EPA-650/3-75-003
TRAVELS  OF  AIRBORNE  POLLEN
                       by

  Eugene C. Ogden, Gilbert S. Raynor, and Janet V. Hayes

           University of the State of New York
                 Albany , New York


                  Grant No. 800677
                  ROAP No. 21ADO
             Program Element No. 1AA009


         EPA Project Officer:  George W .  Griffing

               Meteorology Laboratory
         National Environmental Research Center
       Research Triangle Park, North Carolina 27711


                   Prepared for

        OFFICE OF  RESEARCH AND DEVELOPMENT
       U.S. ENVIRONMENTAL PROTECTION AGENCY
              WASHINGTON, D.C. 20460

                   January 1975

-------
                        EPA REVIEW NOTICE

This report has been reviewed by the National Lnviron^nental Research
Center - Research Triangle Park, Office of Research and Development,
EPA, and approved for publication.  Approval does not signify that the
contents necessarily reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
                   RESEARCH REPORTING SERIES


Research  repoi is of the Office 61 Research and Development, U.S. Environ-
mental Protection Agency, have been grouped into series.  These broad
categories were established to facilitate further development and applica-
tion of environmental  technology.  Elimination of traditional grouping was
consciously planned to foster technology transfer and maximum interface
in related fields. These series are:


          1.  ENVIRONMENTAL HEALTH EFFECTS RESEARCH

          2 . ENVIRONMENTAL PROTECTION TECHNOLOGY

           3. ECOLOGICAL  RESEARCH

          4. ENVIRONMENTAL MONITORING

          5. SOCIOECONOMIC ENVIRONMENTAL STUDIES
          6. SCIENTIFIC AND TECHNICAL ASSESSMENT REPORTS
          9. MISCELLANEOUS

This report has been  assigned to the ECOLOGICAL RESEARCH series.
This series describes research on the effects of pollution on humans,
plant and animal species, and materials.  Problems are assessed for
their long- and short-term influences.  Investigations  include forma-
tion,  transport, and pathway studies to determine the  fate of pollutants
and theif effects.  This work provides the technical basis for setting
standards to minimize undesirable changes in living organisms in
the aquatic,  terrestrial and atmospheric environments.
 This document is available to the public for sale through the National
 Technical Information Service, Springfield, Virginia 22161.
                              11

-------
                             ABSTRACT
From 1957 to 1973, studies were conducted on the transport and dispersion
of airborne pollen.  The major areas of research were:  (a) development
and evaluation of sampling devices for particles in the pollen size range,
(b) development and evaluation of techniques for tagging pollen in
living plants with dyes and radioisotopes, (c) dispersion and deposition
of pollen from known sources of several configurations, (d) study of the
effects of forested areas in removing pollen from the atmosphere, (e)
study of the variation in concentrations of pollen from natural sources
with distance, height, time, and other variables, (f) study of the
feasibility of predicting concentrations of ragweed pollen from unknown
sources, (g) study of the occurrence and concentrations of ragweed
pollen in a large source-free area, and (h) comparison  of the
concentrations of ragweed pollen before and after ragweed eradication
efforts.

This report was submitted in fulfillment of Grant Number R-800677 by
the University of the State of New York under the (partial) sponsorship
of the Environmental  Protection Agency.  Work was completed as of
December, 1974.

-------
                             CONTENTS
Abstract                                                              iii
List of Figures                                                         v
Acknowledgments                                                       vii
Sections
I      Conclusions                                                      1
II     Recommendations                                                  3
III    Introduction                                                     7
IV     Instrument Development, Testing,  and Comparison                   9
V      Tagging Experiments                                             26
VI     Dispersion and Deposition Studies Over Open Terrain              32
VII    Particulate Dispersion Into and Within a  Forest                  40
VIII   Deposition to Vegetated Surfaces: Grassland vs.  Forest          45
IX     Diurnal Patterns of Emission                                    47
X      Temporal Variation in Pollen Concentrations                     48
XI     Spatial Variability in Pollen Concentrations                    50
XII    Variation in Pollen Concentrations with Height                  53
XIII   Enhancement of Pollen Concentrations Downwind
           of Vegetative Barriers                                      54
XIV    Mesoscale Transport and Dispersion of Airborne Pollen           57
XV     Experimental Prediction of Daily Ragweed  Pollen  Concentrations   58
XVI    Occurrence of Ragweed Pollen in a Source-free Area              60
XVII   Concentrations of Ragweed Pollen in Relation
           to Reduction of Source Plants                               79
XVIII  Airborne Pollen From Entomophilous Plants                       82
XIX    Occurrence of Airborne Pollens in Winter                         86
                                IV

-------
                              FIGURES

  No.                                                                  Page
 4-1     Slide-edge-cylinder Sampler                                    11

 4-2     Intermittent Swing-shield Rotoslide Sampler                    13

 4-3     Sequential Swing-shield Rotoslide Sampler                      15

 4-4     Examination of a Slide-edge Sample                             16

 4-5     Variable Air Flow Isokinetic Sampler                           19

 4-6     Automatic Grab Sampler                                         21

 4-7     Wind Tunnel at Brookhaven National Laboratory                  22

 5-1     Cotton Wrap Technique for Introduction of a Radioisotope       28
           Into a Ragweed Plant

 5-2     Autoradiographs of Ragweed Pollen                              30

 5-3     Ragweed Field for Preseason Pollen                             31

 6-1     Aerial View of Area Source of Ragweed Pollen                   32

 6-2     Ground Level  View of Area Source of Ragweed Pollen              34

 6-3     Point Source  Pollen-dispensing Apparatus                        35

 6-4     Movable Line  Source of Kochia Pollen                           37

 7-1     Experimental  Forest                                            41

11-1     Rotoslide Sampler With Revolution Counter                      51

-------
  No.                                                                  Page
13-1     Sampling Array for Studying the Effect of a Hedge on
           Pollen Concentrations                                        55

16-1     Mobile Tower at Blue Mountain Lake and Tower at
           Sagamore Pond                                                63

16-2     Rotoslide Pollen Samplers on the Summit of
           Whiteface Mountain                                           65

16-3     Average Diurnal Patterns of Ragweed Pollen at the
           Sagamore Tower for 1966                                      68

16-4     Average Diurnal Patterns of Ragweed Pollen at the
           Sagamore Tower for 1967                                      69

16-5     Average Diurnal Patterns of Ragweed Pollen at
           Blue Mountain Lake                                           70

16-6     Daily Average Concentrations of Ragweed Pollen at
           Sagamore in 1966                                             72

16-7     Daily Average Concentrations of Ragweed Pollen at
           Sagamore in 1967                                             73

16-8     Daily Average Concentrations of Ragweed Pollen at
           Three Heights at Sagamore for Those Days When
           a Concentration  was at Least 30/m3                          74

17-1     Daily Average Concentrations of Ragweed Pollen at
           Saratoga Springs                                             80

17-2     Average Diirnal Patterns of Ragweed Pollen at
           Saratoga Springs                                             81
                                  VI

-------
                          ACKNOWLEDGMENTS
Many persons and institutions were of valued service in the conduct
of these studies.  We appreciate the assistance of staff members
of the New York State Museum and Science Service and others in the
New York State Education Department.  We are grateful  for help
from many at Brookhaven National Laboratory, especially in the
Biology Department and in the Meteorology Group.  Several of the
staff of the Division of Air Resources, New York State Department
of Environmental Conservation, helped during critical  periods. We
express our appreciation for permission to install  extensive sampling
assemblages at: Brookhaven National  Laboratory; Sagamore Conference
Center of Syracuse University, Raquette Lake; Minnowbrook Conference
Center of Syracuse University, Blue  Mountain Lake;  Adirondack
Museum, Blue Mountain Lake; State University of New York Atmospheric
Sciences Research Center, Wilmington; Whiteface Mountain Authority,
Wilmington; New York State Department of Environmental Conservation
Saratoga Nursery at Saratoga Springs; Lake Eaton Campsite,  and Fish
Creek Pond Campsite; the city of Saratoga Springs;  and the property
of Dr. Donald J. Dean, GUI'Iderland.

Several persons should receive special  mention.  Of the staff of
the New York State Museum and Science Service:  Nancy Farr prepared
materials for sampling and supervised identification and counting
for the up-State locations; Eileen Coulston served  as  our bibliographer;
Donald Lewis advised on pollen identification and lab  techniques;
Kenneth Dean processed many of the samples; these four also operated
samplers in the field as needed.

Of the staff of Brookhaven National  Laboratory:  Maynard Smith authorized
use of meteorological facilities and gave valued advice;  Irving Singer
advised on  mathematical  matters  and  helped program  for computer analysis;

-------
James Stangby supplied services of the Biology Department;  Frank
German and his staff cared for our plants  in the greenhouse and
fields; Charles Meinhold advised on radiation  safety and  supplied
radiation counting machinery; Charles Flood taught us how to handle
radioisotopes and assisted during critical  times.

Among our full-time and part-time employees:  Fay Hyland directed
the studies with vital dyes; Benjamin Graham planned the  initial
studies with radioisotopes; Herbert Mapes  helped design,  construct,
and operate sampling equipment at BNL and  in the Adirondacks;
Philip Walker directed the studies on pine  pollen  and operated
samplers in the Adirondacks; Clifford Lloyd operated samplers
on Whiteface Mountain; William Glider operated samplers at  Blue
Mountain Lake and Raquette Lake.   At Brookhaven  National  Laboratory:
Lester Cohen, Thomas Dowlearn, John McNeil, Anna Kokinelis,
Paul Michael, Joyce Tichler, and Joan Glasmann assisted in  many ways.

Among many who offered assistance that was  gratefully received:
Bruce Darling, caretaker of the Sagamore Conference Center;  Patrick
Collins, caretaker of the Minnowbrook Conference Center;  Jerry
Swinney, Director of the Adirondack Museum; Ray Falconer,  Director
of the Atmospheric Sciences Research Center on Whiteface  Mountain;
Harvey Prins, Stanley House, Jean Hoffman,  and Carl  Kundell  of the
Division of Air Resources, New York State  Department of Environmental
Conservation.
                               vm

-------
                             SECTION I

                            CONCLUSIONS


These studies on the occurrence, transport, and dispersion of
atmospheric pollen include the data from several  separate, but related,
projects.  Conclusions reached from each study are given in the
pertinent Sections of this report.  However, a few summarizing remarks
may be in order here.

Samplers relying on gravity or impingement do not yield data which
reliably indicate concentrations per unit volume of air.  Rotating
impactors are usually satisfactory for general use.  Filter samplers
are highly variable in efficiency for particles in the pollen size
range unless isokinetic sampling is attained.

Concentrations of pollens from area and point sources normalized to
100% at 1 m from the source indicate that at 60 m from the source in
the downwind direction, over open terrain, ragweed pollen averages
±9%, timothy ±6%, and corn ±1%.   Extrapolation suggests that ±99%
of ragweed pollen deposits within a km.

Dispersion into a forest from upwind sources is similar to that
over open terrain.   Concentrations within the forest decrease at a
faster rate than in the open.   Loss of material takes place by
impaction near the forest edge and in the treetops and by deposition
to the ground.  Most loss is to  the foliage rather than the ground.

Distant from local  sources, vertical  pollen profiles up to 100 m show
variability from day to day but, when averaged over one or more pollen
seasons, little change with height is found.

-------
Narrow strips of vegetation markedly alter the low-level pollen
concentration patterns.  A pronounced concentration maximum is found
in the area immediately downwind of a hedge (wake region).   Similar
behavior downwind of other obstacles may be inferred.  Allergic
individuals should avoid wake regions when aeroallergens are present.

Large quantities of pollen are transported in orderly fashion from
their source regions, but pollen often travels in large, discrete
clouds.

Ragweed pollen concentrations in a source-free forested area several
hundred km from the sources of this pollen are mostly low,  but they
may reach high concentrations for short periods, especially in openings
in the forest.

Ragweed eradication over a small area surrounded by a large area with
many ragweed plants is unlikely to result in a significant  reduction
in pollen concentrations.  However, eradication of abundant ragweed
in an area surrounded by an extensive ragweed-free area should cause
a marked reduction in local ragweed pollen concentrations.

The pollens of most flowering plants are not found in the atmosphere
far from their sources.  They are generally ignored when searching
for possible causes of pollinosis.  However, some of these  entomophilous
species at times emit pollen that becomes airborne in appreciable
amounts close to the source plants.  In some situations, a  person
allergic to such pollen might be exposed to concentrations  sufficiently
hiqh to cause pronounced discomfort.

-------
                            SECTION II

                          RECOMMENDATIONS
OPERATING PROCEDURES
Until an isokinetic sampler is developed for use under atmospheric
conditions, rotating impaction samplers should be used for sampling
airborne pollens, large spores, and other particles above about six
microns in diameter.  Smaller particles should be sampled with suction
type samplers.

Sampling surfaces of impaction samplers should be shielded during
nonoperatinq periods to avoid wind impaction.

In pollen surveys, species should be reported  by their scientific names,

A volumetric measurement should be used for pollen counts reported to
the public and the time period during which the count was taken should
be specified.
FURTHER RESEARCH
Collection efficiencies of commonly used samplers should be determined
for at least the more common airborne pollens and other spores  as a
function of particle characteristics, wind speed, and other pertinent
variables.

Research to improve sampling techniques should continue with emphasis
on developing instruments that minimize variation in  efficiency with
particle characteristics (size, shape,  density)  or with meteorological
conditions (wind speed, turbulence, humidity, precipitation).

-------
Development, evaluation, and calibration of the isokinetic sampler for
use in the free atmosphere, which was designed and built under this
program, should be completed.

Development, evaluation, and calibration of the automatic grab sampler,
which was designed and built under this program should be completed.

Samplers should be designed whose efficiency is not significantly
changed by precipitation.

Studies should be conducted to find methods of separating morphologically
similar pollens such as: Comptania from Betula and Chenopodiun from
Amaran thus.

Studies should be conducted to improve methods of separating freshly
emitted from older airborne pollens.

The effectiveness of local ragweed eradication should be adequately
te?t,fK' with sufficiently long before and after sampling in areas
of low, medium, and high concentrations.

Methods for predicting ragweed pollen concentrations should be developed
further and tested in several geographical  areas which differ with regard
to pollen concentrations, location of sources, and meteorological
conditions.

Temporal and spatial variability in airborne pollen concentrations should
be studied at. representative locations in a number of cities in diverse
geographical areas.

The occurrence, concentration, and behavior of common aeroallergens in
and around inhabited structures should be studied as functions of

-------
building size, shape, and function; ventilation and air cleaning
devices; and of particle characteristics and meteorological  parameters.

Long distance pollen transport should be studied by means of a systematic
sampling program to determine the kinds and concentrations of pollen
in the upper air, their temporal and spatial variability, trajectories,
and their relationship to meteorological variables.

Medium-range transport and dispersion of aeroallergens should be
further studied by use of aircraft-mounted isokinetic samplers to
determine relationships with source areas, meteorological conditions,
and other variables with emphasis on transport into source-free areas.

Studies should be conducted to determine if pollen grains equipped with
air bladders are transported through the air with the bladders inflated
or deflated and whether this varies with ambient conditions  such
as humidity.  Similar information should be obtained on thin-walled
pollen grains which could travel either in spherical  form or partially
collapsed.

Studies should be conducted to determine pollen washout efficiency
as a function of particle characteristics, rainfall rate, and drop
size distribution.

Studies should be conducted to determine impaction efficiency of
various vegetative elements as a function of pollen characteristics
and meteorological  conditions.

Studies should be conducted to determine removal  efficiencies of pollen
from vegetative and other surfaces as a function  of pollen characteristics,
wind speed,  and other variables.

-------
Studies should be conducted to determine  more  precisely  the  relationships
between airborne pollen concentrations and pollinosis  in  susceptible
individuals.

Possible synergistic effects between aeroallergens  and other air
pollutants should be explored.

The possibilities of biological  control of ragweeds should be
explored.

-------
                            SECTION III

                           INTRODUCTION
This research was conducted cooperatively by staff members of the
New York State Museum and Science Service and Brookhaven National
Laboratory.  It was supported in large part by Research Grants from
the National Institute of Allergy and Infectious Diseases (E-1958)
and the Environmental Protection Agency (R-800677).

The study areas were all in New York State: primarily at Brookhaven
National Laboratory on Long Island, at Albany, at Saratoga Springs,,
and in the Adirondack Mountains.

The results of these investigations appear in thirteen Progress Reports,
numerous journal articles, books, and special printed ormultilith
documents.  These may be obtained, while the supply lasts, from Botany
Office, New York State Museum and Science Service, Albany, New York
19234, except those having Raynor the senior author which are from
Meteorology 051, Brookhaven National Laboratory, Upton, New York
11973.  For topics described in detail in these publications, only
summaries appear here.  References to pertinent research by others
that appear in these publications are net repeated.  Research not
reported elsewhere, except for preliminary data in one or more Progress
Reports, is presented in greater detail, including references to other
authors.

These studies, which are not being duplicated by any other research group,
have resulted in a wealth of experimental  data, an increased understanding
of many phases of pollen dispersion, and the development of useful  techniques
for predicting pollen concentrations.  The results have practical value

-------
to allergists and public health  officials  concerned with  pollinosis
and weed control  programs and to all  concerned  with dispersion of
airborne participate matter.   The condensed  versions  appear  in suitable
serial publications, the supporting data  in  Progress  Reports or  in
BNL Publications, as indicated in the bibliographic references for
each section.

Research on the occurrence, transport,  and dispersion of  pollen  in the
atmosphere has required extensive studies  on the  sampling techniques
best suited for the various phases of the  work.1   Also, we have  had
to be concerned with the whole field of aerobiology.2

REFERENCES
1.  Ogden, E. C., G. S. Raynor,  J. V. Hayes, D. M. Lewis,  and
    J. H. Haines.  Manual for Sampling  Airborne Pollen.   New York,
    Hafner Press, 1974.  182 pp.

2.  Raynor, G. S., E. C. Ogden,  D. M. Lewis, and  J. V. Hayes.  Aerobiology.
    Submitted for M. B. Rhyne, ed.  Childhood Hay Fever,  to be
    published by C. C. Thomas, Springfield,  Illinois.
                                 8

-------
                            SECTION IV

          INSTRUMENT DEVELOPMENT, TESTING, AND COMPARISON


 IMPINGEMENT SAMPLERS
 Durham Sampler
 This sampler, also known as the gravity slide sampler, consists of a
 mount for positioning a glass microscope slide between two horizontal
 9-inch circular disks.  Pollen collection is by gravitational  settling
 and by turbulent impingement to the adhesive surface of the slide.
 It was adopted as the standard pollen sampler by the American  Academy
 of Allergy in 1946.  It is still in widespread use.  Nearly all of the
 "pollen counts" reported by news media are obtained by its use.

 Studies conducted at eight levels on a 420-foot meteorology tower
 indicated that pollen capture is determined by several factors in
 addition to the concentration of pollen in the air.1

 The amount of pollen captured is influenced by the amount of air passing
 over the slide and also by the orientation of the slide with respect
 to wind direction.

 A comparison of Durham samplers with an intermittent rotoslide sampler
 indicated that variations between the counts from the two sampling
methods were not orderly nor predictable.2  No single factor would
 convert the counts from the Durham sampler to the concentrations
measured with the rotoslide.

Modified Durham Samplers
 Several modifications of the Durham sampler were designed and  tested.3^
 A standard orientation with wind direction was obtained with a wind
 vane.   However, the slide holder is aerodynamically poor and variations
 in catch caused by wind speed and turbulence were still present.   A
 streamlined circular holder accepting a square glass slide offered  two

-------
improvements: equal orientation with any air flow and less modification
of approaching air currents.  It retains the other disadvantages.
Durham samplers and modified Durham samplers were compared at several
different locations.  Four sets of modified Durham samplers, each
containing four units mounted one above the other, were exposed at
four heights on the Brookhaven meteorology tower.  Because of wide
differences between adjacent samplers, unknown variations in catch
with wind speed and turbulence, and impossibility of determining the
volume of air sampled, this sampler appears to be unsuitable for
sampling airborne particles.

Deposition Samplers
For most of the sampling arrays, deposition was measured by one or two
microscope slides placed on a small  board at the base of each sampling
station.  The upper side of the slides was given a thin coating of
silicone stopcock grease.  This adhesive is suitable for dry weather.
During long exposures when rain might occur, AEC fallout paper proved
to be better, as the particles are not as easily dislodged by water.5


WIND IMPACTION SAMPLERS
Slide-edge-cylinder Sampler
This vane-mounted impaction sampler was designed, tested, and perfected
by several modifications.  It combines the principle of wind impaction
on the side of a cylinder with the ease of reading a microscope slide
and was used in quantity during several seasons.6'7'8  It consists
of a glass slide mounted vertically between two aluminum bars on the
front of a vane (Fig. 4-1).  The leading edge of the slide serves  as
the sampling surface.  The bars present a 1/4-inch-diameter cylindrical
surface to the wind.  As efficiency varies with wind speed,  an
anemometer must be used with it.  For open areas (nonforested), one or
a few anemometers can serve many samplers.  This sampler is inexpensive
and easy to use.
                                10

-------
Wind tunnel tests9 with a slide-edge-cylinder sampler mounted beside
a filter sampling isokinetically showed excellent agreement after
applying a calculated efficiency correction.

The theoretical efficiencies for impaction of ragweed pollen (±20u)
on the 1-mm slide in the 1/4-inch holder is approximately as follows:
Air speed  (mps)         J_  _2_  J$_  _4_   5   _6_  _7_  _8_   10    15
Efficiency in percent   43   63   73   79   84   86   88   90    90    95
 Fig. 4-1     Slide-edge-cylinder sampler.

                                 11

-------
ROTATING IMPACTION SAMPLERS
Because the efficiency of stationary impaction samplers is low and highly
variable over the range of wind speeds most commonly encountered in
low-level sampling, the rotating impactor has come into general  use.
Moving the collector through the air is equivalent to allowing the air
to move past the collector.  Sampling at a constant rotational rate with
a fixed radius of rotation gives a constant average linear speed to
the collecting surface which therefore samples a known volume per unit
time.10

Rotoslide Sampler
After many experiments with the rotorod sampler, originally designed by
personnel at the Stanford University Aerosol  Laboratory, the rotoslide
sampler11 operating on the same principle but taking the samples on the
edge of standard microscope slides was designed, tested in the wind
tunnel and in the field, and used extensively.  It consists of two
upright slide holders at the ends of a horizontal  crossarrn mounted on
the upright shaft of an AC motor.  The rotational  speed is about
1600 rpm with a linear speed about 10 mps.  The volume sampled is
about 3.5 m3/hour.  This sampler is essentially independent of wind
speed and d''-action.  The theoretical impaction efficiency of a 1 mm-
diameter cylinder traveling at 10 mps is ±96% for ragweed pollen.
However, the slide edge is not a true cylinder, turbulence is caused
by the rotary action, and the retention of impacted particles is less
than 100%.  Wind tunnel tests indicated an impaction-retention
efficiency of ±68% at. 0 speed and 49% at 10 mps for ragweed pollen.
This is a straight line ratio.  For the usual outdoor average wind
speeds during the ragweed season (±2 mps), we suggest using an
efficiency of 64% when wind speeds are not known.

Intermittent Rotoslide Sampler
If the rotoslide is operated continuously for more than two or three
hours, overloading occurs in most situations.  The intermittent rotoslide
                                12

-------
Fig. 4-2     Intermittent  swing-shield rotoslide sampler.
                               13

-------
is designed to take pollen samples representative of the average
concentration over a 24-hour period by automatically operating
periodically for short periods.  A timing device energizes the motor
for one minute in each twelve, giving a total sampling time of two
hours each day.  As the slide edge is an efficient impactor of pollen
even when not rotating if it happens to stop facing the airflow, it is
necessary to shield it from the wind when not rotating.  The original
model,11 still in widespread use, employs a cylindrical shield,
activated by a reversible motor and timer.  During rotation, the shield
is pulled down to expose the slides to normal airflow.  When rotation
ceases, the shield rises to shelter them from the wind.

Since mechanical  difficulty was occasionally experienced with the
shield-raising mechanism, a more reliable device, the swing-shield12
was designed to eliminate this problem (Fig. 4-2).  The shielding
plates are pivoted on pins through the top and bottom of the
slide-holder frame; the curved ends of the plates are held in
front of the slide edges by spring tension.  Upon rotation, the curved
ends are drawn away from the slide edge by a combination of centrifugal
force and air pressure against the tail section of the plate.  Rotoslides
using this shield are particularly well adapted for sequential sampling
when used with an appropriate timing device.

Sequential Rotoslide Samplers
Several versions  of a sampler employing rotating impactors operating
sequentially were designed, tested, and used in experiments requiring
sequential data.6  To shield the idle units from wind impaction, all
were housed in a  box, with one at a time protruding for operation.
Drive mechanisms  tested included: air jet, magnet, and direct drive.
In most of the studies, a direct flexible drive was used with a delay
relay and a resistance allowing the drive motor to reach half speed
before the full power comes on.
                                14

-------
4-3     Sequential  swing-shield rotoslide  sampler.
                          15

-------
Development of the swing-shield slide holder for the rotoslide sampler12
allowed the construction of rotoslide samplers for sequential  operation
with several advantages over existing machines.   A six-unit,  lightweight
(30 Ibs), portable sequential swing-shield rotoslide sampler  was designed.
Two such samplers were used independently or in  tandem.13  Four 13-unit
sequential swing-shield rotoslide samplers were  constructed for use
in a study of variation in pollen concentrations with time.  They
consist of twelve motors with swing-shield slide holders mounted
on a triangular frame of angle aluminum supported by three legs.  A
thirteenth motor is mounted in the center and a  control  box
containing the timing mechanism below the center of the  frame.   The
whole assemblage is protected by a rain shield (Fig. 4-3). Two
samplers were timed so that the twelve motors operated sequentially
for five minutes each while the central one sampled continuously for
an hour.  In the other two samplers, the twelve  units sampled
sequentially for two hours each.  The central unit sampled intermittently
Fig. 4-4     Examination of a slide edge sample.
                                16

-------
for twenty-four hours on a cycle of one in each twelve minutes.  For
another study, two of the samplers were operated in tandem, taking
consecutive 30-minute samples.  They were re-set by the operator twice
in each 24-hour day.

Pollen samples on the edges of microscope slides proved to be easy
to handle and prepare for study.11  A slide positioner was designed
and used for holding the slides on edge under the microscope.
Identification and counting of pollen grains on more than 100,000
slide-edge samples presented no unusual difficulties (Fig. 4-4).
SUCTION SAMPLERS
Hirst Spore Trap
Samplers in which air, containing material to be sampled, is drawn into
an entrance by suction are used for many air sampling purposes.  Techniques
for collecting and counting particles .drawn into suction samplers have
been subjects of extensive research and experimentation, but little
attention has been given to the problem of getting a representative
Cample into the entrance.  This is partially due to the fact that suction
samplers are mostly used for submicron-sized particles whose entrance
effici-.Mcy is normally high and uniform.  Larger particles, such as
pollen, tend to deviate from the air stream entering the sampler if the
air is forced to change direction or speed.

Among the available suction samplers capable of yielding sequential data,
the Hirst spore trap14 appeared to be the best.   It was used extensively
by us for obtaining diurnal patterns of emission in our dispersion fields15
and for patterns of occurrence elsewhere.  It compared favorably with
the sequential rotoslide samplers, although its time discrimination was
not as precise.
                                17

-------
The efficiencies for sampling ragweed pollen (^20y)  at different
ambient wind speeds were determined to be approximately as follows:
Air speed   (mps)         1    2    4    6    8    10
Efficiency in percent   89   68   58   65   100  124

Filter Samplers
Filter samplers, as commonly used for samoling particles less  than  five
microns in diameter, are not suitable for the larger pollen grains  unless
isokinetic sampling can be attained or at least approximated.   Filter
samplers were used extensively in our wind tunnel  tests under  isokinetic
sampling conditions.

The effect on the entrance efficiency of a filter holder caused by
variations in air speed, flow rate, angle between the air flow and  the
filter holder, and particle size was studied in the  wind tunnel.16
The efficiency varied with all parameters from less  than 1% at the  highest
wind speed (7 mps) and lowest, flow rate (6.4 liters/min) to more than
100% at forward angles where impaction aided suction.  An empirical
formula was developed to model the pertinent parameters.

From this formula, it was predicted that the entrance efficiency of an
orifice at a 45-degree angle to the air stream should be unity and  should
not vary with wind speed, particle size, or flow rate.   Tests  were  made
with 6- and 20-micron diameter particles at wind speeds from 0.75 to
7.0 m/sec and at flow rates of 12.7 and 25.4 liters/min using  both
fixed and vane-mounted filter holders as test orifices.  Early results
confirmed that entrance efficiency at the 45-degree  angle is relatively
constant over the range of experimental conditions in contrast to
results at other angles.  A vane-mounted filter sampler with the
entrance at a 45-degree angle was constructed.17  Later tests, however,
gave more variable results and more testing is needed before the
sampler can be recommended.
                                18

-------
 ISOKINETIC SAMPLERS
 Several existing volumetric samplers have a high efficiency in retaining
 pollen that enters the sampling intake.  However, a significant percent
 of the pollen originally in the air drawn through the intake may deviate
 from this air due to its momentum in its original direction.  This error
 can be overcome only by sampling isokinetically.

 Variable Air Flow Isokinetic Sampler
 A sampler was designed and constructed (Fig. 4-5) which will change
 its flow rate rapidly and automatically to conform to changes in the
 wind speed while the sampling intake is oriented into the wind by a
 vane.  In both field and wind tunnel tests, the sampler indicated
 lower concentrations than comparison samplers.  Reasons for the
 discrepancy have not yet been determined but further tests are planned.
Fig. 4-5     Variable air flow isokinetic sampler.
                               19

-------
Aircraft Isoklnetic Sampler
A light-aircraft-mounted isokinetic sampler was  designed for obtaining
accurate samples of particulate matter in the lower atmosphere.18   The
sampler head is operated in undisturbed air under the winq  and is  drawn
into the cabin along a track for changing filters.   A battery-powered
high volume sampler in the cabin serves as the air  moving device.   The
sampler can draw as much as 0.72 m3/min and match aircraft  air speeds
to 38 m/sec at standard temperature and pressure.  The sampler was  used
two seasons for measuring the vertical  and mesoscale distribution  of
airborne pollens.19
AUTOMATIC GRAB SAMPLER
Grab sampling consists of capturing a known volume of air in  a  way  that
all particles and gases in the volume are taken with it.   A grab sampler
was designed and built (Fig.  4-6)  which periodically encloses a known
volume of air, immediately removes the contaminants of interest, and
repeats the cycle for as long as desired.21  This  sampler is  easy to
operate and is believed to trap atmospheric contaminants  of any size
w'th equal efficiency.  Further tests are necessary to confirm  its
efficiency and utility.
WIND TUNNEL TESTS
                                    •
A wind tunnel was constructed (Fig. 4-7)  to study the air flow patterns
over the various samplers.  Wind speeds may be varied from zero to
45 mph.  Oil fog smoke is produced by a small  generator and introduced
into the tunnel to allow visual  observation and photography of the air
flow past the objects under test.  The true pollen concentration in
the tunnel is measured by a filter sampler sampling isokinetically.
Pollen is emitted into the closed wind tunnel  room and mixed
by four floor fans (one in each  corner) and the wind tunnel  fan.  The
extent of the mixing can be checked by an array of seventeen filter
samplers in the form of a cross  operated in the working section of

                                20

-------
the tunnel.  Once the air enters the honeycomb air straighteners  just
inside the entrance, no further mixing is  possible.
   Fig. 4-6     Automatic grab sampler.

                                   21

-------
Wind tunnel studies confirmed our theories regarding the causes of errors
in the samples from the Durham sampler.  Tunnel tests indicated that the
actual impaction efficiency of the slide-edge-cylinder sampler is
essentially the same as the theoretical efficiency.  Numerous wind
tunnel tests were made to determine the sampling (impaction-retention)
efficiency of the slide-edge-cylinder sampler,9 the rotorod sampler,21
the rotoslide sampler,11 the Hirst spore trap,21 and several  other
sampling devices.
Fig. 4-7     Wind tunnel at Brookhaven National Laboratory.
CONCLUSIONS
Methods  for determining concentrations of particulates in the atmosphere
include  impaction, filtration, liquid impingement, electrostatic
attraction, and thermal precipitation.  However, methods of removing
                                22

-------
particles from the free air and getting them onto a sampling surface
or into a sampler entrance are limited to deposition, impaction,  and
suction.  Impaction is the preferred method for large particles such
as pollen.

Because the efficiency of stationary impactor samplers is low and
highly variable over the range of wind speeds commonly encountered,
the rotating impactor has come into general use.

Data from samplers relying on gravity or impingement (such as the
Durham sampler and modifications of it) may not be converted to
grains per unit volume of air.

None of the samplers in current use is dependable during wet weather
(rain and fog).

REFERENCES
1.  Ogden, E. C.rand G. S. Raynor.   Field Evaluation of Ragweed Pollen
    Samplers.  J Allergy 31(4):307-316, July-August 1960.

2.  Hayes, J. V.  Comparison of the Rotoslide and Durham Samplers in
    a Survey of Airborne Pollen.   Ann Allergy 27:575-584,  November 1969.

3.  Ogden, E. C.  Tagging and Sampling Ragweed Pollen.  Progress Report  1.
    New York State Museum and Science Service, Albany,  N.Y.   May  1960.
    32 p.

4.  Ogden, E. C., G.  S. Raynor, and J.  M.  Vormevik.   Travels of Airbbrne
    Pollen.  Progress Report 5.  New York State Museum and Science
    Service, Albany,  N.Y.   May 1964.   30 p.

5.  Lewis, D. M., and E.  C.  Ogden.   Trapping Methods  for Modern Pollen
    Rain Studies.  In:   Handbook  of Paleontological  Techniques,
    B. Kummel and D.  Raup (eds.).  San Francisco,  W.  H.  Freeman & Co.,
    1965.   p. 613-626.
                               23

-------
 6.   Raynor,  6.  S.»  M.  E.  Smith,  I.  A.  Singer, and E. C. Ogden.
     Pollen Sampling and Dispersion  Studies at Brookhaven National
     Laboratory.  Air Pollut Contr Ass  J  11:557-561 & 584,
     December 1961.

 7.   Raynor,  G.  S.,  J.  V.  Hayes,  and E.  C. Ogden.  Experimental Data on
     Ragweed Pollen  Dispersion  and Deposition From Point and Area Sources.
     Brookhaven  National  Laboratory, Upton, N.Y.  Publication Number
     BNL 50224 (T-564).   February 1970.   33 p.

 8.   Ogden, E. C.,  and 6.  S.  Raynor. Travels of Airborne Pollen.
     Progress Report 3.   New York State Museum and Science Service,
     Albany,  N.Y.  July 1962.   8  p.

 9.   Ogden, E. C., G.  S.  Raynor,  and J. M. Vormevik.  Travels of
     Airborne Pollen.  Progress  Report 4.  New York State Museum
     and Science Service,  Albany, N.Y.  June 1963.  12 p.

10.   Raynor,  G.  S.   Sampling Particulates With Rotating Arm Impaction
     Samplers.  Brookhaven National  Laboratory.  (Presented at Proc.
     Workshop/Conference I.  Ecological  Systems Approaches to Aerobiology.
     Kansas State University, Manhattan.  January 6-9, 1972.)  US/IBP
     Aerobiology Program Handbook 2:82-105.  University of Michigan,
     Ann Arbor.

11.   Ogden, E. C.,  and G.  S.  Raynor. A New Sampler for Airborne  Pollen:
     The Rotoslide.   J Allergy  40(1):1-11, July 1967.

12.   Raynor,  G.  S.,  and E. C. Ogden. The Swing-shield: An Improved
     Shielding Device for the Intermittent Rotoslide Sampler.
     J Allergy 45:329-332, June 1970.

13.   Ogden, E. C.,  G.  S.  Raynor,  and J. V. Hayes.  Travels of Airborne
     Pollen.  Progress Report 11.   New York State Museum and Science
     Service, Albany, N.Y.  February 1971.  25 p.

                                 24

-------
 14.  Hirst, J. M.  An Automatic Volumetric Spore Trap.   Ann  Appl  Biol
     39(2):257-265, June 1952.

 15.  Ogden, E. C., J. V. Hayes, and G. S. Raynor.  Diurnal  Patterns
     of Pollen Emission in Ambrosia, Phleun,  Zea, and Rioinus.
     Amer J Bot 56:16-21, January 1969.

 16.  Raynor, G. S.  Variation in Entrance Efficiency of a Filter
     Sampler With Air Speed, Flow Rate, Angle and Particle Size.
     Amer Ind Hyg Ass J 31:294-304, May-June  1970.

 17.  Raynor, G. S.  Entrance Efficiency of a  Vane-mounted Filter
     Sampler at a 45-Degree Angle to the Wind.   Brookhaven National
     Laboratory, Upton, N.Y.  Publication Number BNL 15909.   1971.
     10 p.

 18.  Raynor, G. S.  An Isokinetic Sampler for Use on Light Aircraft.
     Atmos Environ (Great Britain).  6:191-196,  1972.

 19.  Raynor, G. S., J. V.  Hayes, and E. C.  Ogden.  Mesoscale Transport
     and Dispersion of Airborne Pollens.  J Appl  Meteorol 13(l):87-95,
     February 1974.

 20.  Ogden, E. C., G. S. Raynor, and J. V.  Hayes.  Travels of Airborne
     Pollen.  Progress Report 8.  New York  State  Museum and  Science
     Service,  Albany, N.Y.   February 1968.  36  p.

21.  Ogden, E. C., G. S. Raynor, and J. V.  Hayes.  Travels of Airborne
     Pollen. Progress Report 9.   New York State Museum  and Science
     Service,  Albany, N.Y.   March 1969.  27 p.
                                25

-------
                            SECTION V

                        TAGGING EXPERIMENTS
This phase of the project was concerned with several  methods of marking
pollen grains so that studies could be made on the travels of the
pollen from a known source, including simultaneous emissions from
different positions.
DYES
Experiments were conducted on the preparation of dyed pollen grains,
lyoopodiim spores, and fungus spores for dispersion and deposition
studies.1  The particles are added to a saturated aqueous solution of
a suitable dye with a small quantity of wetting agent and stirred
until the grains are colored and the clumps are separated.   The staining
solution containing the particles is released into the atmosphere
by compressed air-operated atomizing nozzels.2  The liquid quickly
evaporates, leaving the dry particles subject only to atmospheric
motions and their normal settling velocity.  When two or more colors
are to be released from the same container, they are removed from
the staining solution and mixed in a liquid in which the dyes are
not soluble.  Dyed pollens were used extensively as tracers in our
experiments designed to study the behavior of particles in a forested
region.  Dyed pollens also were useful in other point source experiments
and in wind tunnel tests.

Many experiments were performed to determine whether ragweed pollen
could be labeled by vital dyes in the living plants.3  These studies
were conducted with two species of Ambrosia in the greenhouse and in
the field.  Six nonfluorescent and six fluorescent dyes were used.
Methods of application included: saturating the soil; adding dyes to
aerated, buffered, nutrient solutions; introduction directly into the
                               26

-------
conducting system; and immersing inflorescences in the dye solutions.
It was found that dyes in an aqueous solution of 1000 parts per million
could be introduced directly into the xylem with but slight injury to
the living plant.  These dyes moved rapidly to the flower area.  Among
the dyes, rhodamine B was most easily detected because of its bright
red color under white light and its bright red glow under ultraviolet
light.  Also, there was less chance of confusion between the red rays
of the fluorochrome and the yellow-green rays due to autofluorescence
of the pollen.  Although the dye moved rapidly to the flowers, it did
not enter the anthers.  Experiments with immersion of flowers in
alcohol and xylene to break the resistance of the cell membranes to
the dye indicated that a strong xylene mixture or a long immersion
period resulted in deeply dyed pollen but, in all such cases, the
flowers drooped and no pollen was shed.  As it appeared that vital
dyes offered little promise for labeling ragweed pollen in the anthers
in such a way that normal  pollen shed is not modified, these studies
were abandoned in favor of other more promising techniques.
RADIOISOTOPES
Two Isotopes were chosen for these experiments with the two ragweed
species: radiophosphorus (32P) with a half-life of 14.3 days  and
radiosulfur (35S) with a half-line of 87.1  days.  They were not
toxic to the plants in the amounts used, were no hazard to trained
personnel, and they traveled to the pollen  without excessive  loss.
The labeled pollen created no hazard in and around the area.   The
half-life was long enough to permit processing the samples and short
enough so a contaminated area or expensive  equipment need not be
abandoned for an unreasonable length of time, and there was sufficient
radiation from small  amounts to allow detection with available
methods.

The experiments3 were mostly conducted with 32P, using tlmbrosia aptemisiifolia
in the greenhouse in  hydroponic solution to which the isotope was
added and using both  species in the field with the isotope introduced
                               27

-------
into the main stem (Fig.  5-1).   The relative  amounts  of  radiation  from
different parts of the plants were determined by  Geiger  counters and


Fig. 5-1    Cotton wrap technique for introduction of a
            radioisotope into a ragweed plant.
                                28

-------
radiation sealers.  However, the amounts of radiation among individual
pollen grains varied so greatly that autoradiographs were required for
recognition of tagged grains (Fig. 5-2).  These are easily prepared.3
Radiosulfur acted much like radiophosphorus, going as readily to the
pollen.4  Its half-life, which is over six times as long, allowed
longer exposures and less haste in taking and processing samples.
Also, the traces from 35S are shorter than those from 32P, an advantage
when counting crowded grains.

Although radioisotope tagging offered a tracer technique, it was
not necessary for our dispersion studies as the production of preseason
ragweed proved to be easy and satisfactory.
PRESEASON RAGWEED POLLEN
As giant ragweed (A. trifida) normally reaches anthesis one or two
weeks earlier than common ragweed (A.  artemisiifolia], the former
was used for the production of preseason pollen.1*  Greenhouse and
field experiments indicated that increasing the periods of darkness
for five or more days would initiate flower bud formation.  About
300 lignt-tight black cloth bags were used for covering the individual
plants for the preseason pollen dispersion field.  However, after
it was noticed that allowing the plants to remain in small (4-inch)
pots until they are rootbound will  initiate early production of
flowers, this technique was used during succeeding seasons.  The
short-day (actually long-night) technique was an alternate method and
used again one year when animals got into the fenced field and
destroyed the rootbound planting (Fig. 5-3).
CONCLUSIONS
Vital dyes offer little promise for tagging pollen.  Radioisotopes
may be used for special studies.  However, pollen emitted naturally
out-of-season or in quantities far exceeding background pollen and
prestained artificially emitted pollen are more satisfactory for
most dispersion studies.

                                29

-------
    B
                  ,
                  •»"*,*  » ,
                   - *TS  4*'*
                  ..,/**'«
                  -•'
                                     *   ^vv * •  % * ^
                                       »,„ , '>«%'.._ » !'" '
                                      ,  .".<>- IS
Fiq. 5-2     Autoradiographs of ragweed pollen. (A and B)   Pollen

from plant  injected  with  32P.  (C and D)  Pollen from plant injected

with 35S.   (B and  D)   Grains on left show nuclear tracks,  grains on


right are not recognizably tagged.
                                  30

-------
Fig. 5-3     Ragweed field for preseason  pollen,
REFERENCES
1.  Raynor, G.  S., L. A.  Cohen,  J.  V.  Hayes,  and  E.  C.  Ogden.  Dyed
    Pollen Grains and Spores as  Tracers  in  Dispersion and  Deposition
    Studies.  J Appl  Meteor 5(5):728-729, October 1966.

2.  Raynor, G.  S., and M.  E. Smith.   A Diffusion-deposition Tracer
    System.  Brookhaven National  Laboratory,  Upton,  N.Y.   Publication
    Number BNL 859(T-343).  April  1964.   17 p.

3.  Ogden, E.  C.   Tagging  and Sampling Ragweed  Pollen.  Progress
    Report 1.   New York State Museum and Science  Service,  Albany, N.Y,
    May 1960.   32 p.

4.  Ogden, E.  C., and G.  S. Raynor.   Tagging  and  Sampling  Ragweed
    Pollen. Progress  Report 2.  New  York State  Museum and  Science
    Service, Albany,  N.Y.   December 1961.   14 p.
                                31

-------
                            SECTION VI
       DISPERSION AND DEPOSITION STUDIES OVER OPEN  TERRAIN
AREA AND POINT SOURCES
Ragweed Pollen
Dispersion of ragweed (Ambrosia]  pollen emitted naturally from
cultivated plants in 10 circular area sources (Figs.  6-1  and 6-2)
of four sizes and artificially from point sources was studied over
a four-year period.1"3  Concentrations were measured  by slide-edge-cylinder
samplers mounted on 20° radii at four heights (0.5-4.6 m) and four
or five distances from the sources to a maximum of 69 m.   An anemometer
was mounted at each height.  Deposition was measured  by greased
Fig. 6-1
Aerial view of area source of ragweed pollen.
                               32

-------
microscope slides on the ground.  A Hirst spore trap was mounted above
the center of each source to measure the variation of pollen emission
with  time.  Differences in dispersion patterns between point and area
sources were analyzed.  Dispersion patterns, normalized centerline
concentrations, cross-wind integrated concentrations, plume widths,
plume heights, mass flux, and deposition were presented as functions
of distance and related to source size and meteorological variables.3
During the normal ragweed pollination season, the contribution of pollen
from other sources was determined from measurements on the upwind side
of the plot and this quantity subtracted from all downwind measurements
before further analysis.

Point sources (Fig. 6-3) consisted of compressed-air-operated atomizing
nozzles4 which sprayed a known amount of prestained ragweed pollen in
a water suspension.5  Output rates varied from 1  to 5 x 10s grains/sec
but were constant during each run.  Emissions lasted from 24 to 60 min.
The point sources were 0.5 to 3m above the ground.

Concentrations from area and point sources normalized to 100% at 1 m
from the source indicated that from 25 to 65% remain airborne at 60 m.
L/lrapolat.ion of the data to greater distances indicates that about
1% of the po'Hen grains remain airborne at 1000 m.  Since much of the
eastern two-thirds of the United States constitutes a ragweed pollen
source region and a single plant can release more than a million grains
per day during the height of the pollination season, the ubiquity of
airborne ragweed pollen during late summer is not surprising.

The dispersion of pollen from these ragweed sources was also studied
to determine the effect of contributions from such sources upon the
pollen concentrations originating in more distant areas.6  Since ragweed
pollen is uroduced throughout a large region, concentrations measured
at any given location represent contributions from many sources at
various distances along the past trajectory of the air sampled.  A
local  source may produce concentrations several  orders of magnitude
above this background in a small  downwind region.   These concentrations
                                33

-------
decrease with distance and at some point become insignificant in
comparison to background concentrations.  Distances necessary for
concentrations to reach specified fractions of background and areas
covered by concentrations greater than specified multiples of
background are related to source size, surrounding vegetation,  and
meteorological conditions.  The effects of local sources of the
sizes studied (239 to 2000 m2) become negligible at distances greater
than several hundred to a thousand meters.

The areas within which ragweed pollen concentrations equaled or
exceeded selected values were determined from measurements around these
ten sources.7  The size of the polluted area was related directly to
source size and inversely to wind speed and the removal  efficiency of
surrounding vegetation.  Our data indicate that a ragweed source  may
pollute an area from 6 to 100 times its own area with concentrations
above 100 grains/m3 and from 11 to 300 times its own area with concentrations
above 10 grains/m3.  In the downwind direction, concentrations of
100/m3 may be found at distances of 2 to 10 times the source diameter
and concentrations of 10/m3 to distances of 3 to 20 times the source
diameter.  These are averages; day-to-day values may differ greatly.
Fig. 6-2     Ground level view of area source of ragweed pollen,
                                34

-------
Fig. 6-3     Point source pollen-dispensing apparatus,
                                35

-------
Jimpthy Pollen
Dispersion of timothy (Phleun] pollen emitted naturally from three
circular area sources of two sizes of cultivated plants was  studied
over a three-year period.8*9  Pollen was also emitted artificially
from point sources.  The sampling arrays were essentially the same
as for ragweed pollen and the results reported in the same way.

The data indicate that dispersion and deposition of timothy  pollen
(34 microns in diameter) differ little from that of ragweed  pollen
(20 microns) from similar sources despite the difference in  size.
Normalized distances were generally less for timothy than for ragweed.
They range between 2 and 5 source diameters to the 10 grains/m3
isopleth while the ragweed sources give similar concentrations as  far
as 5 to 13 source diameters.  Our data suggest that source configuration
and meteorological conditions are more important variables than
particle characteristics in determining both dispersion and  deposition
rate for ragweed and timothy pollens, but not for the heavier corn pollen.

Corn Pollen
Dispersion and deposition of corn (Zea) pollen (90-100 microns)
emitted naturally from two circular 18-m-diameter plots of cultivated
plants and from pollen emitted artificially from point sources was
studied over a two-year period.8'10  The sampling arrays were essentially
the same as for ragweed and timothy.  The results were reported  in the
same way and compared with the data for dispersion and deposition  of
ragweed and timothy pollens.  These studies show that corn pollen  is
not transported as far by the wind as smaller pollens, does  not  disperse
as widely in either the horizontal or the vertical direction and
settles to earth more quickly, much of it within the source  itself.

At 60 m from the source in the downwind direction, concentrations
averaged about }% of those at 1 m, compared to 6% for timothy and
9% for ragweed.  The total amount of corn pollen remaining airborne
at 60 m was 5% of that at 1 m, compared to 37% for timothy and 50%
for ragweed.

                                 36

-------
LINE SOURCES
Dispersion and deposition of pollens emitted naturally from 80-m-long,
crosswind, line sources of living plants  (Fig.  6-4)  and of  pollens
sprayed from a moving vehicle were studied in a set  of 45 experiments.11*12
Concentrations were measured by 276 slide-edge-cylinder samplers1
mounted in a square array which included  four heights  at each  of 69
positions.  Deposition was measured by greased microscope slides on
the ground.  Wind speeds were measured by anemometers  at each  sampling
height.

Plants used as pollen sources were giant  ragweed (Ambrosia  trifida),
summer cypress (Koohia sooparia), and castor bean (Hicinus  camunis].
The ragweed pollen was spherical with low spines and about  20  microns
in diameter; the summer cypress pollen was spherical,  smooth,  and about
30 microns; the castor bean pollen was ellipsoidal,  smooth, and
24 x 38 microns.  Each source included 240 plants grown in  12-quart
pails and placed in an 80-m-long line about one meter  wide.
Fig. 6-4     Movable line source of Koohia pollen,
                                 37

-------
Ten runs were made with ragweed,  seven with  sumner cypress,  and
twenty-three with castor bean.   Prestained ragweed pollen  was  emitted
from a vehicle moving back and forth along a line in  five  runs and
from a point source to determine  angle of spread in seven  runs.

The data which were presented as  functions of distance from  source,  pollen
type and release height included  decrease in concentration,  plume
height, mass flux, deposition,  and deposition velocity.  The results
were compared with those from previous point and area source
dispersion experiments.


REFERENCES
1.  Raynor, G. S., J. V. Hayes, and E. C. Ogden.   Experimental Data
    on Ragweed Pollen Dispersion  and Deposition  From  Point and Area
    Sources.  Brookhaven National Laboratory, Upton,  N.Y.  Publication
    Number BNL 50224(T-564).   February 1970.  33 p.

2.  Raynor, G. S., and E. C.  Ogden.  Twenty-four-hour Dispersion of
    Ragweed Pollen From Known Sources.  Brookhaven National  Laboratory,
    Upton, N.Y.  Publication  Number BNL 957(T-398).  December 1965.
    17 p.

3.  Raynor, G. S., E. C. Ogden, and J. V. Hayes.   Dispersion and
    Deposition of Ragweed Pollen  From Experimental Sources.
    J Appl Meteor 9(6):885-895, December 1970.

4.  Raynor, G. S., and M. E.  Smith.  A Diffusion-deposition  Tracer
    System.  Brookhaven National  Laboratory, Upton, N.Y.   Publication
    Number BNL 859(1-343).  April 1964.  17  p.

5.  Raynor, G. S., L. A. Cohen, J. V. Hayes, and E. C. Ogden.   Dyed
    Pollen Grains and Spores  as Tracers in Dispersion and  Deposition
    Studies.  J Appl  Meteor 5(5):728-729, October 1966.
                                38

-------
 6.  Raynor, G. S., E. C. Ogden, and J. V.  Hayes.   Effect of a Local
     Source on Ragweed Pollen Concentrations From  Background Sources.
     J Allergy 41:217-225, April 1968.

 7.  Raynor, G. S., J. V. Hayes, and E. C.  Ogden.   Areas Within Isopleths
     of Ragweed Pollen Concentrations From  Local  Sources.  Arch Environ
     Health 19(l):92-98, July 1969.

 8.  Raynor, G. S., J. V. Hayes, and E. C.  Ogden.   Experimental Data
     on Dispersion and Deposition of Timothy and  Corn Pollen From
     Known Sources.  Brookhaven National  Laboratory,  Upton, N.Y.
     Publication Number BNL 50266(1-595).   October 1970.  32 p.

 9.  Raynor, G. S., E. C. Ogden, and J. V.  Hayes.   Dispersion and
     Deposition of Timothy Pollen From Experimental Sources.
     Agr Meteor (The Netherlands).   9(1971/1972):347-366, 1972.

10.  Raynor, G. S., E. C. Ogden, and J. V.  Hayes.   Dispersion and
     Deposition of Corn Pollen From Experimental Sources.
     Agron J 64(4):420-427, July-August 1972.

11.  Raynor, G. S., J. V. Hayes, and E. C.  Ogden.   Experimental Data
     on Dispersion and Deposition of Pollens From  Low-level,  Crosswind
     Line Sources.  Brookhaven National Laboratory, Upton,  N.Y.
     Publication Number BNL 16334.   November 1971.  17 p.

12.  Raynor, G. S., E. C. Ogden, and J. V.  Hayes.   Dispersion of  Pollens
     From Low-level, Crosswind Line Sources.   Agr  Meteor (The Netherlands)
     11(1973):177-195, 1973.
                                 39

-------
                            SECTION VII

          PARTICULATE DISPERSION INTO AND WITHIN A FOREST
Particulate dispersion into and within a 10- to 13-m tall pine forest
(Fig. 7-1) was studied using stained ragweed pollen and other tracers
ranging from 14 to 54 microns in diameter.1~l|»6

During 42 tests, 119 continuous point source releases lasting 20 to
40 minutes were made from atomizing nozzels at various locations from
within the forest to 60 m upwind and at heights of 1.75 to 14 m.  In
most experiments, differently colored ragweed pollen was released
simultaneously from three locations.  Thirty-six other tests from
2 to 5 hours in length were made using pollen released naturally
from area sources of ragweed planted upwind of the forest anc1 three
with pollen from distant sources.  All tests were made during the
day with steady winds and unstable lapse rates outside the forest.
The sampling network consisted of 119 rotoslide samplers mounted at
heights from 0.5 to 21  m on 57 towers and lower-level supports along
seven rows 10 m apart extending 100 m into the forest.  Deposition
was sampled by greased microscope slides at each sampling position.
Meteorological measurements were taken in and near the forest.5

Data were classified by particle characteristics;  source type,
distance, and height; and by meteorological parameters-.- Isopleths
were drawn on scale diagrams of the sampling grid in the horizontal
at each sampling level, along each tower row in the downwind-vertical
plane and at each distance in the crosswind-vertical plane to illustrate
concentration patterns.  Changes in centerline concentration, crosswind
integrated concentration, mass flux, plume width, plume height, and
deposition were related to distance within the forest and other variables.
Results were compared to those of similar releases over open terrain
and those of previous forest dispersion studies elsewhere.
                                40

-------
DISPERSION FROM UPWIND POINT SOURCES
All particles used in the upwind point source dispersion experiments
were emitted in a liquid spray at locations  from just within the
forest edge to 60 m upwind and from 1.75 to  14 m above the ground.1,4
In most tests prestained ragweed pollen was  used and in nearly all
tests simultaneous emissions of differently  colored pollen were made
from three locations.  In a few tests other  particles were used:
timothy (Phleun] pollen, paper mulberry (Broussonetia) pollen, club
moss (lyoopodiym] spores, fern (Osmunda and  Dryopteris] spores, rust
(CTonaTtiun] spores, and copper spheres.  Seventy-two releases were
made in 27 tests.
Fig. 7-1     Experimental  forest showing some of the  towers
                                41

-------
The plume approaching the forest was broadened both vertically and
horizontally by divergence at the forest edge and flowed mainly into
the trunk space and above the forest.  Lateral spread was slow within
the forest but vertical spreading in the equilibrium region was greater
than in the open.  Particles became mixed uniformly below the canopy
while appreciable interchange took place through this layer.   Concentration
within the forest decreased at a faster rate than in the open, but
change in total mass flux within and above the forest was not
significantly different.  Loss of material took place by impaction
near the forest edge and in the treetops and by deposition within
the forest where plume movement was slow.  Most loss took place to
the foliage rather than the ground and larger particles were lost
faster than smaller ones.
DISPERSION FROM UPWIND AREA SOURCES
Dispersion of ragweed pollen into the forest from a rectangular
30- x 15-m plot of cultivated preseason and inseason ragweed plants
upwind of the forest was studied in a series of 36 tests.2»k  After
the ragweed season, three additional tests were made with background
pollen from distant sources.

The sampling grid and experimental procedures were the same as those
used in point source experiments except that test periods were longer,
usually several hours in length.  Data were classified by the same
variables and analyzed in similar fashion.  Results were qualitatively
similar except that plumes were wider due to the size of the source
and greater wind variability during the longer sampling periods.
DISPERSION FROM WITHIN THE FOREST
Dispersion from within and above the forest was studied during 47
releases in 17 tests.3)6  Sources were located at heights from 1.75
to 14.0 m at locations near the north end of the sampling grid.
Procedures were similar to those used for point source releases
                                42

-------
upwind of the forest and data were analyzed in the same manner.
These tests documented interchange through the canopy and some
channeling within the forest.

Centerline and crosswind integrated concentrations decreased more
rapidly with distance in the forest than over open terrain.
Source-height center!ine concentrations decreased most rapidly from
sources in the canopy, least rapidly from sources above the trees
and at intermediate rates from sources in the trunk space.

The plumes spread less in the crosswind direction than plumes from
similar sources over open fields but spread faster in the vertical
direction.  Rate of vertical spread was correlated with wind speed
above the forest but horizontal spread showed an inverse relationship
near the source and no correlation at greater distances.

Total, airborne particles from 1.75 m sources decreased at about the
same rate as in the open field but decreased at rates inversely
proportional  to wind speed.   Large particles were lost faster
than smaller ones.
FOREST METEOROLOGICAL STUDIES
Measurements of wind speed and temperature were taken in and near
the forest during all dispersion tests.   Cup anemometers were located
at 25 locations at selected distances within the forest at heights
from 1.75 to 21 m.  Temperature sensors  were mounted at 4 heights
within the forest and 6 heights in the field.  Data were also taken
continuously for a period to compare the micrometeorology of the
forest to that of an open location.   This study included humidity
as well as wind and temperature measurements.  Much of the earlier
data has been published.5  Measurements  of turbulence were made with
sensitive bivanes on several occasions and assisted in the interpretation
of dispersion data.
                                43

-------
With winds penetrating the forest edge, speeds in the trunk space  are
greater than those in the canopy for a distance of ±60 m.   With a
longer fetch through the forest, wind speeds vary little with height
to midcanopy.  During the day, a temperature inversion is  found
beneath the canopy and a negative lapse rate above.   During the night,
an isothermal layer or a slight lapse below the canopy and an inversion
above are typical.

REFERENCES
1.  Raynor, G. S.,  J. V. Hayes, and E. C.  Ogden.  Experimental  Data
    of Particulate  Dispersion Into and Within a Forest.  Part I:
    Dispersion From Upwind Point Sources.   Brookhaven National
    Laboratory, Upton, N.Y.   Publication Number BNL  17750.
    December 1972.   95 p.

2.  Raynor, G. S.,  J. V. Hayes, and E. C.  Ogden.  Experimental  Data
    on Particulate  Dispersion Into and Within a Forest.  Part II:
    Dispersion From Upwind Area Sources.  Brookhaven National
    Laboratory, Upton, N.Y.   Publication Number BNL  18064.
    June 1973.  39  p.
3.  Raynor, G. S.,  J. V. Hayes, and E. C.  Ogden.  Experimental  Data
    on Particulate  Dispersion Into and Within a Forest.  Part III:
    Dispersion From Within and Above the Forest.  Brookhaven National
    Laboratory, Upton, N.Y.   In press.

4.  Raynor, G. S.,  J. V. Hayes, and E. C.  Ogden.  Particulate
    Dispersion Into and Within a Forest.  Boundary Layer Meteorology.
    In press.

5.  Raynor, G, S.  Wind and Temperature Structure in a Coniferous
    Forest and a Contiguous  Field.  Forest Sci 17(3):351-363,
    September 1971.

6.  Raynor, G. S.VJ. V. Hayes, and E. C.  Ogden.  Particulate
    Dispersion From Sources  Within a Forest.  In preparation.
                                44

-------
                           SECTION VIII

      DEPOSITION TO VEGETATED SURFACES: GRASSLAND VS.  FOREST


Deposition measurements were made in 395 tests in connection with
dispersion studies, 240 in the open grassy field and 155 in the pine
forest.  From 57 to 90 sampling locations were utilized in each test.1

Although airborne pollen grains are often transported long distances
and concentrations may be heavy downwind of stands of anemophilous plants,
most pollen grains deposit close to their source.  Our studies of
ragweed pollen released from area sources of various sizes show that
±50% deposited in the first 60 m of travel.  Extrapolation of these data
suggests that +99% deposit within a km.  However, the remaining 1%
can account for the high pollen concentrations during the ragweed
season when the number and wide distribution of sources and the amount
of pollen per plant are considered.

Corn pollen, at the other extreme, deposits closer to its source.
Using 18-m diam plots, we found that more corn pollen reached the
ground within the plot than outside while additional  amounts deposited
on the leaves.  Of that portion which did become airborne downwind of
the plots ±95% deposited within 60 m.

Deposition velocity over grass (mowed field) was determined from area,
point, and line sources.  Individual measurements varied widely between
locations and between tests.  However, they tended to average about
three times the terminal velocity of the pollen grains.

Deposition velocity in the forest was studied using point source releases
outside and inside the forest and an area source adjacent to the forest
edge.  Deposition velocity averaged close to the calculated terminal
                               45

-------
velocity.  This is probably due to the light wind speeds and low level
of turbulence near the forest floor which minimize turbulent impingement.
Mass flux calculations showed that most particles were lost to the foliage
rather than to the ground.

Assuming that the measurements are reasonably representative of the
natural surfaces, the data show that velocity of deposition to short grass,
during unstable conditions with light to moderate winds, appreciably
exceeds the terminal velocity of the particles.   Turbulent impingement
is added to gravitational settling.  In the forest with the same conditions
outside, a slight inversion or an isothermal  layer is present under the
canopy, wind speeds are light and turbulence levels low.2  Here,
gravitational settling seems to be the predominant deposition mechanism.
REFERENCES
1.  Raynor, G. S.  Experimental Studies of Pollen Deposition to Vegetated
    Surfaces.  Brookhaven National  Laboratory.   (Presented at Proc.  of
    Atmosphere-Surface Exchange of Particulate  and Gaseous Pollutants.
    1974 Symposium, Richland, Washington.  September 4-6,  1974.)
    Brookhaven National Laboratory, Upton, N.Y.  Publication Number
    BNL 19219.  September 1974.  32 p.

2.  Raynor, G. S.  Wind and Temperature Structure in a Coniferous
    Forest and a Contiguous Field.   Forest Sci  17(3):351-363,
    September 1971.
                                46

-------
                             SECTION IX

                    DIURNAL PATTERNS OF EMISSION

Hourly measurements  of pollen emission were made from a total of
15  cultivated  plots  of ragweed  (Ambrosia], timothy (Phleum], corn (Zea),
and castor  bean  (Eioinus] during several pollination seasons.1

A Hirst spore  trap was mounted  at the center of each circular plot for
this  purpose.  A characteristic diurnal emission pattern was found for
each  genus, but the  emission patterns for an individual  day sometimes
differed appreciably from the seasonal mean.

Ambrosia pollen emission normally begins an hour or two after sunrise,
peaks a few hours later, decreases in the afternoon, and is minor
during the  night.  Emission most typically begins when stable, moist
nighttime conditions change to  unstable and drier air shortly after
sunrise.

Phlewn emission starts during the night, peaks about two hours after
sunrise, and slowly  declines during the day.  Pollen emission may begin
anytime from 2100 to 0600 EST.  If emission started during the night,
it  usually ended by  0800 but, if the beginning was delayed until after
sunrise, appreciable amounts continued until noon.
Zea emits pollen fairly uniformly during the period from two hours
after sunrise  to about sunset, with only small amounts at night.  The
peak varied from midmorning to midafternoon.

Riainus pollen emission began two to three hours after sunrise,  with a
peak at 0900 to 1100, and decreased gradually through the afternoon.
REFERENCE
1.  Ogden, E. C., J. V. Hayes, and G. S.  Raynor.   Diurnal  Patterns of
    Pollen Emission in Ambrosia.,  Phleun,  Zea,  and Riainus.
    Amer J Bot 56:16-21, January 1969.
                                47

-------
                            SECTION X

           TEMPORAL VARIATION IN POLLEN CONCENTRATIONS
Diurnal patterns of occurrence close to a source are influenced
primarily by the patterns of emission but distant from sources
by meteorological variables.  In several different phases of the
study on the travels of airborne pollen, the patterns of occurrence
were routinely determined, primarily with sequential  rotoslides and
Hirst spore traps.  The principal  locations were Brookhaven, Albany,
Saratoga Springs, Raquette Lake, Blue Mountain Lake,  and Whiteface
Mountain near Wilmington.  For most of these studies, the sequential
data are described in other sections.

However, four sequential swing-shield rotoslide samplers (Fig. 4-3)
were operated to study the variation with time of natural pollen
concentrations, not directly influenced by local sources, to determine
peak to mean concentration ratios over various sampling intervals and
to assess the representativeness of single station measurements.1

Sampling was conducted in Albany,  Saratoga Springs, and Brookhaven
to examine patterns in both urban and rural areas.  Sequential rotoslide
samplers with time periods of 5 minutes plus one hour and 2 hours
plus 24 hours were used in each location.  Saratoga Springs was used
in 1971, Albany in 1972, and Brookhaven in both years.  Sampling was
conducted during tree pollen and ragweed pollen seasons.  Two hundred
ninety-four tests were made, 72 in Saratoga, 95 in Albany, and 127 at
Brookhaven.  Since tree pollens were sometimes separated by genera,
567 separate cases were available for analyses.  These were composed
of 415 tree and 152 ragweed pollen cases.  Two hundred seventy-nine
were taken in 1971 and 288 in 1972.  Three hundred three were five minute
and one hour tests and 264 were 2 and 24 hour sampling periods.
                                48

-------
Analyses of short-period samples show that agreement in catch is  good
between consecutive 5-minute periods, but slowly becomes poorer over
the course of the hour.  Little consistent difference was evident
between species and locations.

Agreement between consecutive 2-hour samples is slightly poorer
than between 5-minute samples an hour apart.  Agreement decreases
over the next 12 hours and usually becomes better again toward the
end of the 24-hour period.  This is attributed to the diurnal cycle.
Again, similar patterns were found for all species and for all  locations
except Albany where an improvement toward the end of the period was
not evident.

Coefficients of variation averaged about one for 2-hour periods and
0.3 to 0.5 for 5-minute samples.  Values of maximum/mean usually  ranged
from 3 to 5 for 2-hour samples and from 1-2 for 5-minute periods.
Values of minimum/mean were usually less than 0.3 for 2-hour  samples
and about 0.5 for 5-minute samples.
REFERENCE
1.  Raynor, G. S., E. C.  Ogden,  and J.  V.  Hayes.   Temporal  Variation
    in Airborne Pollen Concentrations.   In preparation.
                                49

-------
                            SECTION XI

           SPATIAL VARIABILITY IN POLLEN CONCENTRATIONS
Tests were conducted to determine the relationship between airborne
pollen concentrations and distance.1  Simultaneous samples were taken
in 171 tests, using sets of eight rotoslide samplers, 1.5 m above the
ground and 1 to 486 m apart in straight lines.  Use of all possible
pairs gave 28 separation distances.  These tests were conducted over
a two-year period in urban (Saratoga Springs and Albany)  and rural
(Brookhaven) locations during both tree and ragweed pollen seasons.
Each run was of short duration to minimize the smoothing  effect of
longer exposures, most lasted 10 minutes but 5-min and 20-min periods
were also used.  During the second summer, to improve accuracy,
mechanical counters were added for determining the actual number of
revolutions each sampler made during each run, and four slides were
rotated instead of the usual  two (Fig.  11-1).  Forty-two  tests were
made during the two tree pollen seasons and 129 during the ragweed
seasons.

The tests were grouped by pollen type,  location, and wind direction
relative to the line.  The data were analyzed to evaluate variability
without regard to sampler spacing and variability as a function of
separation distance.  The mean, standard deviation, coefficient of
variation, ratio of maximum to the mean, and ratio of minimum to the mean
were calculated for each test, each group of tests, and all  cases.

The average coefficient of variation is 0.21, the maximum over the
mean 1.39, and the minimum over the mean 0.69.  No relationship was
found with experimental conditions.  Samples taken at the minimum
separation distance had a mean difference of 18%.  Differences between
pairs of samples increased with distance in 10 of 13 groups.
                                50

-------
Fig. 11-1
Rotoslide sampler with revolution  counter.
                               51

-------
This increase in variability with distance appears to be caused by
actual differences in pollen concentrations.   They may be due to
inequalities in emission which atmospheric mixing has not yet
smoothed out or they may be caused by the action of local obstacles
in modifying the pollen content of previously well-mixed air.  The
data make the first possibility more plausible since the variability
at Albany with many large, local obstacles was not significantly
different from that at Brookhaven, where the  terrain is relatively
smooth and major obstacles absent.

These results suggest that airborne pollens are not always well mixed
in the lower atmosphere and that a sample becomes less representative
of concentrations at increasing distances from the sampling location.

REFERENCE
1.  Raynor, G. S., E. C. Ogden, and J.  V. Hayes.  Spatial Variability
    in Airborne Pollen Concentrations.   J. Allergy Clin Iminunol.
    In press.
                                52

-------
                            SECTION XII

          VARIATION IN POLLEN CONCENTRATION WITH HEIGHT
Variation in concentration with height from near and distant pollen
sources was studied on the 128-m tower in an open area at Brookhaven
on Long Island, on the 38-m tower in an extensive forest at Raquette
Lake in the Adirondack Mountains, on the 16-m mobile tower at various
locations, and by the use of aircraft up to 1000 m over the Long Island
area.  Most of the data are only a part of other studies and are discussed
elsewhere in this report.

Ragweed pollen concentrations were measured at five levels from
1.5 to 108 m on the meteorology tower at Brookhaven.1  Samples were
taken over an 11-year period using both rotoslide and slide-edge-cylinder
samplers.  Vertical  pollen profiles showed great variability from
day to day but, when averaged over pollen seasons or longer periods,
little systematic change with height was found.   These results suggest
that long-term patient exposure at upper levels  of tall  buildings
would be similar to that at ground level in the  absence of nearby
pollen sources.  This would not apply close to a local  source where
low-level  concentrations would be much greater than those at higher
elevations and may not apply in cities or other  dissimilar locations
where vertical  mixing and removal  patterns may differ from those
at Brookhaven.

REFERENCE
1.  Raynor,  G.  S., E.  C.  Ogden, and J. V.  Hayes.   Variation  in Ragweed
    Pollen Concentration to a Height of 108 Meters.   J  Allergy Clin
    Immune1  51:199-207,  April  1973.
                                53

-------
                           SECTION XIII

ENHANCEMENT OF POLLEN CONCENTRATIONS DOWNWIND OF VEGETATIVE BARRIERS


Studies on dispersion into a forest gave information on the removal  rate
of a large vegetated area which extended far beyond the sampling array
in the downwind direction.  Information on the effects of narrower
strips of vegetation was sought by sue of sampling arrays on the two
sides of a hedge.1  The area selected for study in 1971 was at a tree
nursery near Saratoga Spring (Fig. 13-1), where large fields are
bordered and separated by dense hedges of arbor-vitae (Thuja ooddentalis]
The hedge selected was about 100 m lonq, 6 m tall, 3.7 m thick, and
oriented in a north-south direction.  The upwind and downwind distances
to the next obstacles were the same, about 100 m.

Two parallel rows of masts supporting swing-shield rotoslide samplers
were installed 5 m apart at right angles to the hedge.  Four pairs of
masts were located on each side of the hedge.  Samplers were mounted
at, heights of 1.5, 3.0, and 4.5 m on each mast, giving a total of
48 samplers.  A totalizing anemometer, 50 m upwind, recorded the air
travel during each run.  As these runs were taken only when the mean
wind direction was approximately at right angles to the hedge, only
four were obtained during the spring tree pollen season and five during
the ragweed season,

A hedge of broad-leaved shrubs and trees was chosen for 1972.  The
sampling design was the same.  Seven runs were taken during the tree
and eight during the ragweed pollen season.

To visually determine the effects of these hedges on air flow, oil
fog smoke was released upwind of each hedge.  These observations
indicated that much of the air mass goes up and over the hedges
rather than through, particularly at the arbor-vitae hedge.
                                54

-------
                                                             o
                                                             QJ
                                                             c
                                                             o
                                                             c
                                                             QJ
                                                             O
                                                             C
                                                             o
                                                             o
                                                                  0)
                                                                  QJ
                                                                  s_
                                                                  ro
                                                                  i.
                                                                  QJ
                                                                 •r—
                                                                  i-
                                                                  s_
                                                                  re
                                                                 -Q
 o
 CL

 C.
 O

 QJ
 C7!
 -o
 QJ
                                                                  re
                                                            re   -i->
                                                                  QJ
                                                            M-    CD
                                                            O    QJ
                                                                  O)
                                                            o
                                                            QJ
                                                            II     I ^
                                                            M-
                                                            QJ   M-
                                                                  O
                                                            QJ


                                                                  re
                                                            CD   CL
                                                            C
                                                            •r-    re
                                                            i-
                                                            o
                                                                 o
                                                                 sr
                                                            re
                                                            s-
                                                            s-
                                                                 to
                                                                 S-
                                                                 0)
            c:
            o

           -P

            s-

            c
            QJ
            O
            C
            o
            o

            c
            QJ
                                                                       O
                                                                       Q.
            C
            O
                                                                 CL  T-
                                                            CD   «/l
i—   -O
 CL  T-

 re    i/>
c^    o

      o
                                                           ro
                                                                 c
                                                                 QJ

                                                                 cr
                                                                 QJ
                                                                 c/i
            re
           •r-
            i-
            re
                                                                       re
                                                                       o
                                                                       CL
                                                                       QJ
                                                                       O
55

-------
Examination of the data showed that the low-level  concentration  pattern
was markedly altered by both hedges.  A pronounced concentration
maximum was found in the cavity region (also called wake eddy or
standing zone eddy) downwind of each hedge.   Apparently airborne
particles as large as pollen, because of their significant settling
rates and appreciable inertia in a moving airstream, accumulate  in
the cavity downwind of vegetative barriers.   Similar behavior downwind
of other obstacles may be inferred.  The pollen grains appear to move
more readily into the cavity than out of it and accumulate there.
The process must result from inertial effects so gases and small
particles would not behave in similar fashion.

Several  applications are apparent: (1) other large airborne particles
such as pesticides and plant disease spores should be affected in
similar fashion; (2) maximum deposition should occur in the same
region as maximum concentration and deposition sampling could indicate
the location of the cavity; (3) multiple impaction surfaces, such as
scattered trees and bushes in the cavity region of a large barrier,
probably would be more efficient removers of particles than the
barrier itself; and (4) allergic individuals should avoid wake regions
when aeroallergens are prevalent.

REFERENCE
1,  Raynor, G. S., E. C. Ogden, and J. V. Hayes.  Enhancement of
    Particulate Concentrations Downwind of Vegetative Barriers.
    Submitted to: Agric Meteor.
                                56

-------
                            SECTION XIV

      MESOSCALE TRANSPORT AND DISPERSION OF AIRBORNE POLLENS


Pollen transport and dispersion from generalized area sources were
studied by 29 flights to distances of 100 km and heights to 3 km
using an aircraft-mounted isokinetic sampler.1  Tree pollens and ragweed
pollen were sampled.  Four types of flights were made to study various
aspects of pollen transport: (1) ascents over a fixed location to
investigate vertical distribution, (2) flights over a source-free
area to document change of concentration with distance, (3) east-west
flights along Long  Island to study the influx of pollen from the
mainland with westerly winds and, (4) vertical ascents and horizontal
flights during sea breeze flows to determine their effect on pollen
concentrations.

It was found that large quantities of pollen are transported in
orderly fashion  from their source regions,  but pollen often travels
in large, discrete clouds.   Pollen is transported to Long Island
from the mainland in some quantity.   Sea breeze flows greatly decrease
low-level concentrations, but pollen is carried aloft at the sea breeze
front and recirculated in the return flow aloft.  Vertical  distribution
is reasonably well  related to lapse  rate, although secondary
concentration peaks, which often occur below elevated inversions,
cannot be explained by the data obtained.

REFERENCE
1.  Raynor,  G. S.,  J. V.  Hayes, and  E. C. Ogden.  Mesoscale Transport
    and Dispersion of Airborne Pollens.  J  Appl  Meteor 13(l):87-95,
    February 1974.
                                57

-------
                            SECTION XV

  EXPERIMENTAL PREDICTION OF DAILY RAGNEED POLLEN CONCENTRATIONS
A method was developed by which daily ragweed pollen concentrations
were predicted on an experimental basis over a four-year period.1
Predictions were made by referring to the mean annual  concentration
curve of earlier years, previous concentrations of the current year,
and the weather forecast.  The predicted weather was evaluated primarily
for: (1) its suitability for ragweed pollen emission,  locally and  in
potential source regions; (2) whether expected air trajectories would
pass from source regions over the forecast site; and (3) whether
precipitation, which might remove previously emitted pollen from the
air, would occur at the site or between there and the  source region.
More accurate results were obtained than by use of alternative methods
such as using the average for the date in past years (climatology)
or using the previous day's concentration (persistence).

Forecast^ were verified by comparisons with concentrations  measured
with an intermittent rotoslide sampler at a height of  five  feet
in an open area.  Verification criteria were selected  arbitrarily
as a factor of 2 or 10 grains/m3.  Thus, a forecast of 100  grains/m3
would be considered correct if the measured concentration fell
between 50 and 200.  The standard of 10 grains/m3 was  included to
take care of low concentrations where the forecast could be very
close but not within a factor of 2, such as forecast of one and an
actual  concentration of three.
                                58

-------
Results for the four years are given in the following table for all
actual pollen forecasts, and for all forecasts for days on which the
weather forecast was correct.  Also, given are the results which
would have been obtained if climatology and persistence had been used.
The averages in the last column are weighted by the number of cases
in each year.

               PERCENT OF FORECASTS WITHIN A FACTOR
                       OF 2 OR 10 GRAINS/m3
Climatology
Persistence
Forecast
Forecast with correct
  weather forecast
Number of cases
1966
72
49
71
80
27
1967
55
50
71
82
17
1968
58
62
58
62
36
1969
68
59
70
76
30
Average
62
56
67
75

REFERENCE
1.
Raynor, G.  S.,  and J.  V.  Hayes.
Ragweed [Pollen]  Concentration.
December 1970.
Experimental Prediction of Daily
Ann Allergy 28:580-585,
                                59

-------
                            SECTION XVI

        OCCURRENCE OF RAGWEED POLLEN IN A SOURCE-FREE AREA


From mid-August to late September, ragweed (Ambrosia) pollen originates
in a multitude of sources throughout eastern and central  United States  and
southern Canada.  Although most of the pollen which becomes airborne
returns to earth near its source, the small  percentage from many sources
which reaches altitudes above surface obstacles and travels long distances
causes significant concentrations over vast areas.   Pollen concentrations
at some distance downwind from these sources during dry weather can be
expected to be appreciable.  Rain may wash most of the pollen from the
atmosphere.  A frontal passage may replace a pollen-laden air mass with
one relatively pollen-free or vice-versa.

In the heart of the Adirondack Mountains of northern New York, local
sources of ragweed pollen are negligible.  It may be assumed that the
po1!en present would have come from distant sources and,  for the most
part, from areas several  hundred kilometers to the southwest, this
being the direction of greatest ragweed production and of prevailing
winds at this time of year.  Most of the Adirondack area  is covered
with forest.  Ragweeds, especially A.  ariemisii folia, can grow here
in waste lots and along roadsides.  However, this is a recognized
hayfever resort area where the stray ragweed plants that appear are
removed by an alert citizenry before the pollen is shed.   Numerous
persons, afflicted by ragweed pollinosis, have attested to the relief
they experience in this region.  Ragweed pollen counts, reported for
several localities in this region over many years1'2 would seem to
indicate low concentrations.  Several  questions arise.  Were the
sampling techniques sufficiently trustworthy to indicate the true
concentrations?  What are the true concentrations here?  Do the forests
remove large quantities of pollen?  Are the concentrations at ground
                               60

-------
level and above the forest materially different?  How do meteorological
variables such as wind direction, wind speed,  turbulence, humidity,
and precipitation affect the pollen concentrations?  Do the diurnal
patterns of concentration show correlation with time of day or any
meteorological parameters?  Is this pollen, after transport one or more
days in varying humidities and exposed to cosmic rays and solar radiation,
as allergenic as fresh pollen?

To answer some of these questions, pollen sampling assemblages were
operated at three locations in the Adirondack  Park during 1963, 1966,
and 1967.  Valuable data were obtained,  but there were frequent periods
of rain during the ragweed season each year.   In 1968, sampling was
continued at one of the locations with the hope of obtaining data
during longer periods of dry southwest winds.

SAMPLING SITES
Samplers were installed on a 100-km line beginning near Raquette Lake
and running northeast to Whiteface Mt. near Wilmington.   The most
southwesterly location was at Sagamore Pond, six km south of Raquette
Lake Village.  To the southwest are many km of unbroken forest without
any nearby land elevations to greatly modify the air flow,   The next
location was at Blue Mt. Lake, 22 km northeast, with forest intervening.
The third location was on Whiteface Mt., 78 km further along.

INSTRUMENTATION
Samplers employed were intermittent rotoslides, sequential  rotoslides,
and Hirst spore traps.  The intermittent rotoslide sampler3 took pollen
samples representative of the daily average concentration by sampling
for one minute in each 12 during a 24-hr period.  The efficiency of
catch varies with wind speed from 50-70% but with an average of
approximately 64% at the wind speeds mostly encountered.   This efficiency
                               61

-------
correction was used for all  rotoslide samples.   The  corrected  count  from
each sample represented the number of pollen  grains  in  4.5 m3  of air;
the data were then reduced to grains/m3.

The sequential rotoslide sampler4 automatically took 12 consecutive
samples, each continuous over a chosen time period.   For most  of the
sampling, a 2-hr time period was chosen.   Two units  were operated  in
tandem on occasion, for obtaining 24 one-hour samples  during the
24-hr day.

The Hirst spore trap5 is an automatic sequential  volumetric air sampler.
A vacuum pump draws air through a narrow  orifice which  is oriented into
the wind by a vane tail.  The pollen is impacted on  a microscope slide,
coated with a suitable adhesive, which moves  past the  orifice  at 2 mm/hr.
The 24-hr sample is spread in a 48-mm-long band from which counts  are
made that represent each hour.  The efficiency varies  with wind speed,
from about 60-120% for ragweed pollen, with a probable  average efficiency
around 70% at the wind speeds of that time of year.

Sagamore
At the Sagamore Pond location, a tower was installed in the forest at
the Sagamore Conference Center of Syracuse University  (Fig. 16-1B).
This tower is of open construction steelwork, allowing  the air to  pass
through freely.  There are four working levels: 1.5, 12, 26, and 38
meters.  Treetop level here is approximately  20 m.   Two of the levels
are below the forest canopy.  The 26-m level  is above  the trees but
where air turbulence may be affected by the forest.   The 38-m  level
is at a height where the flow of air is less  affected by the vegetation.
Samplers at the three upper levels were raised and lowered on  tracks
by means of electrically powered winches.

Sampling was conducted at six positions,  of which three were near  the
ground (1.5 m): in a large clearing at the Conference Center which is
                               62

-------
Fig.  16-1      Mobile  tower at Blue Mountain Lake (A) and tower at
              Sagamore  Pond  near Raquette Lake (B).
                               63

-------
0.5 km southwest of the tower (C), in a small  glade near the  tower (G),
and in the dense forest at the tower (SI).   The other three positions
were on the tower: at 12 m (S2), 26 m (S3), and 38 n (S4).  An intermittent
rotoslide obtained daily average pollen concentrations at each of the
six positions.  In 1963, sequential rotoslides determined average
concentrations for each 2-hr period at positions SI and S4.  For 1966
and 1967, the sequential rotoslides were replaced by Hirst spore traps
which allowed determination of concentrations  for each 1-hr period.

Blue Mt. Lake
In 1963, an intermittent and a sequential  rotoslide were operated at
the 1.5-m level  in a large clearing at the  Minnowbrook Conference
Center of Syracuse University.  In 1966, a  mobile tower sampling
assemblage (Fig. 16-1A) was located on a bluff, approximately 15 m
above the lake level, at the Adirondack Museum.  Sampling positions
were at 1.5 m (Bl), 6 m (B2), 11 m (B3), and 16 m (B4).  An intermittent
rotoslide was at each level.  At the 1.5-m  level, a sequential
rotoslide took consecutive 2-hr samples.  A sensitive anemometer
recording system,  using counters that print out each hour,  recorded
airflow at the Bl  and B4 levels.  In 1967,  the assemblage was identical
to that in 1966 but with the addition of a  Hirst trap.  In 1968, the
tower assemblage was located near lake level in an open area, approximately
50 m from the surrounding forest.  These three locations are  essentially
one, as they are near together at the east  end of the lake.  The 1968
assemblage included all samplers used in 1967, except that two sequential
rotoslide machines were operated in tandem  to  take consecutive 1-hr
samples.  In addition, at the Bl level, two intermittent rotoslides,
coupled with a switch timer, each operating on alternate days, obtained
daily samples beqinng at 0000 EST.
                                64

-------
                                                               c

                                                               03
                                                              O)
                                                              u
                                                              to
                                                              M-
                                                              O)
                                                              l/>

                                                              O)
                                                              .c
                                                              +->

                                                              c
                                                              o

                                                              to
                                                              S_
                                                              O)

                                                              o.
                                                              C
                                                              O)
                                                              o
                                                              Q-
                                                              CJ
                                                              •
                                                              o
                                                              4->
                                                              o
                                                              CM
                                                               I
                                                              to
65

-------
Whiteface Mt.
During the 1963 season, an intermittent rotoslide and a  Hirst  trap
were operated on an exposed cliff on the southwest side  near the
summit, approximately 1480 m (4850 ft) above sea level  (S)  and an
intermittent rotoslide at the Atmospheric Sciences Research Center
Building (A) on the northeast side of the mountain.

In 1966 and 1967, intermittent rotoslides were at these  two positions,
In 1966, a sequential rotoslide was installed at the  summit position
but almost continuously developed mechanical troubles so was moved  to
the roof of a building, called the turnhouse, near the  summit  on
the northeast side (T), a slightly less exposed position.   In  1967,
after several modifications, it was again operated at the sunmit
(Fig. 16-2).
RESULTS
The average concentrations during the ragweed season  for the  four
years at the three regions are the daily averages  at  the thirteen
locations.   The concentrations for each hour or two-hour period  at
four of these locations are charted in detail  in several Progress
Reports.6'9

Seasonal Patterns
The daily average concentrations varied greatly, from day to  day,
from mid-August to mid-September.  On most days, the concentrations
were low at all stations.  What minimum number of  grains per  cubic
meter would be judged a high concentration is difficult to choose.
However, a concentration of 25/yd3 has been generally considered by
many allergists to produce marked symptoms of hayfever among  most
persons allergic to ragweed pollen.  Although this figure has never
been adequately tested, it will serve as a reasonable guess.   This
                               66

-------
would be approximately 30/m3.

Assuming an average daily concentration of 30 or more per cubic meter
to be a hayfever day, the following chart indicates the number of such
days at each of the three sampling stations in the open at 1.75 m, as
determined with intermittent rotoslides.
Year
1963
M
II
1966
ii
H
1967
II
1)
1968
Station Time
Sagamore Aug. 19 - Sept. 8
BML
Whiteface
Sagamore Aug. 18 - Sept. 14
BML
Whiteface
Sagamore Aug. 14 - Sept. 14
BML
Whiteface
BML Aug. 15 - Sept. 19
Total
days
21
21
21
28
28
28
32
32
32
36
Hayfever
days
5
0
3
4
4
2
4
6
2
4
Diurnal Patterns
At the Sagamore tower, two Hirst traps obtained the average concentration
for each hour at 1.75 m in the dense forest and at 38 m well  above the
treetops.  These data are graphically shown in Figs.  16-3 and 16-4.
Similar data for Blue Mountain Lake were obtained with sequential  rotoslides
and are shown in Fig. 16-5.

Variation With Vegetation
At the Sagamore tower in the dense forest,  two sampling positions  were
below (1.75 and 12 m) and two were above (26 and 38 m) the tops  of the
                                67

-------
Fiqure  16-3.  AVERAGE  DIURNAL PATTERNS OF RAGWEED POLLEN AT THE
             SAGAMORE TOWER FOR 1966.  Figures at left indicate
             mean percent of daily total.
    1
      _Pattern at  1.75 m in the forest
      _for the same days as above
    t
    1
       Pattern at 38 m above the forest for the 20 days
       when the average concentration was 10/m3
       or more at this level
                       n
      -Pattern  at 38 m for the  10 days when  the average
      ..concentration was 10/m3  or more at the
       1.75-m level
                                 I  I  I  I  I   I  I  I  I   I  I
      -Pattern at 1.75 m for
      _the same  10  days
           Ln_
EST  0   2
I   I  t  I  I  I   I  I  I  I   I  I  I  I   I  I
8   10  12   14    16   18  20   22  24
                           68

-------
Figure 16-4.
    9
    7
    5
    3
    1

   11
    9
    7
    5
    3
    1


   13
   11
    9
    7
    5
    3
    1
   13
    11
       AVERAGE DIURNAL PATTERNS  OF RAGWEED POLLEN AT THE
       SAGAMORE TOWER FOR 1967.  Figures at left indicate
       mean percent of daily total.
 Pattern at 38 m above the forest for the 16 days when
 the average concentration was  10/m3 or more at this
J evel
 Pattern at 1.75 m  in the forest for the
_same days as above
 Pattern at 38  m for the 8 days when the average
 concentration  was 10/m3 or more at the 1.75-m level
-Pattern at 1 .75 m for the same
.8 days
J\.
             4
                 I   I  I  I  I  |   I  I  I  I   I  I  1  I
                 8   10    12   14   1€   18   20  22   24
                           69

-------
Figure 16-5.
AVERAGE DIURNAL PATTERNS OF  RAGWEED  POLLEN AT THE
1.75-m LEVEL IN AN OPEN AREA AT  BLUE MOUNTAIN LAKE
FOR THOSE DAYS WHEN THE AVERAGE  DAILY CONCENTRATION
WAS 10/m3 OR MORE.  Figures  at left  indicate mean
percent of daily total .
        _1966  (13 days)
      5 -
      3

      1
         -1967  (12 days)
                               1   I  I  I   I  I  I   I  I  I   I  I   I  I
        -1968  (20 days)
        -1966-68  (45 days)
  EST   0
            8   10    12   14   16   18   20   22   24
                               70

-------
trees (20 m).  The effect of this forest on the pollen concentrations
is difficult to assess because the daily comparisons were so variable.
Excluding those days when the concentrations at neither of the levels
being compared was at least 10/m3, the following data apply:

                                          Range in %
Year        No. of days           at 1.75 m of that at 38 m         Av.
1963            6                       11  to 325                     45
1966           13                       12  to 169                     25
1967           17                        0  to 800                     48
                                          Range in %
Year        No. of days           at 12m of that at 38 m           Av.
1963            8                       46 to 425                     91
1966           12                       35 to 136                     95
1967           15                        0 to 2850                   106
                                          Range in %
Year        No. of days           at 26 m of that at 38 m            Av.
1963            8                       53 to 600                     93
1966            9                       59 to 140                     90
1967           15                        9 to 2300                    78

Variation in daily average concentrations for three positions  at  the
Sagamore location are graphically shown in Figures 16-6  and 16-7.   It
is seen that while the concentrations are usually highest above the
forest, lowest in the forest at ground level, and intermediate in  the
open area at ground level, this is not always so, especially when
concentrations are low at all three positions.   Figure  16-8 illustrates
the comparative concentrations for those days during the two years
when the pollen grains per cubic meter were at least 30  (presumably
a hayfever day) at one or more of the positions.
                               71

-------
  i      i      i      i      i      \      r
                                                   i      i      r
UD
D;
•r—
Ll_
                                                                  LT>
  J_...  I  _.  I'   I      I      I      I     I.__!__!	I
                                                                  o
                                                                  ro
                                                                  oo
                                                                  CM
                                                                  UD
                                                                  CM
                                                                     OL
                                                                     UJ
                                                                     CO
                                                                     Q.
                                                                     UJ
                                                                     OO
                                                                  CM
                                                                  CVJ
                                                                   O
                                                                   CsJ
                                                                   00
                                                                      to

                                                                      CD
  O
  O
        O
        CT>
O
CO
                          o    o    o
o
CO
o
CM
                                72

-------
      I      I      I     I     I
I      I     I     I     I     I
 I
UD
cn
                                                                                                     en
                                                                                                     co
                                                                                                     LO
                                                                                                        CO
                                                                                                     CM
                                                                                                        Q-
                                                                                                        LU
                                                                                                        00
                                                                                                     00
                                                                                                     O
                                                                                                     oo
                                                                                                    en
                                                                                                    c\J
                                                                                                    CO
                                                                                                    CVJ
                                                                                                    CM


                                                                                                    UD
                                                                                                    LO
                                                                                                    CM
                                                                                                    CNJ
                                                                                                    oo
                                                                                                    c\j
                                                                                                    CM
                                                                                                    CM
                                                                                                    CM
                                                                                                    O
                                                                                                    CM
                                                                                                    cn
                                                                                                    CO
                                         oo

                                         CD
      I     I     I     1    I     I     I      I      I     I     I     I     I     I      I

     Z>    VQ    CM    OO^l-    O   tD   CM    CO     «d"O(J3CM    CO^J-
                                                73

-------
          Figure 16-8
180

170
160

150

140

130

120

110

100

 90
 80

 70

 60
 50

 40

 30

 20
 10
  0
O
DAILY AVERAGE CONCENTRATIONS PER irf OF RAGWEED
POLLEN AT THREE POSITIONS AT SAGAMORE FOR THOSE
DAYS WHEN THE CONCENTRATION WAS AT LEAST 30/m3
AT ONE OR MORE OF THE  POSITIONS
                         Q  ABOVE THE FOREST AT 38 m
                             LARGE OPEN AREA AT 1.75 m
                             DENSE FOREST AT 1.75 m
       23   29
       AUG.
      1     5
      SEPT.
         10   12

             1966
 20   21   27
 AUG.
1967
28   29   30
8
SEPT.
                                  74

-------
Comparison of Sites
The 1966 and 1967 data allow some comparison of agreement at the
sampling positions at the Sagamore forest and between the Sagamore and
Blue Mountain Lake locations.  For determining the probability of agreement
between pairs of positions, correlation coefficients (r) were obtained,
using the method of least squares.  For determining the relationship
between two positions, the constants m and b for the line of best fit
(y = m x + b) were calculated by the method of least squares.  Using
the above formula, it is possible to use each experimental  value of
x to calculate the expected corresponding value of y.  It was
arbitrarily decided that the correlation between individual  sample
pairs was sufficiently close if the experimental value of y  was not less
than half nor more than twice the calculated value, or the two differed
by no more than 10 pollen grains per cubic meter,   p is the  fraction
of samples which met these criteria.

Positions          Year          r          p          m          b
SI & S4
SI & S4
 G X S4
 G & S4
 G & SI
S4 & B4
S4 & B4
SI & B4
S4 & W
 W & A
1966
1967
1966
1967
1967
1966
1967
1966
1967
1967
0.85
0.63
0.99
0.51
0.88
0.96
0.39
0.86
0.36
0.61
0.86
0.64
1.00
0.60
1.00
0.92
0.60
0.90
0.64
0.57
4.06
1.41
2.02
0.72
0.56
0.79
0.45
3.42
0.32
0.75
-1.51
4.42
0.66
7.96
2.02
4.41
18.44
0.92
4.38
13.20
                               75

-------
SI = Sagamore tower in the forest at 1.75  m
S4 = Sagamore tower above the forest at 38 m
 6 = Small open area in the Sagamore forest near the  tower
B4 = Blue Mountain Lake tower at 16 m
 W = Whiteface Mountain at the surrmit
 A = Atmospheric Sciences Research Center  on the east slope  of
        Whiteface Mountain

The above chart shows the relationships of simultaneous  (hourly)  pollen
concentrations at different locations.   We assumed that  if any
relationship existed, it would be linear,   r (correlation coefficient)
is a measure of the extent to which two variables are related,  p is
the fraction of sample pairs in which the  experimental and calculated
concentrations agreed within a factor of two or 10/m3.   m is the  actual
ratio of the concentrations at one chosen  position with  the  concentrations
at another position.  For example: in 1966, the concentrations  at the
38-m level above the forest at Sagamore averaged 4.06 times  the
concentrations at the 1.75-m level in the  forest directly below;
in 1967 the comparable ratio was 1.41.   b  is the intercept,  or  the
value of y when x is zero.  For example: in 1966, with a concentration
of 100/m3 at the 38-m-level at the Sagamore tower, the expected
concentrations at the 16-m-level at Blue Mountain Lake would be
0.79 x 100 + 4.41 = 83.41/m3.

To use examples which actually occurred: with a concentration of  175
at 38 m at Sagamore, the expected concentration would be 142 at BML;
it was 141.  With a concentration of 118 at Sagamore; the expected
concentration at BML would be 97; it was 88.  However, with  a
concentration of 77 at Sagamore, the expected concentration  at  BML
would be 65; it was 95.  Thus, it is evident that a formula  based on
a seasonal average, may not be reliably used for estimating  a day's
                                76

-------
average concentration at one of these locations based on the
concentration at the other.

Correlation With Weather
Official weather data for wind direction, wind speed, and precipitation
during the ragweed seasons of 1966 and 1967 were obtained and charted
for eight locations surrounding the Adirondack area.   These indicate
much variation among these locations at identical  times.  As expected,
dry southwest winds were correlated with high pollen  concentrations
but little more can be inferred.

Conclusions
The average daily ragweed pollen concentrations at all  13 positions
at the three locations varied from zero to more than 50/m3 and in some
situations to more than 100.  On some days the concentrations reached
two or three hundred for short periods.  Usually, but not always, the
concentrations above the trees or at ground level in large open areas
were higher than in the forest.  The frequent rains usually lowered
the concentrations to zero but, during the dry periods  that followed,
the concentrations increased rapidly, especially if the winds were from
the southwest sector.  It is possible that the lower concentrations in
the forest are due, at least in part, to less air speed and less turbulence,
allowing the grains to settle to earth more rapidly.

Thus, it is apparent that there may be, at times, rather high
concentrations of ragweed pollen distant from the sources of this pollen,
and while the extensive forest may have but little influence on removal
of pollem from the air mass passing high above the forest, it seems obvious
that there is significant removal  by the forest below treetop level.
                               77

-------
REFERENCES
1.  Durham, 0. C.   Pollen Prevalence  and  Pollen-free Areas.
    J Amer Med Ass 148:716-720,  March 1952.

2.  Ogden, E.  C.,  and D.  M.  Lewis.  Airborne  Pollen and Fungus
    Spores of New York State.  New  York State Museum and Science
    Service Bulletin 378, Albany, N.Y.  January  1960.  104 p.

3.  Ogden, E.  C.,  and G.  S.  Raynor.   A New  Sampler for Airborne Pollen:
    The Rotoslide.  J Allergy  40(1):!-!!, July 1967.

4.  Ogden, E.  C.,  G. S.  Raynor,  and J. M. Vormevik.  Travels of
    Airborne Pollen. Progress  Report  4.   New  York State Museum and
    Science Service, Albany, N.Y.   June 1963.  12 p.

5.  Hirst, J.  M.   An Automatic Volumetric Spore  Trap.  Ann Appl Biol
    39:257-265, June 1952.

6.  Ogden, E.  C.,  G. S.  Raynor,  and J. M. Vormevik.  Travels of
    Airborne Pollen. Progress  Report  5.   New  York State Museum and
    Science Service, Albany, N.Y.   May 1964.  30 p.

7.  Ogden, E.  C.,  G. S.  Raynor,  and J. V. Hayes.  Travels of Airborne
    Pollen. Progress Report  7.   New York  State Museum and Science
    Service, Albany, N.Y. March 1967.  25  p.

8.  Ogden, E.  C.,  G. S.  Raynor,  and J. V. Hayes.  Travels of Airborne
    Pollen. Progress Report  8.   New York  State Museum and Science
    Service, Albany, N.Y.  February 1968.   36 p.

9.  Ogden, E,  C.,  G. S.  Raynor,  and J. V. Hayes.  Travels of Airborne
    Pollen. Progress Report  9.   New York  State Museum and Science
    Service, Albany, N.Y. March 1969.  27  p.
                               78

-------
                               SECTION  XVII

CONCENTRATIONS OF RAGWEED POLLEN  IN  RELATION  TO  REDUCTION  OF  SOURCE  PLANTS
    The  mobile tower assemblage  was  operated  in  Saratoga  Springs  in  cooperation
    with a planned program of citywide ragweed eradication.  This  small city
    (population 20,500)  has the  usual  abundance  of  ragweed  plants  in  its
    unattended areas and in the  surrounding countryside.  Sampling was
    conducted before and after the eradication activities.   However,  the
    eradication,  conducted by another  agency, was less  than  complete  and
    the  average pollen  concentration was  higher  during  1971  following the
    attempts  at eradication than during 1970  (Fig.  17-1).  The  diurnal
    patterns  of occurrence for the two years were essentially the  same
    (Fig.  17-2).   The data, detailed in Progress Reports,1'2 indicate
    that the  hourly concentrations of  pollen  varied greatly, usually more
    than half of the pollen occurred between 0700 and 1300 EST, and  the
    highest concentrations were  between 0800 and 1000.  There was a  tendency
    for  the daily peak  concentrations  in  1971 to be an  hour  later  than
    in 1970.   Perhaps this is due to the  ragweed eradication activities
    causing the diurnal  patterns of  occurrence in the city to lag behind
    the  diurnal  patterns of emission in the areas outside the three-mile
    radius of ragweed reduction. If so,  then it is obvious  that these
    more distant  ragweed pollen  sources supply pollen to  the city air
    in far greater amounts than  what is produced locally.  The  higher
    average concentrations for 1971, in comparison  with 1970, apparently
    are  due to higher regional concentrations.   Whether the  1971
    concentrations in the center of  the city were lower than would have
    occurred  with no eradication is  not known, but  it seems  likely that
    any  difference would be negligible.
                                   79

-------
 CONCLUSIONS
 Ragweed eradication over a  small  area  surrounded by a large area
 with many ragweed plants is unlikely to result in a significant reduction
 in ragweed pollen average concentrations.   We  offer no opinion with
 regard to any possible effect on  incidence  of  pollinosis.   However, it
 is probable that eradication of abundant ragweed in an area that is
 surrounded by an extensive  ragweed-free area would have a  marked effect
 on local ragweed pollen concentration  reduction.   See Section XVI.
                                         Saratoga Springs Ragweed Pollen
                                                  	1970 (average 40/M3
                                                     1971 (average 62/M3)
 17
                                                9    H    13
                                                SEPTEMBER
Fig. 17-1     Daily Average Concentrations  of  Ragweed Pollen
              at Saratoga Springs.
                                 80

-------
                                        Saratoga Springs Ragweed Pollen
                                            Average Diurnal Patterns
                                             AUGUST 17 thru SEPTEMBER 21
                                            Sequential rotoslide   1970	
                                                            1971
                                            Hirst spore trap   1970	
                                                          1971 	
1234   5  6  7  8   9   10 ' U ' 12  13 '  14 ' 15 ' 16  17   18 ' 19 ' 20 21  22  23 ' 24
                             Eastern Standard Time
   Fig. 17-2
   REFERENCES
Average  Diurnal  Patterns of Ragweed Pollen
at Saratoga  Springs.
   1.  Ogden, E.  C.,  G.  S.  Raynor, and J.  V.  Hayes.   Travels of Airborne
       Pollen.  Progress  Report 11.  New York  State Museum and Science
       Service, Albany,  N.Y.  February 1971.   25  p.

   2.  Ogden, E.  C.,  G.  S.  Raynor, and J.  V.  Hayes.   Travels of Airborne
       Pollen.  Progress  Report 12.  New York  State Museum and Science
       Service, Albany,  N.Y.  February 1972.   26  p.
                                   81

-------
                           SECTION XVIII

            AIRBORNE POLLEN FROM ENTOMOPHILOUS PLANTS


The pollens of most flowering plants are not found in the atmosphere
far from the source plants.  They are seldom seen in atmospheric samples
and not reported in airborne pollen surveys.  Such pollens are transported
by agencies other than wind, primarily by insects.  As a consequence,
they are generally ignored when searching for possible causes of
pollinosis.  However, some of these entomophilous species emit
pollen that becomes airborne in appreciable amounts close to the source
plants.  In some situations an individual allergic to such pollen might
be exposed to concentrations sufficiently high to cause pronounced
discomfort.  These concentrations would vary greatly according to time
of year, time of day, weather conditions, abundance of flowers, pollen
characteristics, and other factors.  Thus, the concentrations in the
vicinity of such plants unde<^ :'.:vorable conditions during anthesis
would vary from zero to some appreciable number.

Durham1 called attention to several of these.  He reported rather high
concentrations of their pollen in the air near the entomophilous
species: Melilotus alba, Pyrus oommunis3 Chrysanthemum leuoanfhemwn,
Solidago oanadsnsis,  Cichoriun intybus, and Brassica niger,   Burchill2
reported high concentrations of apple pollen (Pyrus mains] in an apple
orchard, especially during early afternoon, with a maximum of 2385/m3.

There are many published references to airborne pollen in the vicinity
of source plants which are primarily, at least, pollinated by insects.
These data are seldom from volumetric measurements.  Daily rhythms of
pollen presentation were reported by Percival3 for several entomophilous
species.  These suggest when maximum concentrations might be expected.

As time permitted, concentrations of pollen in the air were measured
close to a large number of species of flowering plants known to be
entomophilous or essentially so.  Battery-powered rotoslide samplers

                                 82

-------
were operated 1.5 m above the ground (average adult nose level) and
1 m or more  from the nearest flowers.  In all cases, 1 m3 or more of
air was  sampled.  Although the rotoslide samples ragweed pollen at an
efficiency of ±64% in the wind speeds usually encountered, its
efficiency for the larger entomophilous pollens was estimated to
average  ±80%, which was used in all calculations.  In a few cases,
concentrations were measured also 0.1 m from the flowers.  Attempts
were made to determine the maximum concentrations that might occur by
sampling downwind of large plants or large numbers of plants, choosing
the time of  season, time of day, and kind of weather thought to be
most favorable for pollen emission.  Several common species were
sampled  several times with greatly different results.  Many times
the number of samples taken would be needed for properly determining
maximum  possible concentrations.  Although the concentrations of
airborne entomophilous pollen do not closely approach the concentrations
of pollen found close to many anemophilous plants, the following data
indicate significant amounts.  All figures represent the maximum
recorded at  1.5 m above ground and at least 1 m from nearest flowers.
Those below  10/m3 are not listed.
Sambueus eanadensis 125,000, Sorbaria sorbi folia 3736, Spiraea
vanhouttei 1272, Primus serotina 305, Hypericum perforation 286,
Daucus carota 281, Cleame Lutea 253, Pyrus mains 208, Galium vemm 178,
Lotus oorniaulatus 160, Rubus allegheniensis 160, Solidago juncea 157,
Thaliotr-im polygamwn 130, Potentilla reata 121, Ceanothus ameriaanus 114,
Aster novae-angliae 113, Viburnum cassinoides 111, Eupatoriun
maculatum 106, Galium palustre 103, Spiraea latifolia 98, Ranunculus
aaris 77, Centaurea maeulosa 76, Aesaulus hippooastanum 75, Hesperis
matronalis 72, Prunus virginiana 66, Anthemis ootula 65, Brassiaa kaber  64,
Barbarea vulgaris 57, Taraxacun offiainaie 57, Rosa sp.  55, Soiidago
rugosa 55, Smilacina raaemosa 54, Syringa vulgaris 53, Sisymbriun
altissimun 52, Ciohorium intybis 52, Senecio gldbellus 51, Rudbeokia
hirta 39, Erigemn strigosus 38, Pastinaaa sativa 38, Rosa rugosa 34,
Conyza eanadensis 33, Sambueus simpsonii 31, Polygonum ouspidatun 31,
Chrysanthemum leuoanthemum 23, Saxifraga virginiensis 22, Helianthus
annuus 21, Amelanc'hier oanadensis 18, An.aphalis margaritacea 18,
Erigeron annuus 17,  Prunus inaisa 16, Berteroa inoana 16, Melilotus

                                 83

-------
alba 14, Carvun aarvi 14, Philadelphus aoronarius 13,  Cornus  racemosa 12,
Eupatorium rugosum 12, Lonicera tatariaa 12, Ceanothus ovatus 11,
Meli,lotus offioinalis 10, Tanacetwn vulgare 10, Dipsaous laeiniatus 10.

The surprisingly high concentration of Sambuaus eanadensis  pollen
(125,000/m3) was downwind of and below a large patch of the plants
on a side hill and approaching the end of its pollination season.
The air was calm at the beginning of sampling but became gusty during
the sampling period.  The sample was processed too late for repeated
sampling and for obtaining counts from the earlier flowering  Sambuaus
pubens.  These await the next pollination season.  The concentration
of 31/m3 for Sambuaus s-impsonii was also unexpected as the  bush was
small and isolated.

The Spiraea was sampled five times at 1.5 m with pollen grains/m3
of 1272, 450, 374, 184, and 79.  The highest count was a morning sample
with many grains in clumps; the others were afternoon  samples with
less clumping.

Prunus serotina concentrations were 305 at 1600 EST, 91 and 21 from
separate trees at 0730, and 17 at 1240.

Hyperioum perforate midday concentrations were 286, 282, 206, and 67.

Dauaus aarota midday concentrations were 281, 30, 26,  and 18.

Solidago juncea afternoon concentrations were 157, 64, 43,  and 32.
A forenoon concentration of 38 was measured.

Lotus Gorniaulatus midday concentrations were 160, 148, 117,  98, 78,
65, 61, 40, and 13.  At first glance, it would seem strange that
pollen from this plant could become airborne.  Its papilionaceous
flowers have the stamens completely enclosed.  Pollination  is effected
by insects heavy enough to depress the keel petal.  However,  in age
the petals wither and expose the stamens and the now dry pollen.
                                84

-------
Simultaneous samples at 1.5 m and at 0.1 m from the nearest flowers
indicated concentrations of grains/m3 as follows:
                                        1.5          0.1
   Gal-imn vemm                         178          518
   Lotus cornisulatus                   148          587
   Potentilla Teota                     121          303
   Lotus oorniculatus                   117          308
   Ceanothus amerioanus                 114          233
   AnthemLs cotula                       65          157
   Brassiaa kaber                        64           54
   Taraxacum officinale                  57          153
   Daucus carota                         26           94
     11                                   18           37
   Berteroa ineana                       13           20
   Chrysanthemum leuoanthemwn            11           35
   Ranunculus aaris                      10           51
   Melilotus offiainalis                 10           19
   Origanum vulgare                       9           40
   Saponaria offioindlis                  3           20

Although more than 200 samples were taken from 120 species,  it
should be emphasized that these data are preliminary and might be
greatly changed through further sampling.  Most of the samples were
obtained and processed during odd moments in  a busy schedule.   Time
did not permit as full a coverage as was hoped and originally  planned.

REFERENCES
1.  Durham, 0. C.  The Volumetric Incidence of Atmospheric Allergens. V,
    Spot Testing in the Evaluation of Species.  J  Allergy 18:231-238,
    July 1947.

2.  Burchill, R. T.  Air-borne Pollen in Apple Orchards. Rept East
    Mailing Res Sta for 1962 (London),   p.  109-111, 1963.

3,  Percival, M. S.  Floral  Biology.   New York,  Pergamon Press,
    1965.  243 p.
                                85

-------
                            SECTION XIX

             OCCURRENCE OF AIRBORNE POLLENS IN WINTER
                         Donald M. Lewis*
Deposition of pollen on the rooftop of the Education Building Annex
in Albany, New York has been monitored continuously since March 1966.
The discovery of pollen types several weeks in advance of the flowering
of local species producing pollen of the same or similar morphology
posed the question as to the origin of these pollens and whether exotic
types originating in the southern and western United States might
be detected in midwinter.  Pollen recovered from November through
February accounts for only 0.1  or 0.2% of the yearly pollen catch.
Most of this is damaged to some degree and is likely refloated from
local sources, but some is definitely fresh and can be vaguely
attributed to distant source areas.  Although totals for this winter
season are commonly low (less than 200/cm2), the few types found
in fresh condition are occasionally trapped during brief periods
in amounts comparable to seasonal maxima for the same pollen types
produced locally but many weeks later.  So far, these fresh pollens
have never trr-n recovered in fair weather and failure of other
workers to record their presence is probably due to the use of
samplers which do not function properly in wet or cold weather.

The sampling location (EBA) is in downtown Albany, 38 m above street
level.  The building is immediately bordered on the north by a
small valley with a waste strip of weedy herbaceous and shrub growth
and a few mature trees of Aae? platanoides, A.  sacohaYin-m,
Ailanthus (iltissima, Robinia pseudoaaaoia, and Satix fragiZi-s',
on the west by the blocklong State Education Building (slightly higher
than the EBA); on the south by the State Capitol; and on the east
by a small park and municipal and business buildings of downtown
*New York State Museum and Science Service, Albany, New York

                                86

-------
Albany at gradually lower elevations.
METHODS
Daily and weekly deposition samples were taken in replicates throughout
the local flowering season and for at least two or three weeks before
and after those dates, or from about March 1  to October 15.   Weekly
samples only were taken through the winter months except for special
short-term samples taken in conjunction with  periods of unusual
weather.

These samples were AEC-fallout papers1 attached to 1 x 3-inch
microscope slides with DC-269 pressure sensitive adhesive.   They were •
fastened to a 1 m2 redwood board which is moored in the center of the
roof.  The roof itself is a layer of crushed  stone on an asphalt base.
Pollen becomes washed into the loose material  and samples from the
stones are removed and deposited on Mi Hi pore  filters, from  time
to time, as checks for types and physical conditions of the  most likely
contaminating pollens.  The sampling board is  cleaned at intervals
to prevent a build-up of contamination.  Regular sample positions are
assigned to the board as 1,2,3,4,...,etc.  The long axis of  each
microscooe slide is oriented in an east-west  direction.  Sample  number  3
in the northwest quadrant of the board was examined in this  study
with occasional comparisons with sample number 4 in the southeast
quadrant and other special samples.  In the laboratory, the  slides
were mounted in glycerine jelly prestained with basic fuchsin under
a 22 mm square coverglass so that something less than one-half of the
sample surface is prepared.  The remaining portion can be examined
for uniformity of pollen distribution or for other studies.

Particles adhere firmly to AEC-fallout paper,2 (perhaps requiring a
short residence time), and are not subsequently moved by mounting
procedures.  The retention of all particles coming in contact with
                                87

-------
the adhesive is likely not uniform.   Previous tests  indicate  high
"bounce-off" when this adhesive is used with powered impaction  samplers
and although tests have shown that pollen carried down  in  rain  is  at
least sometimes sampled with high efficiency, it is  likely that drop
size, intensity of precipitation, temperature,  and other factors
influence catch.1  The effectiveness of catch during storms has been
compared to Mi Hi pore-filtered samples of rain  collected in battery
jars on numerous occasions, but these have not  been  fully  analyzed.

Most pollen falling in rain must be directly impacted on the  adhesive.
The adhesive is not wetted and it is probable that pollen  which is not
easily wetted and is carried on the surface of  the drop is selectively
sampled over hygroscopic particles.3  Sampling  of pollen in snow on
AEC-paper is probably often a matter of chance.  Snow on this roof
melts sooner than that of surrounding rooftops, but  commonly  through
late December and January brief periods of slow thawing and many
hours of below freezing temperatures will ice the board in, so  that
on two occasions samples were not changed for two to four  weeks.
The insulating effect of the board accentuates  this  condition,  so
that it may be covered with a dome of ice three to four inches  thick
at the center.  Under these conditions, the sampler  is  unretentive
to new po'ilen and will lose much of that in the ice.

Unsheltered slides covered with AEC-fallout paper have  been compared
directly with Modified Durham samples on numerous occasions in  the
tree flowering seasons of 1961 and 1962 at Brookhaven National
Laboratory.  The catch on AEC-fallout paper was 8 to 16 times that
of the Modified Durham.  The samples discussed  here  caught 5  to 8
times as much pollen over the local  flowering season as that  collected
by us1* on standard Durham samplers on the Education  Building  in 1953,
1954, and 1955 and on the EBA in 1962 by Hayes.5  The greater catches
may be due to a number of factors acting singly or in combination.
                                88

-------
With experience it is usually possible to determine if pollen and other
spores are strictly fresh or somewhat older (i.e., refloated).  Pollen
is recorded on data sheets according to condition by means of superscripts
which indicate if pollen is: eroded, degraded, crumpled, broken, obscured,
or exine-thinned.  Criteria for determining these categories are
modified from those used by fossil pollen workers.6  In addition,
conditions concerning only fresh pollen are employed: degeneration
of intine and cell contents, natural pigmentation of cell contents
and/or exine, changing staining reaction of exine and/or cell contents,
and capacity of the pollen to imbibe water.  These effects such as
staining or characteristic breakage patterns are often specific for
different types of pollen.

AEC-fallout paper provides an adhesive surface superior to any of
those in common use (silicone grease, rubber cement, DC-269, etc.)
for the trapping and retention of pollen carried down in rain and
also for the retention of pollen sampled before rains.  Dry
deposition and washout by rain should be about equally effective
over long periods in humid climates.7  The unsheltered AEC-fallout
paper will collect some of this pollen in rain, but the effectiveness
of the process relative to dry deposition is unknown.  More data
are required for a fuller understanding of the dimensions of this
effect.  The rain-sheltered Durham sample will collect virtually
no pollen brought down in rain.

It is known that horizontal  microscope slides in holders similar to
the Durham sampler produce an edge effect in moderate winds which can
strongly alter pollen catch.8  The random distribution of pollen over
the slides and the remarkable uniformity of catch between replicate
slides on the sampling board indicate that under average conditions
edge effects are nil  for these deposition samples.  Silicone grease
slides (and other adhesives) have been compared with AEC-fallout paper
in parallel  tests on the sampling board during times of abundant
                               89

-------
airborne pollen.  There is little difference in catch during periods
of fair weather indicating that "bounce-off" of particles from the
paper is not normally a factor under these conditions.   In stormy
weather, however, catches on silicone grease slides are frequently
much less than those on AEC-fallout paper and at times  the grease
is almost completely removed by rain.  Samples previously collected
during fair weather and re-exposed during stormy periods commonly
show high losses for silicone grease slides, but no detectable loss
for AEC-fallout paper.

The reasons, then, for the much higher catch of pollen  on unsheltered
AEC-fallout paper samples compared to silicone greased, horizontal
slides in Durham-type holders seem to be these:
     (1)  In moderate wind speeds an edge effect reduces catch on
     the 1.5-meter-high Durham samples relative to the  roof level
     deposition samples.
     (2)  Rain or evaporation of condensed moisture will remove
     adhesives such as silicone grease and consequently reduce
     pollen catch.
     (3)  Pollen carried down in rain is sampled by unsheltered
     AEC-fallout paper, but is not sampled by rain-sheltered
     greased slides.  The efficiency of this collection is
     probably highly variable under different weather conditions.
     (4)  There is greater opportunity for contamination by
     redeposited pollen on deposition samples than on Durham
     samples, although precautions are taken to prevent it.
It seems that reason No. 1 would be most important in explaining these
differences in catch, although there are not enough experimental data
for samples taken in periods of fair weather, moderate  wind speeds,
and abundant pollen to corroborate this.  At somewhat increased wind
speeds, wind tunnel experiments have shown that the "edge shadow"
                                90

-------
effect decreases for horizontal slides and pollen trapping efficiency
increases due to turbulent impaction.7  Under these higher wind speeds,
AEC-fallout paper may trap relatively less pollen due to the high
"bounce-off" factor which has been demonstrated in its use as an
adhesive for impaction samples.

Pollen carried down in rain (reason No.  3) is an unknown quantity.
On one occasion, AEC-fallout paper collected with nearly 100%
efficiency as compared to Millipore-filtered rain water.  At other
times, additional large areas of the paper had to be examined to
find the same pollens collected on the filtered samples, with collection
efficiency probably somewhere around 10%.   It would seem that this
method of collection would not account for the large differences in
total seasonal pollen.

RESULTS
Weekly samples for the four winters of 1966 through 1970 and many
subsequent samples into the winter of 1973-74 have been examined.
Total pollen recorded for the months of  November through February
have been:  155 grains/cm2 for 1966-67; 94/cm2 for 1967-68; 155/cm2
for 1968-69; and 71/cm2 for 1969-70.  Pollen recovered in January
and February has consistently outnumbered  that for November and
December by about two to one in spite of the fact that sampling
may not be active due to weather conditions in the later months.
Pteridophyte spores are recorded,  but not  included in the totals.
Lycopodium spores reach their yearly maximum in the first half of
November.

Pollen frequency is expected to be low on  most winter samples and
a minimum of 4.84 cm2 sample surface is  examined for each of these.
The pollen caught in the months of November and December is almost
all from redeposited pollen, mostly from those local  plants which
                                91

-------
last produced airborne pollen in quantity:  Ambrosia,  miscellaneous
Compositae of several types; Chenopodiaceae and Amaranthaceae
(Cheno-Ams), and Gramineae.  The ubiquitous pollen of Finns  is
found practically year around in at least small  amounts.   Some  pollens
are identifiable much longer after normal  flowering than  others
because of more recognizable features and innate chemistry which
allows them to withstand degenerative processes.   Soon  after their
normal flowering periods, the pollen of most plants are either  not
represented in the record or are found in sporadic trace  amounts.
The tendency of pollen to be refloated (after once being  impacted
on vegetation, for instance) is slight.7  Large pollens (e.g.,  winged
pollen of conifers)  and clumped pollen might be expected  to  be
refloated more often than others, since it would be easier for  them
to escape through the laminar boundary layer.  Common airborne  pollens
of spring flowering  trees are found sporadically throughout  the
summer and fall (Betula, Quereus); but others, produced at about
the same time in fairly large amounts, are  rarely found after their
normal season (U1must Populus,  Fraxinus].   It is felt that later
recovery of most of  these is a measure of their ability to withstand
degradation.  An increase in refloated pollen is seen in  the fall.
This is mostly from  fall-flowering plants,  but also includes many
of the spring-flowering trees.   This increase is thought  to  be  due
to increased wind speeds and particularly to senescence of vegetation.
Occasionally, portions of leaf epidermal tissue, stomates, etc.
can be recognized on the samples.  Pollen and fungus  spores  are
sometimes rafted with these tissues and it is deduced that many of
the tree pollens which show effects of age are released by leaf fall
and the death and shattering of herbaceous  vegetation.  Also, the
depletion of foliage will create better conditions for  the dispersal
of particles by the  increase of wind speed through trunk  spaces, etc.9
There are differences in condition of refloated pollen  and these should
be expected when the variety of surfaces from which they  were dispersed
and their microenvironments are considered.  Fungus spores and  hyphae
                                92

-------
may be intimately associated with "old" pollen.

Most of the fall-collected pollen appears old.   Ambrosia is frequently
eroded with the spines blunted and sometimes with one side abraded flat.
The same conditions are common for other pollens.  The exceptions are
rare collections of Artemisia, Cheno-Ams, or pollens of unknown
affinity which appear strictly fresh.  Rarely a grain or two of some
member of the Cupresssaceae is found.  This type of pollen shows age
quite rapidly by a reversal of staining reaction and fresh pollen is
usually easily recognized.  The "out-of-season" Cupressaceae appears
fresh; and those rare grains found in November  and December may have
been derived from southwestern or southern states.  These grains become
somewhat more common in January and February at the same time that
records show increased catches in the deep south.10'11  The four
winters analyzed show only an average of 40 pollens per cm2 for
November and December.  Of these, probably no more than four or five
would be considered "fresh" from late composites or Cheno-Ams in
frost protected situations, and probably less than one might have
been derived from very distant (one or two thousand km) sources.

Some minor constituents of the spring pollen flora are occasionally
found in fall  in amounts higher than would be expected.  Old pollen of
Pioea and Abies, for example, are sometimes found in low amounts in
November and December, but these amounts are often equivalent to
their local  in-season maxima.  The possibility  of refloatation of
these pollens from a source area perhaps 100 km to the north where
the producing plants are much more numerous should not be ignored.

Old pollen of Zea is found as commonly in the fall as in season,
although in  very low amounts.  Erdtman12 sampled pollen of Zea in
December which was presumably derived from old, desiccated plants
a short distance away.  Source areas here are some distance away,
but late fall  winds or late harvesting of corn  for grain may release
old pollen trapped on the plants.

                                93

-------
Most of the "fresh" pollen sampled in winter is trapped from late
January through February and is, so far, always associated with stormy
weather.  The types commonly encountered are:  Taxodivm, Cupressaceae,
Alnus, and ulmus.   One of the first significant records for distant
"fresh" pollen in winter was from a rain sample of January 27,  1967.
The rain was deposited on a Millipore filter.   Numerous pollens were
seen, some obviously fresh, in a tangle of plant hairs and industrial
debris.  A number of cupressaceous pollen types appeared to have
"papillae", but these were not taken seriously since the optical
qualities of the sample were poor.  The filter was dissolved in acetone
and the pollen was recovered by centrifugation.  It was felt that a
considerable amount of pollen was lost by this method.  About 600 fresh
pollen were counted on four standard microscope slides, or about one
fresh pollen per square centimeter of the sampling surface.  The sample
was collected during a seven-hour rain storm following a week of
unusually warm weather (highs of 50°F.).  The majority of this  pollen
was of cupressaceous types.  The "papillae" seen were leptoma of
pollen of the Taxodiaceae.  Unfortunately, centrifugation broke
most of these, but it was estimated that perhaps one-fourth of
the cupressaceous types were actually Taxodiaceae.  In later years,
identification techniques have been employed which allow a
preliminary screening of probable taxodiaceous pollens at low
magnifications.  About one-fourth of the total fresh pollen was Alnus.
By chance, these rain samples were collected at the same time the
weekly AEC-fallout paper samples were changed, so that the total
rain is represented by two weekly samples.  These weekly samples
average twice the total pollen trapped in the preceding and following
weeks.

In the period from February 10-17, 1967, there were 38 pollen grains/cm2
for the week.  This was a period of mild weather with rain early
on February 16.  The weather became cold and so windy that the
sampling board was removed until the regular weekly change the  next
day.  The rain samples are inadequate and it is not known if the pollen
                               94

-------
recorded occurred in rain or through the mild part of the week.
Experience in subsequent years would indicate that the pollen arrived
with the rain on the morning of February 16.  Of the total, 13 were
cupressaceous pollen and 12 were Alnus.   There are 212 pollens of
Cupressaceae recorded per cm2 for the year; 48 of these occur
"out-of-season".  There are 90 Alnus pollen recorded per cm2 per year;
21 are "out-of-season".

Most of these "out-of-season" pollens of the four taxa mentioned
as common are "fresh" and are generally recovered in two to four
episodes of precipitation in each of the four winters studied.  Spot
samples in subsequent winters have shown that Taxodiun pollen can be
expected at least once in February and occasionally in late January.
Local airborne pollen records and regional  floras would indicate that
the source of this pollen is somewhat south of the Carolinas.  ulmus
is occasionally found in low amounts at about the same time as Taxodium.
It is thought to have a similar source area.  Alnus pollen and that of
the Cupressaceae is found along with that of the other taxa and later,
sporadically, until the beginning of their local flowering seasons.
Other pollens are found well in advance of their local flowering
season, but this is after March 1 and they are not the subject of
this report.

Some winter rains and snows are exceptionally clean.  There may be
some early washout of homeheating and industrial pollutants, but
little else.   Some snowstorms accompanied by strong winds may bring
fairly large  quantities of "old" pollen.  In many cases this is
thought to be scoured from available pollen-holding surfaces (rooftops,
branches, etc.)  and redeposited at the sampling site.  But sometimes
most of these surfaces are already covered with snow which does  not
contain pollen.   The pollen appears to be associated with the storm
system, but its  source is puzzling.
                                95

-------
CONCLUSIONS
A small amount of pollen was recorded on samplers between  November 1
and March 1.   Most of that for November and December was  probably
refloated from locally derived, "old" pollen.   About twice this  amount
of pollen was trapped during January and February and about one-half
of that was "fresh".  So, for the entire winter season about one-third,
which would amount to a maximum of 0.1% of the yearly pollen total,
would be considered to be "fresh" pollen of distant origin.   The
point of origin, based on flowering time, regional  floras, and
aeropalynological records is probably the southern  to southwestern
United States at distances of over 1,000 km from the sampling site
in Albany, New York.  Some of the best records of "fresh"  pollen in
winter are at the end of weeklong warm periods followed by rain  or
snow.  This does not mean that unusually early flowering  times  occurred
only a few hundred km to the south.  Taxodim, flowering  in late
January and early February for instance, is six weeks in  advance of
its normal flowering in North Carolina.  Most  of this "fresh" pollen,
perhaps all of it, is carried down in precipitation.  These records
are for pollen sampled on AEC-fallout paper.   It is known  that  the
efficiency of this material  is quite variable  for particles carried
in rain drops.  A better method of assessing airborne pollen in
winter would be to take all-weather samples on AEC-fallout paper
and compare them to precipitation samples taken in  jars and deposited
on molecular membrane filters.  It is likely that the winter totals
and those for "fresh" pollen would be appreciably higher.


REFERENCES
1.  Posinski, J.  Some Studies on the Evaluation of Gummed Paper
    Collectors Used in Determining Radioactive Fallout.  Amer Geophys
    Union Trans 38(6):857-863, December 1957.
                               96

-------
2.  Lewis, D. M.  Comparisons of Adhesives and Mounting  Media.
    In:  Travels of Airborne Pollen. Progress Report 8,  Ogden,  E.  C.,
    G. S. Raynor, and J. V. Hayes.   Albany,  New York State  Museum
    and Science Service, 1968.   p.  4-8.

3.  Gregory, P. H.  The Leaf as a Spore  Trap.   In:   Ecology of
    Leaf Surface Micro-organisms, Preece,  T.  F.  and C. H. Dickinson
    (eds.).  New York, Academic Press, 1971.   p.  239-243.

4.  Ogden, E. C., and D. M. Lewis.   Airborne  Pollen and  Fungus  Spores
    of New York State.  Albany, New York State Museum and Science
    Service Bulletin 378, January 1960.   104  p.

5.  Hayes, J. V.  Comparison of the Rotoslide and Durham Samplers  in
    a Survey of Airborne Pollen.  Ann Allergy 27:575-584, November 1969.

6.  Gushing, E. J.  Redeposited Pollen  in  Late-Wisconsin Pollen Spectra
    from East-central  Minnesota.  Amer J Sci  262(9):1075-I088,
    November 1964.

7,  Chamberlain, A.  C.  Deposition  of Particles  to  Natural  Surfaces.
    In:  Airborne Microbes, Gregory, P.  H. and J.  L.  Monteith (eds.).
    London: Cambridge University Press,  1967.   p.  138-164.

8.  Gregory, P. H.  Microbiology of the  Atmosphere, 2nd  ed.  New York,
    John Wiley and Sons, 1973.   377 p.

9.  Raynor, G. S., J.  V. Hayes, and E. C.  Ogden.   Particulate Dispersion
    Into and Within  a  Forest.   Accepted  by:   Boundary Layer Meteorology.
                               97

-------
10.  Wodehouse,  R.  P.   Hayfever Plants.   Waltham,  Chronica  Botanica Co.,
     1945.  245  p.   Table on seasonal  occurrence  on  page  180.  This table
     is deleted  in  the revised edition,  Hafner  Press,  1971.

11.  Wawrzyn, B.,  and  J.  Tomb!in (eds.)   Statistical  Report of the
     Pollen and  Mold Committee of the  American  Academy of Allergy,
     1973.  Columbus,  Ross Laboratories, 1974.

12.  Erdtman, G.  An Introduction to Pollen  Analysis.  Waltham,
     Chronica Botanica Co.,  2nd printing, 1954.   239 p.
                                 98

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 1. REPORT NO.
   EPA-650/3-75-003
                              2.
                                                           3. RECIPIENT'S ACCESSI ON«NO.
 4. TITLE ANDSUBTITLE
                                                           5. REPORT DATE
                                                             January 1975
  Travels  of Airborne Pollen
                                                           6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
  Eugene C.  Ogden, Gilbert  S.  Raynor  and Janet V.
    Brookhaven Natl. Laboratory	
                                                           .8. PERFORMING ORGANIZATION REPORT NO
        Hayes
9. PERFORMING ORGANIZATION NAME AND ADDRESS

   University of the State  of New York
   Albany,  New York  12234
              10. PROGRAM ELEMENT NO.

               1AA009     (21ADO
              11. CONTRACT/GRANT NO.

               800677
 •2. SPONSORING AGENCY NAME AND ADDRESS

  Office of Research  and  Monitoring
  U.  S.  Environmental Protection Agency
  Washington, D. C.   20460
              13. TYPE OF REPORT AND PERIOD COVERED
               Final  10/1/70  -  12/31/ 74
              14. SPONSORING AGENCY CODE
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT

        The following studies  were conducted on  the  transport and dispersion  of air-
   borne pollen:  (a)  Development and evaluation  of sampling devices for  pollen;
   (b}   development and evaluation techniques for  tagging pollen in living plants with
   dyes  and radioisotopes;  (c)   dispersion and deposition of pollen from known  sources
   of various configurations;  (d)   effects of forested  areas on the removal  of  pollen
   from  the atmosphere; (e)  concentration variations of pollen from natural  sources
  with  distance, height, time and other variables;  (f)   feasibility of predicting
   ragweed pollen concentrations from unknown sources;   (g)  measurements  on  ragweed
   pollen concentrations  in  a  large source-free  area; and (h) comparisons  of the
   ragweed pollen concentrations before and after  ragweed eradication efforts.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
   Air Pollution
   Pollen Dispersion
b.lDENTIFIERS/OPEN ENDED TERMS  C. COSATI Field/Group
  Dispersion of Airborne
  Pollen
Primary
06 Biological  ancj
Medical Sciences
0603 Biology
Secondary
04 Atmospheric S<
0401 Atm. Physics
 3  DISTRIBUTION STATEMENT
  Release Unlimited
                                              19. SECURITY CLASS (This Report)
                                                Unclassified
                           21. NO. OF PAGES

                            106
20. SECURITY CLASS (Thispage)
  Unclassified
                                                                         22. PRICE
EPA Form 2220-1 (9-73)
                                            99

-------
                                                          INSTRUCTIONS

    1.    REPORT NUMBER
         Insert the EPA report number as it appears on the cover of the publication.

    2.    LEAVE BLANK

    3.    RECIPIENTS ACCESSION NUMBER
         Reserved for use by each report recipient.

    4.    TITLE AND SUBTITLE
         Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
         type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
         number and include subtitle for the specific title.

    5.    REPORT DATE
         Each report shall carry a date indicating at least month and  year.  Indicate the basis on which it was selected (e.g., date of issue, date of
        approval, date of preparation, etc.).

    6.    PERFORMING ORGANIZATION CODE
         Leave blank.

    7.    AUTHOR(S)
         Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.).  List author's affiliation if it differs from the performing organi-
        zation.

    8.    PERFORMING ORGANIZATION REPORT NUMBER
         Insert if performing organization wishes to assign this number.

    9.    PERFORMING ORGANIZATION NAME AND ADDRESS
         Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hirearchy.

    10.   PROGRAM ELEMENT NUMBER
         Use the program element number under which the report was prepared.  Subordinate numbers may be included in parentheses.

    11.   CONTRACT/GRANT NUMBER
         Insert contract or grant  number under which report was prepared.

    12.   SPONSORING AGENCY NAME AND ADDRESS
         Include ZIP code.

    13.   TYPE OF REPORT AND PERIOD COVERED
         Indicate interim final, etc., and if applicable, dates covered.

    14.  SPONSORING AGENCY CODE
         Leave blank.

    15.  SUPPLEMENTARY  NOTES
         Enter information not included elsewhere but useful, such as:  Prepared  in cooperation with, Translation of, Presented at conference of,
        To be published in, Supersedes, Supplements, etc.

    16.  ABSTRACT
         Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the report contains a
        significant bibliography or literature survey, mention it here.

    17.   KEY WORDS AND DOCUMENT ANALYSIS
         (a) DESCRIPTORS • Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
         concept of the research  and are sufficiently specific and  precise to be used as index entries for cataloging.

        (b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
        eniioJ terms written in descriptor form for those subjects for which no descriptor exists.

        (c) COSATI MELD GROUP - Field and gtoup assignments are to be taken from the 1965 COSATI Subject Category List.  Since the ma-
        jority of documents are multidisciplmary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
        endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
        the primary postmg(s).

    18.   DISTRiBUTION STATEMENT
         Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
         the public, wiui address and price.

    19. e. :•:.•  :>EC!jRiTY CLASSIFICATION
         DO NuT ubmit ciarsfY.d  reports to the National Technical Information service.

    21.   NUMBER OF PAGES
         Insert the total number  of pages, including this one and  unnumbered pages, but exclude distribution list, if any.

    22.   PRICE
         Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1  (9-73) (Reverse)
                                                      100

-------
 -o

 07
cn
O
CO
O
O
CO
                                      NO 3,
" o
&£„
P * 2
                                       0) (V
                                       Q. —
                                       Q. _.
                                       10 CX
                                       -, o
 pi
 .-. o

 ^s
 O (D
 S5-
 tO (D
 £5
 g«
 
                                                              o
                                                              m

                                                              n

-------