Integrating the biological and physical components of maize

pollen dispersal

Mark Westgate, Ray Arritt, and Susana GoggiIowa State University

Biological and Physical Components of Out-crossing

• Biological (source)– Pollen shed characteristics

• Timing, intensity, viability

• Physical (delivery system)– Topography

• Distance, elevation, wind breaks, border rows– Atmospheric conditions

• Wind speed, wind direction, stability index, mixing height, air temperature, relative humidity

• Biological (receiver field)– Pollen shed characteristics– Synchrony with female and adventitious pollen source – Specific combining ability

Structure of pollen production, Structure of pollen production, dispersal, and outdispersal, and out--crossing modelcrossing model

MeteorologicalAnalysis or

Model

Pollen Pollen TransportTransport Pollen ViabilityPollen Viability

WeatherData

TopographicDataCrop Data

Pollen Shed

ReceptorRisk

Pollen production Pollen production

MeteorologicalAnalysis or

Model

Pollen Transport Pollen Viability

WeatherData

TopographicDataCrop Data

Pollen Shed

ReceptorRisk

Hybrid A

0.0E+00

3.0E+05

6.0E+05

9.0E+05

1.2E+06

1.5E+06

1.8E+06

24-Jul 26-Jul 28-Jul 30-Jul 1-Aug 3-Aug 5-Aug

Polle

n gr

ains

she

d pe

r tas

sel

1 pl/m28 pl/m218 pl/m2Poly. (1 pl/m2)Poly. (8 pl/m2)Poly. (18 pl/m2)

Pollen shed intensity typically peaks Pollen shed intensity typically peaks 2 days after first anthers appear 2 days after first anthers appear

192 196 200 204 208Pol

len

shed

rate

(gra

ins

cm-2

d-1

)

Actual Predicted

Day of the year194 198 202 206 210

0

200

400

600

800

Per

cent

of p

opul

atio

n (%

)

0

20

40

60

80

100BeginMax End

P3978 P3925

100% MF 75% MF

PI

• Pollen shed characteristics• Timing, intensity, viability

Timing of pollen shed can be Timing of pollen shed can be simulated from tassel development simulated from tassel development

and population dynamicsand population dynamics

Diurnal patterns of pollen shed vary

7/29

020406080

100120140160180

6:00 8:00 10:00 12:00 14:00 16:00

Time of day

Polle

n gr

ains

/cm

2Temp, RH, and wind speed affect the initiation and

intensity of pollen shedShed typically begins at RH < 90%

Pollen dispersal Pollen dispersal

MeteorologicalAnalysis or

Model

Pollen Pollen TransportTransport Pollen ViabilityPollen Viability

WeatherData

TopographicDataCrop Data

Pollen Shed

ReceptorRisk

• Physical (delivery system)– Topography

• Distance, elevation, wind breaks, border rows– Atmospheric conditions

• Wind speed, wind direction, stability index, mixing height, air temperature, relative humidity

Adapted from Di-Giovanni and Kevan, 1991

Gaussian plume

Statistical

Diffusion

Adapted from Di-Giovanni and Kevan, 1991Adapted from Di-Giovanni and Kevan, 1991

Gaussian plume

Statistical

Diffusion

Pollen transport models• Lagrangian random-walk approach

– the pollen cloud is represented as a population of virtual “particles”

– each virtual particle can be traced according to its source, path, or other property of interest

• Gaussian plume approach– the pollen cloud is ‘constrained’ to a normal

distribution that is modified hourly by local atmospheric conditions

Physical parameters for modeling maize pollen dispersal are fairly well established

---- weight/grain: weight/grain: 250 to 350 250 to 350 ngng---- diameter: diameter: 50 to 90 50 to 90 µµMM---- density: density: 1.25 to 1.45 g cm1.25 to 1.45 g cm--33

---- settling velocity: settling velocity: 20 20 –– 32 cm s32 cm s--11

---- number per plant: number per plant: 0.5 to 6 x 100.5 to 6 x 1066

---- duration of shed: duration of shed: 4 to 6 days4 to 6 days

Field evaluation of the pollen transport models:

• 9 monitoring locations within the source field

• Receptors (sticky traps) placed at 5, 10, 30, 90, 165, 330, and 660 feet from the edge of the source field in the 8 major directional axes

• Pollen dispersal monitored from 0730 to 1700 hours.

Daily and seasonal patterns of pollen dispersal from a source field can be simulated fairly accurately

July 21, 2000

g/m2

1 pollen grain/4 cm2

at 0.25 pollen grain/cm2dpotential contamination ~ 0.28%

or ~ 3 kernels per 1000 exposed silks

100m

July 21, 2000

g/m2

1 pollen grain/4 cm2

at 0.25 pollen grain/cm2dpotential contamination ~ 0.28%

or ~ 3 kernels per 1000 exposed silks

100m100m100m

Lagrangian numerical simulation-- Arritt et al. ISU

Gaussian plume– ISCST3/AERMOD EPA models

…but both models tend to overestimate deposition near the source field, and underestimate deposition at greater distances

from Riese, 2004

Lagrangian model

Values summed for eight cardinal directions from source fieldPo

llen

depo

sitio

n (g

rain

s/cm

2da

y)

ISCST3

Need to account for modification of the 3D flow field by the crop (windbreak)and atmospheric turbulence on a larger scale (large eddy simulation).

Pollen viability decreases linearly with moisture content.

Pollen of the ‘average’ maize genotype loses viability completely at about 30% moisture

Average values for 11 genotypes

Initial viability 80%PMC at 0% 30%

Pollen/tassel 3.1E6

Loss of pollen moisture is an exponential function of VPD and time

tVPDePMC 0012.02.63 −=

0

200

400

600

800

1000

1200

0 20 40 60 80

Viability (percent)

Peak

Par

ticle

Hei

ght (

m)

Topeka PM Omaha PM

“Terminal viability” of lofted maize pollen

• Pollen lofted through two representative atmospheric soundings (updraft at 3 m/s, fall at 20 cm/s)

• Viability adjusted accounting for VPD through the profile until the pollen grain returned to the ground.

Brunet et al., 2004

Y-axis normalized to daily convective boundary layer, 800-2000m

OutOut--crossing crossing

MeteorologicalAnalysis or

Model

Pollen Transport Pollen Viability

WeatherData

TopographicDataCrop Data

Pollen Shed

ReceptorRisk

• Biological (receiver field)– Pollen shed characteristics– Synchrony with female and adventitious pollen source – Specific combining ability

““Nick ManagerNick Manager” converts daily estimates of pollen shed and silk emergence to ” converts daily estimates of pollen shed and silk emergence to simulate kernel set for any given field condition simulate kernel set for any given field condition

Temporal profile of silk exsertion (blue), temporal profile of pollen shed (black), and simulated daily values of kernel set (red).

Lizaso et al. 2003

Simulated kernel set in 13 seed production fieldsSimulated kernel set in 13 seed production fields

Fonseca et al. (unpublished)

Loss of pollen viability and pollen trapping by leaves not taken into account

0

10

20

30

40

50

60

Perc

ent o

f tot

al

obse

rvat

ions

0 2 4 6 8 10Percent outcross

199819992000

3-year seed industry study

Ireland et al 2004

A few mid-field samples had > 10% out-crosses

About 50% of the mid-field samples were free of out-crosses

Risk of OutRisk of Out--crossingcrossing

Field BRisk Index: 5.68Field BRisk Index: 5.68

Field CRisk Index: 4.65Field CRisk Index: 4.65

Early and late appearing silks at risk

Late appearing silks at risk

An example of simulated out-crossingresulting from “adventitous presence” late in flowering

Nick Manager Summary

0

5

10

15

20

25

30

195 200 205 210 215 220 225 230

Day of the year

Silk

s, K

erne

ls (h

a x

106 )

0

100

200

300

400

500

Pollen grains cm-2

Silks

Kernels

Silks at Risk

Local pollen

Adventitious pollen

Adventitious pollen peaked at 7.5 gr/cm2 on 216

Total kernel production21.8 mil kernels/ha

76.5% silks were pollinated98.4% genetically pure seed

1.6% out-crossed seed

Contamination by Pollen Movement

Seed FieldWest

Source

Durant, IA :1998Durant, IA :1998

2.0 2.0 2.0 2.0 0.0 0.0 0.250.25Percent OutPercent Out--crossingcrossing

12 14 12 14 16 16 18 18 20 20 22 2422 24JulyJuly

East PollenEast Pollen

West PollenWest Pollen

Seed field silkingSeed field silkingSeed fieldSeed field pollenpollen

WindRoseWindRose for Durant, IAfor Durant, IA1515--21 July 1998: 0800 21 July 1998: 0800 –– 1400 h1400 h

EastSource

100m

6:1:4:1

71a

100m

160a

80a

InterInter--industry Isolation Standards Studyindustry Isolation Standards Study

Seed FieldWest

Source

Durant, IA :1998Durant, IA :1998

2.0 2.0 2.0 2.0 0.0 0.0 0.250.25Percent OutPercent Out--crossingcrossing

12 14 12 14 16 16 18 18 20 20 22 2422 24JulyJuly

East PollenEast Pollen

West PollenWest Pollen

Seed field silkingSeed field silkingSeed fieldSeed field pollenpollen

12 14 12 14 16 16 18 18 20 20 22 2422 24JulyJuly

East PollenEast Pollen

West PollenWest Pollen

Seed field silkingSeed field silkingSeed fieldSeed field pollenpollen

WindRoseWindRose for Durant, IAfor Durant, IA1515--21 July 1998: 0800 21 July 1998: 0800 –– 1400 h1400 h

EastSource

100m

6:1:4:1

71a

100m

160a

80a

InterInter--industry Isolation Standards Studyindustry Isolation Standards Study