :80-82, February 1967.
/
62. Erby, W. A., W. E. Erner, J. S. Skaptason, and R. A.
Walde. Defoliation and Desiccation of Cotton with
cis-2,3,5,5,5-Pentachloro-4-keto-2-pentanoic Acid.
U.S. Patent 3,472,004 (to Air Products and Chemicals,
Inc.), October 14, 1969.
119
-------
63. Wcilworth, B. L. 3-Amino-3-carboxypropylmethylsulfoximine
Salts as Nonselective Water-Soluble Defoliants. U.S.
Patent 3/323,895 (to American Cyanamid Co.), June 6, 1967.
64. Neuville, M. L., and R. B. Carroll. Cacodylic Acid
Plant Defoliants. U.S. Patent 3,378,364 (to Ansul Co.),
April 16, 1968.
65. Hensel, J., and D. W. Gier. Defoliating and Desiccating
Plants with Substituted Triphenyl Phosphates and
Phosphites. U.S. Patent 3,416,911 (to Chemagro Corp.),
December 17, 1968.
66. A Natural Defoliant. Agricultural Research. 14(5)ill,
November 1965.
67. Miller, C. S., L. H. Wilkes, E. L. Thaxton, and J. L.
Hubbard. Cotton Wilt-Harvest and Wiltant Defoliation
Effectiveness in Texas. Texas Agricultural Experiment
Station. Miscellaneous Publication No. MP-1010.
October 1971. 12 p.
68. 1973 Handbook of Agricultural Charts. Washington, U.S.
Department of Agriculture, October 1973.
69. Saunders, J. M., and H. R. Cams. The Usage of Harvest-
Aid Chemicals, 1952-1960. In: Proceedings of the 16th
Annual Beltwide Cotton Defoliation and Physiology
Conference. Memphis. January 9-10, 1962. p. 8-12.
70. Eichers, T., P. A. Andrilenas, R. Jenkins, and A. Fox.
Quantities of Pesticides Used by Farmers in 1964. U.S.
Department of Agriculture. Washington. Agricultural
Economic Report No. 131. January 1968. 37 p.
71. Eimutis, E. C. Source Assessment: Prioritization of
Stationary Air Pollution Sources, Model Description.
Monsanto Research Corporation. Dayton. Report No.
MRC-DA-508. Environmental Protection Agency,
EPA-600/2-76-032a. February 1976. 77 p.
72. Meteorology and Atomic Energy 1968. Slade, D. H. (ed.).
U.S. Atomic Energy Commission. (NTIS TID-24190).
July 1968. 445 p.
73. Handbook of Chemistry and Physics, 33rd Edition.
Cleveland, Chemical Rubber Publishing Co., 1951.
p. 1651.
120
-------
74. Chemical Engineers' Handbook, 4th Edition. Perry, J. H,
(ed.). New York, McGraw-Hill, Inc., 1969. p. 5-62.
75. Metric Practice Guide. American Society for Testing
and Materials. Philadelphia. ASTM Designation:
E 380-74. November 1974. 34 p.
121
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-77-107
fed
2.
3. RECIPIENT'S ACCESSION- NO.
a. TITLE ANDsuBT.TLE SOURCE ASSESSMENT: DEFOLIATION
OF COTTON, State of the Art
5. REPORT DATE
July 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
J.A. Peters and T.R. Blackwood
8. PERFORMING ORGANIZATION REPORT NO.
MRC-DA-708
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Monsanto Research Corporation
1515 Nicholas Road
Dayton, Ohio 45407
10. PROGRAM ELEMENT NO.
LAB015; ROAP 2LAXM-071
11. CONTRACT/GRANT NO.
68-02-1874
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERE.D
Task Final; 2/75-2/76
14. SPONSORING AGENCY CODE
EPA/600/13
is. SUPPLEMENTARY NOTES IERL-RTP task officer for this report is David K. Oestreich,
Mail Drop 62, 919/541-2547.
is. ABSTRACT Tne repOr|. describes a study of air pollutants emitted during the defoliation
or desiccation of cotton prior to harvest. (Defoliation is the process by which leaves
are abscissecl from the plant by the action of topically applied chemical agents. Desi-
ccation by chemicals is the drying or rapid killing of the leaf blades and petioles with
the leaves remaining in a withered state on the plant.) Emissions of defoliants were
DEF, Folex, and sodium chlorate. Emissions of desiccants were arsenic acid and
paraquat. Source severity for emissions from a representative source were 0.69 + or
-0. 32 for arsenic acid, 0. 30 for paraquat, 0.44 for sodium chlorate, and 0. 67 for
DEF. (Source severity is a measure of the hazard potantial of a representative
emission source; for this source type, it was defined as the ratio of the time-averaged
ground level concentration of the species emitted at the downwind perimeter of a
representative field undergoing spraying for defoliation or desiccation, to a time-
adjusted exposure factor related to TLV.) Existing control technology for aerial
application of pesticides includes the use of fluid additives and nozzle design/orien-
tation to reduce chemical drift. Future control technology considerations include the
use of foam spray systems, multiple needle nozzle systems, and the replacement of
chemical defoliation with thermal defoliation.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Cotton Plants
Defoliation
Desiccation
Sodium
Chlorates
Arsenic Organic
Acids
Herbicides
Pyridines
Sulfonic Acids
Pesticides
Air Pollution Control
DEF
Folex
Sodium Chlorate
Paraquat
Source Severity
13 B
02D
07A
07C
06F
IS. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
133
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
CPA Form 2220-1 (9-73)
122
------- |