4>EPA
                      EPA/600/R-14/191 | August 2014 | www.epa.gov/research
United States
Environmental Protection
Agency
    Atmospheric LiDar Coupled
    with Point Measurement Air
   Quality Samplers to Measure
     Fine Particle Matter (PM)
    Emissions from Agricultural
Operations. Part 2 of the California
  2007 - 2008 Tillage Campaigns:
    Spring 2008 Data Analysis
      RESEARCH AND DEVELOPMENT

-------

-------
 Atmospheric LiDar Coupled with Point
  Measurement Air Quality Samplers to
    Measure Fine Particle Matter (PM)
Emissions from Agricultural Operations,
    Part 2 of the California 2007 - 2008
             Tillage Campaigns:
         Spring  2008 Data Analysis
                     David J. Williams
               Office of Research and Development
              National Exposure Research Laboratory
                Environmental Sciences Division
               Landscape Characterization Branch
              Research Triangle Park, North Carolina
                      Jerry Hatfield
                 National Soil Tilth Laboratory
                 Agricultural Research Service
              United States Department of Agriculture
                       Ames, Iowa
 Although this work was reviewed by EPA and approved for publication, it may not necessarily reflect
 official Agency policy. Mention of trade names and commercial products does not constitute
 endorsement or recommendation for use.
              U.S. Environmental Protection Agency
               Office of Research and Development
                    Washington, DC 20460

-------

-------
Atmospheric LiDAR coupled with point measurement air quality samplers to measure fine
particulate matter (PM) emissions from agricultural operations. Part 2 of the California 2007-
2008 Tillage Campaigns: Spring 2008 Data Analysis
Authors:

Jerry Hatfield
National Soil Tilth Laboratory
Agricultural Research Service
United States Department of Agriculture
Ames, Iowa

David Williams
Environmental Sciences Division
National Exposure Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC

James Sweet
Planning Division
Central Region Office
San Joaquin Valley Air Pollution Control District
Fresno, CA

Sona Chilingaryan
Air Division
Region 9
U.S. Environmental Protection Agency
San Francisco, CA
* Corresponding author. (919) 541-2573; williams.davidj@epa.gov
Disclaimer Notice: The United States Environmental Protection Agency through its Office of
Research and Development partially funded and collaborated in the research described here
under Interagency Agreement: DW 12922568 to Space Dynamics Laboratory. It has been
subjected to Agency review and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.

-------
Report Prepared By:

Dr. Gail Bingham
SDL Chief Scientist & Civil Space and Environment Division Leader
Space Dynamics Laboratory
1695 North Research Park Way
North Logan, UT 84341

Jennifer Bowman
Senior Manager, Internal Communications
ATK Aerospace Systems
P.O. Box 98
5000 S. 8400 W.
Magna, UT 84044

Christian Mar chant
Energy Dynamics Laboratory
1695 North Research Park Way
North Logan, UT 84341

Dr. Randal Martin
Research Associate Professor
Department of Civil and Environmental Engineering
Utah State University
41 10 Old Main Hill
Logan, UT 84322

Kori Moore
Environmental Engineer
Energy Dynamics Laboratory
1695 North Research Park Way
North Logan, UT 84341

Dr. Philip Silva
Environmental Chemist
Animal Waste Management Research Unit
USDA Agricultural Research Service
230 Bennett Lane
Bowling Green, KY 42 104

Dr. Michael Wojcik
Branch Chief for Environmental Measurement
Energy Dynamics Laboratory
1695 North Research Park Way
North Logan, UT 84341

-------
        Space Dynamics
                      LABORATORY
        Utah State University Research Foundation
California Spring  2008  Tillage Campaign:  Data

Analysis

A project performed for the San Joaquin Valleywide Air Pollution Study
Agency, Contract 07-1 AG

Submitted To:
James Sweet
Planning Division
Central Region Office
San Joaquin Valley Air Pollution Control District
Fresno, CA

David Williams
Environmental Sciences Division
National Exposure Research Laboratory
U.S. Environmental Protection Agency
RTF, NC

Sona Chilingaryan
Air Division
Region 9
U.S. Environmental Protection Agency
San Francisco, CA

Dr. Jerry Hatfield
National Laboratory for Agriculture and the Environment
Agricultural Research Service
United States Department of Agriculture
Ames, IA

Submitted By:
Space Dynamics Laboratory/ Utah State University Research Foundation
1695 North Research Park Way
North Logan, Utah 84341
                                                   DOCUMENT NUMBER:      SDL/08-556
                                                   REVISION:
                                                   DATE:                 JUNE 21,2013

-------
                             TABLE OF CONTENTS
TABLE OF ABBREVIATIONS/VARIABLES	ix
EXECUTIVE SUMMARY	1
1.  Background	3
   1.1    Literature Review	4
2.  Experiment Design	12
   2.1    Site Description	12
   2.2    Operation Description	12
   2.3    Tillage Operation Data	18
3.  Measurements and Methods	20
   3.1    Measurement Overview	20
   3.1.1      Meteorological Measurements	21
   3.1.2      Wind Profile Calculations	25
   3.1.3      Soil Characterization	26
   3.1.4      Air Quality Point Samplers	27
   3.1.5      Lidar Aerosol Measurement and Tracking System	32
   3.2    Dispersion Modeling Software	37
   3.3    Statistical Analysis of Data	40
4.  Results and Decisions	41
   4.1    General Observations	41
   4.1.1      Soil Characteristics	41
   4.1.2      Meteorological Measurements	43
   4.2    Aerosol Characterization Data	46
   4.2.1      Minivol Filter Sampler Data	46
   4.2.2      PM Chemical Analysis	48
   4.2.3      Aerosol Mass Spectrometer	54
   4.3    Optical Characterization Data	56
   4.3.1      Met One Optical Particle Counter	56
   4.3.2      Optical To PM Mass Concentration Conversion	61
   4.3.3      Lidar Aerosol Concentration Measurements	66
   4.4    Fluxes and Emissions Rates	67
   4.4.1      Lidar Based Fluxes and Emission Rates	69
   4.4.2      Inverse Modeling Calculations	73
   4.5    Derived Emission Rate Comparison	81

SDL/08-556                     California 2008 Tillage Campaign                              i

-------
5.  Summary and Conclusions	85
6.  Lessons Learned	90
7.  Acknowledgments	92
8.  Publications	93
9.  References	94
10. Appendices	99
   10.1   Appendix A: Data and Settings Tables	99
   10.2   Appendix B: Investigations into and conclusions from filter-based data	108
   10.3   Appendix C: Responses to comments received re: California Spring 2008 Tillage
          Campaign: Data Analysis Report	120
   10.3.1     DraftDate: 16 March 2011 (Organization: U.S. EPA)	120
   10.4   Draft Date:  15 April 2013 (Organization: San Joaquin Valley Air Pollution Control
          District)	126
   10.4.2     Presentation of Results and Discussion with the San Joaquin Valley Air Pollution
             Control District's Ag Technical Group: 22 April 2013	131
SDL/08-556                     California 2008 Tillage Campaign

-------
                               TABLE OF FIGURES
Figure 1. Histogram of disc operation emissions factors and both the empirical and the Weibull
cumulative distribution functions (CDFs)	11
Figure 2. Shaded relief map of the State of California, USA, with the location of the selected
sample site shown by the white star. Image from geology.com [33]	13
Figure 3. Satellite image of the  study location with soil types shown in white. The study fields
are outlined within the two rectangles. Soil type 130 represents Kimberlina fine sandy loam,
saline-alkali [34]	14
Figure 4. Photo taken standing near the southern edge of Field 5 looking north across Field 5
towards Field 4. Note the relative flatness of the terrain	15
Figure 5. Orthman 1-tRIPr in operation during this field experiment	16
Figure 6. Wind rose for May and June of 2005 - 2007 as recorded by the CEVIIS Station # 15
(Stratford). No calm periods were recorded	20
Figure 7. Sample layout used for Field 4	22
Figure 8. Sample layout used for Field 5	24
Figure 9. Two Airmetrics MiniVol Portable Air Samplers and a Met One Instruments Optical
Particle Counter (OPC) deployed for field sampling during Spring 2008 tillage study	28
Figure 10. The three wavelength Aglite lidar at dusk, scanning a harvested wheat field	32
Figure 11. The Aglite lidar retrieval algorithm flow chart, showing the input locations for the in
situ data	33
Figure 12. (A) Conceptual illustration of the method for using lidar to generate time resolved
local area particulate fluxes. (B) An example of a "staple" lidar scan over the facility showing
aerosol concentration on the three sides of the box	34
Figure 13. Example  of a lidar scan profile used to monitor PM concentrations around and
emissions from conventional tillage operations in Field 4. Each data point represents a 0.5
second averaging time, therefore data point 1000 was taken at time  = 500 seconds	36
Figure 14. Example  of a lidar scan profile used to monitor PM concentrations around and
emissions from conservation tillage operations in Field 5. Each data point represents a 0.5 second
averaging time, therefore data point 1000 was taken at time = 500 seconds	36
Figure 15. Soil sample collection locations in fields under study	43
Figure 16. Wind rose of wind speed and direction measured during  the May-June 2008 campaign
	44
Figure 17. Cup anemometer measurements shown with the wind speed profiled calculated using
the measured wind speed	46
Figure 18. Contour plot of measured PMio concentration for the June 25 sample period	47
Figure 19. PM2.s OC/EC time series concentrations as collected at the downwind AQT location.
The shaded sections indicate the observed agricultural practices. It should also be noted that the
SDL/08-556                      California 2008 Tillage Campaign                             iii

-------
raw instrument OC concentrations have been multiplied by 1.7 to account for potential non-
carbon functional groups	49
Figure 20. PM2.5 organic matter and EC concentrations during specific sampling periods (parallel
to filter-based sampling)	50
Figure 21. PM2.5 organic matter and elemental carbon concentrations during specific sampling
periods (parallel to filter-based sampling)	51
Figure 22. Soluble ionic mass concentrations of AQT downwind PM2.5 filters	52
Figure 23. Average soluble ionic mass percentage composition of the AQT downwind PM2.5
filters. The error bars represent the 95% confidence interval	53
Figure 24. Average compositional mass percentage of the AQT downwind PM2.5 filters	53
Figure 25. Average chemical composition of particles detected by the AMS from May 14-15... 54
Figure 26. Representative AMS mass spectrum of particles detected during the study. Mass-to-
charge (m/z) ion assignments include nitrate (m/z 30 NO+ and m/z 46 NC>2+), sulfate (m/z 48 SO+,
64 SO2+, 81 HSO3+, and 98 H2SO4+), and carbon (e.g. m/z 55 C4H7+, 57 C4H9+, 77 C6H5+, 91
C7H7+)	54
Figure 27. (a) Image plot of nitrate particles on May 14, 2008 using m/z 30 (NO+). This shows
the formation of a mode of nitrate particles in the early morning hours of 5/14/2008, with peak
mass concentration at -3:00 AM Pacific Standard Time, (b) Integrated size distribution of nitrate
particles from 2:00-4:00 AM on May 14, 2008  showing the peak in the mass distribution at
-0.65 |im	55
Figure 28. Comparison of AMS PMi and OPC  PMi (assuming a MCF of 1.0 g/cm3) data for the
morning of May 15,2008	56
Figure 29. Particle volume size distributions measured from upwind (background) and
downwind (background plus emissions) locations, with the difference being the aerosol emitted
by the tillage activity: (a) strip-till operation, conservation tillage method; (b) disc 2 operation,
conventional tillage method, (c) plant operation, conservation tillage method; and (d) plant
operation, conventional tillage method	58
Figure 30. OPC PM time series, created by multiplying the volume concentrations (Vk) by the
daily MCF, as measured at elevated locations in a) an upwind and b) a downwind positions
during the May  19 R2  sample period	59
Figure 31. Contour plots of average  OPC a) number concentration (#/L) for particles larger than
1 jam  and b) PMio concentration (|ag/m3) across the field for the third cultivator pass on June 25.
	60
Figure 32. Average daily measured MCF values with error bars representing the 95% confidence
intervals	61
Figure 33. Sample period MCF values compared with the length of each sample period	63
Figure 34. PM2.5, PMio, and TSP mass concentrations retrieved from collocated lidar and OPC
during the 'stare' time series for 6/18. Data acquisition time of the lidar data is 0.5 s while OPCs
were set to 20 s  accumulation time. Measurements were done on the upwind side of facility
(location Tl)	64
SDL/08-556                      California 2008 Tillage Campaign                              iv

-------
Figure 35. PM2.5, PMio, and TSP mass concentrations retrieved from collocated lidar and OPC
during 'stares' at downwind locations. Data acquisition time of the lidar data point is 0.5 sec
while OPCs were set to 20 sec accumulation time. Measurements were done on the downwind
side of field (location T2) on 6/18/2008	65
Figure 36. Wind speed, wind direction, upwind and downwind plume area average particulate
volume concentrations, for the May 17, 2008 strip-till pass of the conservation tillage method. 68
Figure 37. Wind speed, wind direction, upwind and downwind plume area averaged particulate
volume concentrations for the May 19, 2008 first disc pass of the conventional tillage operation.
	68

Figure 38. Lidar-derived fluxes (|ig/m2/s) of PM2.5, PMio, and TSP for the May 17, 2008 strip-till
pass of the conservation tillage method	69
Figure 39. Lidar derived fluxes (|ig/m2/s) of PM2.5, PMio, and TSP for the May 19, 2008 first
disc pass of the conventional tillage operation	70

Figure 40. ISCST3-modeled results for the third cultivator pass of the conventional tillage
operations on June 25, 2008 with northwest winds. The area of operations and sampler locations
are denoted in black and contour line numerical values are in |ig/m3	75
Figure 41. AERMOD modeled results for third cultivator pass of the conventional tillage
operations on June 25, 2008 with northwest winds. The area of operations and sampler locations
are denoted in black and contour line numerical values are in |ig/m3	76

Figure 42. A comparison of PM2.5, PMio, and TSP emission rates ± 95% confidence intervals
derived through lidar scanning techniques and inverse modeling using ISCST3 and AEJAMOD
dispersion models and measured PM concentrations	83
Figure 43. Four adjacent microscopic images taken with the lOx lens that have been combined
for determining the particle size distribution and count at this location on a PMio sample filter.
The largest particle with dpA   31.94 jam is shown by the blue circle	Ill
Figure 44. Number distributions based on projected area diameter as measured via microscopy
for two downwind and one upwind PM2.5 filters  and two downwind PMio filters used during the
Lister pass on May 20, 2008	112

Figure 45. Line graphs showing the number of filters in each ring particle density category for
each sample period	117
Figure 46. Categorization of filter samples to determine suitability for use in emission rate
calculations in inverse modeling	119
SDL/08-556                      California 2008 Tillage Campaign

-------
                                  LIST OF TABLES
Table 1. Emission rates ± 95% confidence intervals found by SDL [7] using lidar measurements
for a conventional fall tillage sequence and a Combined Operations CMP sequence	4
Table 2. Emission factors and uncertainties for land preparation as reported by Flocchini et al.
(2001) [12]	6
Table 3. Emission factors used by the California Air Resources Board in estimating agricultural
tilling PMio emissions [13]	6
Table 4. Conventional and conservation tillage emission rates reported by Madden et al. (2008)
for tillage in a dairy forage crop rotation [16]. ST= standard tillage method, CT = conservation
tillage method	7
Table 5. Estimated distribution parameters and goodness-of-fit, as root mean square error
(RMSE), for the lognormal and Weibull distributions fit to tillage operation datasets with > 8
data points. Values in bold represent model with a better fit	11
Table 6. Emissions factor values corresponding to statistical measures of interest for the
distributions fitted to the various tillage operation datasets	11
Table 7. Tillage operations and dates performed for the comparison study	16
Table 8. Agricultural equipment used to perform the tillage operations	17
Table 9. Operation data for both the conventional and conservation tillage studies as recorded by
field personnel	18
Table 10. Sample period, total tractor operation time, and the sample period-to-tractor operation
time ratio for all sample periods	19
Table 11. Summary of instruments located at each position for the conventional tillage study of
Field 4. All heights given as above ground level (agl)	22
Table 12. Summary of instruments located at each position for the Conservation tillage study of
Field 5. All heights given as above ground level (agl)	24
Table 13. Statistics of soil characteristics measured for both fields	41
Table 14. Stable aggregate analysis results for both fields	42
Table 15. Average ± la temperature, relative humidity, wind speed, and wind direction for each
sample period as measured at the WM tower. Temperature, relative humidity, and wind speed
were measured at 9.7 m agl and wind direction was measured at 11.3 m agl	45
Table 16. PM2.5 filter ion concentrations averaged over seven downwind samples collected at
AQT (uncertainty represents the 95% confidence interval)	52
Table 17. Mass  conversion factors (g/cm3) used to convert  optical particle measurements to mass
concentrations for each day and averaged for the whole campaign. Error values represent the
95% confidence interval for ri>3	62
Table 18. Comparison of PM mass concentrations (|ig/m3) as reported by MiniVol samplers and
mean values measured by collocated OPCs and lidar at Tl  (upwind) and T2 (downwind) for
6/18/2008	64
SDL/08-556                      California 2008 Tillage Campaign                              vi

-------
Table 19. Total number of upwind and downwind vertical lidar scans and the number of those
scans determined to be valid for emission rate calculations. No lidar data exists for the 6/25 run
because of instrument problems after the 6/18 run	66
Table 20. Mean fluxes (|ig/m2/s) ± 95% confidence interval from quality controlled samples for
each tillage operation	71
Table 21. Aerosol mass transfer (± 95% confidence interval) from each field (flux normalized by
operation duration and area tilled) as calculated from lidar data for all tillage operations	72
Table 22. Tillage operations, number of passes, area tilled per pass during monitoring period, and
the seed emission rate used for modeling particle dispersion using ISCST3 and AERMOD for
each sample period	74
Table 23. Mean emission rates per unit  area per unit time (± 95% CI for n > 3) for each operation
as determined by inverse modeling using ISCST3	78
Table 24. Mean emission rates per unit  area (± 95% CI for n > 3) for each operation as
determined by inverse modeling using ISCST3	79
Table 25. Mean emission rates per unit  area per unit time (± 95% CI for n > 3) for each operation
as calculated by inverse modeling using AERMOD	80
Table 26. Mean emission rates per unit  area (± 95% CI for n > 3) for each operation as calculated
by inverse modeling using AERMOD	81
Table 27. Calculated PMio emission rates (± 95% confidence interval) from the lidar and inverse
modeling using two dispersion models	84
Table 28. A comparison of PMio emission rates herein derived and found in literature	87
Table 29. Particulate emissions, from Lidar data and inverse modeling with OPC data, and tillage
rate comparison between conventional and conservation tillage	88
Table 30. Settings for the ISCST3 and AERMOD dispersion models for the tillage study in the
ISC-AERMOD View software by Lakes Environmental, Inc. All settings were held constant
across the sample periods except the source area size and shape, which changed each day, and
the downwind receptor locations, which were specific to each field studies. (— = not applicable)
	99

Table 31. Calculated PM concentrations (ng/m3) measured during May 2008 at all sample
locations	101
Table 32. Calculated PM concentrations (|ag/m3) measured during June 2008 at all sample
locations	102
Table 33. PM concentrations (ng/m3) used in emission rate calculations from sample periods in
May 2008	103

Table 34. PM concentrations (ng/m3) used in emission rate calculations from sample periods in
June 2008	104

Table 35. PM concentrations (ng/m3) as measured by OPCs from sample periods in May 2008.
                       	105
SDL/08-556                      California 2008 Tillage Campaign                             vii

-------
Table 36. PM concentrations (ng/m3) as measured by OPCs from sample periods in June 2008.
                                                                                    106
Table 37. PM concentrations (ng/m3) as measured by OPCs used in emission rate calculations
from sample periods in May 2008. (OPC PM,t = V* x MCF,t) [[[ 107

Table 38. PM concentrations (ng/m3) as measured by OPCs used in emission rate calculations
from sample periods in June 2008. (OPC PM,t = V* x MCF,t) [[[ 108
Table 39. Size fractionated results of the visual inspection of annular filter rings ..................... 115
Table 40. Results of the visual inspection  of filter rings by sample date ................................... 116
Table 41. Sample period filter datasets used in calculating emissions, with reasons for why some

-------
              TABLE OF ABBREVIATIONS/VARIABLES
Abbr./Var.
PE
Pv
n
ng
HL
pirn
Pv
o
Ow
°v
Oz
2D
3D
,4
ACS
Activity
AERMOD
agl
AMS
AQT
ARS
Avg
AvgHp
bhp
CIQ
CD
n
^ measured
n
^ modeled
Co
c
CE
CFR
/^ +2
Ca
CARB
CCF
CCV
CF
CHrs
CI
CIMIS
cr
CMP
CO2
CT
d
Definition
backscatter values measured by lidar in vector form (1/m-steradians)
particle normalized backscatter from OPC data in vector form ( 1/m-steradians)
control efficiency
micrograms (IxlCT6 gram)
microliter (IxlCT6 liter)
micrometers (IxlCT6 meter)
density of water vapor
standard deviation
standard deviation of the vertical wind (m/s)
horizontal plume spread parameter (meter)
vertical plume spread parameter (meter)
two-dimensional
three-dimensional
acres of land tilled
American Chemical Society
amount of time the engine is active during the year (hr/yr)
American Meteorological Society/Environmental Protection Agency Regulatory Model
above ground level
aerosol mass spectrometer
air quality trailer
Agricultural Research Service
average
maximum rated average horsepower (bhp)
brake horsepower
10 minute average pollutant concentration (ng/m3)
downwind PMK concentration (ng/m3)
measured PM concentration (ng/m3)
modeled concentration (ng/m3)
scan area average upwind PMK concentration (ng/m3)
Carbon dioxide concentration
constant, 4.8 Ib/acre-pass
Code of Federal Regulations
Calcium ion
State of California Air Resources Board
counting correction factor (unitless)
continuing calibration verification
conversion factor from (g/yr) to (tons/day)
cumulative engine operation hours (hr)
confidence interval
California Irrigation Management Information System
Chloride ion
conservation management practice
Carbon dioxide
conservation tillage
particle diameter (\m\)
SDL/08-556
California 2008 Tillage Campaign
                                                                 IX

-------
Abbr./Var.
daero
QZ, upper
QZ, /ower
4
dpA
DDI
DF
dr
E
E
ECT
f
^estimated
EPM
Tf
l-'seed
ESTT
EC
EC
EC/OC
EF
EFacjj(pM)
FF
L-'i ss
Emissions
EPA
F
F
FRM
GMD
GPS
H
H
H2SO4
HC1
hp
lirtractor
1C
IRGA
ISCST3
k
r
LE
Load
Lpm
M
m
MCF
MDL
Mg+2
Definition
aerodynamic diameter (nm)
upper edge of bin / (nm)
lower edge of bin / (nm)
particle of diameter k (\m\)
projected area diameter (\m\)
double-distilled, de-ionized water
engine deterioration factor (unitless)
engine deterioration rate (g/bhp-hr2)
east
flux of water vapor from the soil surface
emission rate for the conventional tillage method
derived emission rate (ng/s-m2)
PM emission (Ibs)
seed emission rate (ng/s-m2)
emission rate for the strip-till tillage method
elemental Carbon particulate matter
eddy covariance
elemental Carbon/organic Carbon particulate sampler
emission factor (g/bhp-hr)
adjusted PM emission factor (g/hp-hr)
steady-state measured emission factor (g/hp-hr)
amount of pollutant released (tons/day)
United States Environmental Protection Agency
particulate flux in lidar scans (ng/s)
Fluoride ion
federal reference method
geometric mean diameter (nm)
global positioning system
sensible heat
effective stack height (meter)
Hydrogen sulfate
Hydrogen chloride
horsepower
tractor operation hour
ion chromatography
infrared gas analyzers
Industrial Source Complex, Short Term Ver. 3
dimensionless particle size multiplier
Potassium ion
latent heat
load factor
liters per minute
molar
meter
mass conversion factor (g/cm3)
minimum detection level
Magnesium ion
SDL/08-556
California 2008 Tillage Campaign

-------
Abbr./Var.
N
N,,
Nt
n
n(r)
Na+
NAAQS
NaOH
NC
NEVES
ng
NH4+
NO2"
NO3
NOX
NONROAD
NRCS
OC
OFFROAD
OPC
P
Pii
PM
PM,0
PM25
PM10
PM*
PMK(r)
ppm
PSL
Q
1
to
RARE
RD
S
S
SDL
SO42
SpMadj
ST
t,
tp
TAP
TD
TSP
Definition
north
average number concentration per bin /' for OPC j (number/cm3)
mean of the averages across all OPCs for bin / (number/cm3)
number of samples
relative amplitude of the aerosol component of total atmospheric backscatter at range r
(unitless)
Sodium ion
National Ambient Air Quality Standard
Sodium hydroxide
not calculated
Nonroad Engine and Vehicle Emission Study
nanogram (IxlCT9 gram)
Ammonium ion
Nitrite ion
Nitrate ion
Oxides of nitrogen
U.S. EPA Nonroad Emissions Model
Natural Resources Conservation Service
Organic Carbon
CARB Offroad Emissions Model
optical particle counter
power law wind speed coefficient (unitless)
raw OPC particle counts in bin / (number)
particulate matter
particulate matter with an aerodynamic diameter less than or equal to 1.0 micrometers
particulate matter with an aerodynamic diameter less than or equal to 2.5 micrometers
particulate matter with an aerodynamic diameter less than or equal to 10 micrometers
particulate matter with an aerodynamic diameter less than or equal to k micrometers
particulate matter with an aerodynamic diameter less than or equal to K micrometers,
calculated from lidar data, at range r
parts per million
polystyrene latex sphere
emission rate (ng/s)
water vapor concentration
average measured sample flow for OPC j (cm3/min)
Regional Applied Research Effort
respirable dust
south
silt content of surface soil (%)
Space Dynamics Laboratory
Sulfate ion
diesel sulfur content adjustment factor (g/hp-hr)
standard tillage
sample time for OPC j (min)
number of passes or tillings per year
transient adjustment factor (unitless)
total dust
total suspended particulate
SDL/08-556
California 2008 Tillage Campaign
                                                                                                  XI

-------
Abbr./Var.
M
U*
Ui
U2
UGM
u w
UC
USDA
UWRL
v£
v,
VK(r)
V
^(r,h)
vw
W
W
W
w
wss
y
z
Zl
Z2
ZH
Definition
wind speed in the streamwise direction (m/s)
friction velocity (m/s)
Measured reference wind speed (m/s)
Calculated wind speed based on the reference wind speed (m/s)
mean wind speed at the release height H (m/s)
momentum term of the streamwise and vertical wind directions (m2/s2)
University of California
United States Department of Agriculture
Utah Water Research Laboratory
particle normalized volume concentration vector (nm3/cm3)
OPC cumulative volume concentration up to a diameter of k (nmVcm3)
cumulative volume concentration from lidar data at range r (nm3/cm3)
wind speed in the lateral direction (m/s)
average wind speed component at range r and height h that is parallel to the long axis of
the lidar staple box
momentum term of the lateral and vertical wind directions (m2/s2)
west
watt
diagonal weighting matrix used in lidar data analysis (unitless)
wind speed in the vertical direction (m/s)
web soil survey
horizontal distance of the modeled receptor from the centerline of the plume (meter)
height of modeled receptor above ground level (meter)
height at which the reference wind speed was measured (m)
height of the calculated wind speed (m)
zero-hour engine emission rate (g/bhp-hr)
SDL/08-556
California 2008 Tillage Campaign
                                                                                                 xn

-------
The statements and conclusions in this report are those of SDL and not necessarily those of the
California Air Resources Board, the San Joaquin Valleywide Air Pollution Study Agency, or its
Policy Committee, their employees or their members. The mention of commercial products, their
source, or their use in connection with material reported herein is not to be construed as actual or
implied endorsement of such products.
SDL/08-556                      California 2008 Tillage Campaign                             xiii

-------
                              EXECUTIVE SUMMARY
Airborne particles, especially particulate matter 10 micrometers (um) or smaller in aerodynamic
diameter (PMio) and fine participate matter 2.5 jim or smaller in aerodynamic diameter (PM^.s),
are microscopic solids or liquid droplets in the air that can cause serious health problems (e.g.,
coughing, difficulty  breathing,  decreased  lung function,  asthma, heart attacks,  and premature
death), especially in people with heart and lung disease.  Concern with health effects resulting
from  PMio  exposure is drawing  increased regulatory  scrutiny and research toward  local
agricultural tillage operations. To  investigate the  control effectiveness of one of the  current
Conservation Management Practices (CMPs) written for agricultural land preparation on the
generation of particulate matter (PM) levels, the San Joaquin Valleywide Air Pollution Study
Agency funded a project to  study:
       1)       The magnitude, flux, and transport of PM emissions produced  by agricultural
                practices for row crops where tillage CMPs are implemented vs. the magnitude,
                flux, and transport of PM emissions produced by agricultural practices where
                CMPs are not implemented.
       2)       The control efficiencies  of equipment used to implement the  "conservation
                tillage"  CMP. If resources allow assessing  additional  CMPs, what  are the
                control  efficiencies of the "equipment  change/technological improvements"
                CMP?
       3)       If these CMPs for a specific crop can be quantitatively compared, controlling
                for soil type, soil moisture, and meteorological conditions.
This study used advanced measurement technologies, which link lidar systems with conventional
point-measurement air quality samplers, to  map  PM emissions at high spatial and temporal
resolution in order to accurately compare CMPs with conventional tillage systems. The purpose
of this field study was to determine if and how much particulate emissions differ between the
conventional  method of agricultural  fall  tillage and a  conservation tillage CMP.   It is a
companion study  to an  earlier  study performed in  October  2007 near Los Banos,  CA
investigating the control efficiency of combined operations CMP versus conventional tillage in
the fall tillage sequence.  Findings from that study are detailed in a previous report [7].
The test location  and CMP to be evaluated were chosen  in discussion with stakeholders,
regulatory agencies, and researchers. The current study was performed in the San  Joaquin Valley
of California in May and June of 2008 during the spring tillage sequence between a winter wheat
crop and corn.  The conventional 13-pass spring tillage  sequence for  a field going from winter
wheat into corn was: two passes for in-field irrigation border breakdown, chisel, two disc passes,
lister, two cultivator passes, roll, plant, fertilizer injection, and two more cultivator passes. The
spring conservation  tillage  CMP  sequence  consisted of only three  tractor passes: strip-till,
plant/fertilize, and herbicide spray; specifically, the CMP  implement investigated was the strip-
till implement.
An extensive network of measurement systems were used during this study, including a scanning
lidar,  a full meteorology suite, four sonic  anemometers (for turbulence information),  and filter
and optical aerosol  point samplers.  Two  additional aerosol  chemical analysis  systems were
employed from a sampling trailer located on the downwind side of the field under test.
Tillage particulate emission rates  were determined using two methods:  1) inverse  modeling
coupled with observed facility-derived concentrations from filter- and  optical-based instruments,
SDL/08-556                      California 2008 Tillage Campaign

-------
and 2) a mass balance approach applied to upwind and downwind PM concentrations measured
by the lidar. The tillage emissions were modeled using two different air dispersion models:  the
Industrial   Source Complex Short-Term  Model,  version 3  (ISCST3) and the  American
Meteorological Society/Environmental  Protection  Agency Regulatory  Model  (AERMOD).
Emission data calculated for each measurement method for the conventional and conservation
tillage operations  are presented herein. The  study showed that the conservation practice required
< 1/4 of the number of tractor passes when compared to conventional tillage;  similar reductions
in fuel use and tractor exhaust associated PMio emissions were expected to have occurred.
Lidar-derived and inverse  modeling emission rates  for  PM2.s, PMio,  and total suspended
particulate  (TSP) by operation,  as  well as the average tillage rate in hours per hectare are
summarized herein. Based on lidar data, the conservation tillage method reduced PM2.5 emission
by 91%, PMio by 94%, and TSP by 91%, which were all statistically significant differences.
Reduced emissions as  calculated using inverse modeling  and  optical  particle counter data are
very  close to lidar-derived reductions  at  85%,  87%,  and 90% for  PM2.5, PMio, and TSP,
respectively.  The time per hectare required to perform the conservation tillage was  about 14% of
the conventional method.  The control efficiency of the Conservation Management Practice for
particulate  emissions was 0.905, 0.937, and 0.909 for PM2.s, PMio, and  TSP, respectively, based
on lidar data and 0.853, 0.872,  and 0.903 for PM2.5,  PMio, and TSP, respectively, based on
inverse modeling with optical particle counter data.
Some of the  derived PMio emission rates from this experiment agree with those reported in the
literature, as  well as  the emission factors used by the State of California Air Resources Board
(CARB) to calculate  area source PMio contributions from agricultural tilling [14];  others do not
agree and are significantly higher, such as the emission rates for the chisel, disc 1,  disc 2, and
lister passes. Emissions factor values estimated through inverse modeling for these operations
are below the 95% level predicted by statistical distributions fitted to published data; emissions
estimates from lidar for the same  operations  are  above the 95% level. While values from
published studies are generally not in close agreement, they are within the range of the variability
expected from measurements made  under different  meteorological   and  soil conditions,  as
demonstrated by the wide range of values in the literature [12][16]. In general,  the emission rates
calculated from this study are higher than others reported but are within the variability of other
reported emission rates.
SDL/08-556                      California 2008 Tillage Campaign

-------
                               1.     BACKGROUND
Airborne particles, especially parti culate matter 10 micrometers (um) or smaller in aerodynamic
diameter (PMio) and fine participate matter 2.5 jim or smaller in aerodynamic diameter (PM^.s),
are microscopic solids or liquid droplets in the air that can cause serious health problems (e.g.,
coughing or difficulty breathing, decreased lung function, aggravated asthma, development of
chronic bronchitis, irregular heartbeat, heart attacks, and premature death), especially in people
with heart or lung disease [3]. Particles larger than 1.0 um tend to be prevented from entering the
lungs by the nose and throat [4]. The U.S. Environmental Protection Agency (U.S. EPA) has
established limits for PIVb.s and PMio levels in order  to protect public health as  part of the
National Ambient Air  Quality Standards (NAAQS) [5][6].  The  U.S. EPA requires state air
quality  management agencies  to monitor ambient PM2.5 and PMio concentrations  in order to
identify possibly hazardous conditions for  the population; to report areas  that  exceed the
NAAQS beyond the allowed number of times; and to establish  procedures to reduce particulate
concentrations to meet the standards.
To address the problems associated with exposure to high particulate matter  (PM) levels, the
U.S. EPA has  been working  with the  San  Joaquin Valley  Air Pollution Control District to
support PM research. In 2007 the Environmental Sciences Division, National Exposure Research
Laboratory was awarded a Regional Applied Research Effort  (RARE) grant to determine the
control  effectiveness of a Conservation Management Practice (CMP) for agricultural  tillage,
targeted toward reducing PMio emissions, using advanced measurement technologies such as
atmospheric light  detection and ranging (lidar) systems. The  field  study was  carried out in
October 2007 in the San Joaquin Valley [7]. This report describes the 2008 tillage comparison
project that was supported by the San Joaquin Valley wide Air Pollution Study Agency to follow-
up the 2007 RARE project. The lidar system, when coupled with point-measurement air quality
samplers, can map PM emissions at high spatial and temporal resolutions, allowing for accurate
comparisons of various  CMPs for a variety of agricultural practices  [8]. The purpose of this
project was to deploy an elastic lidar system, together with a network of air samplers, to measure
PM emissions from agricultural operations in order to answer the following research questions:

    1.  What are the magnitude,  flux, and transport of PM emissions produced by  agricultural
       practices for row crops  where tillage CMPs are implemented vs. the magnitude, flux, and
       transport of PM  emissions  produced by  agricultural practices  where CMPs  are not
       implemented?

    2.  What are the control efficiencies of implementing the "conservation tillage" CMP?

    3.  Can this CMP for a specific crop be  quantitatively compared, controlling for soil type,
       soil moisture, and meteorological conditions?
In November 2008, EPA redesignated the San Joaquin  Valley  to  attainment for the
NAAQS, but sources in the Valley must continue to implement measures that helped the District
attain the PMio standard, including CMPs. The  Valley continues to violate the PM2.s NAAQS.
The  CMP  chosen for comparison against  the  conventional tillage method  was conservation
tillage, which is defined as a method in which  the soil is  being tilled or cultivated to a lesser
extent compared to a conventional system [9].  There are several different conservation tillage
systems that vary in the amount of the soil tilled; the system chosen for examination in this study
SDL/08-556                      California 2008 Tillage Campaign

-------
was the strip-till system. Data collected May 17 - June 25, 2008 at a site in the San Joaquin
Valley of California are included in this report.

1.1    LITERATURE REVIEW
A handful  of published articles pertain  to  PM emissions from agricultural tillage, with the
majority of the studies performed in the State of California. There are also several papers that
collectively examine impacts of a variety of conservation tillage practices with respect to soil
characteristics, fuel consumption, cost of production, and air emissions.
The companion study to this experiment  was conducted in October 2007 near Los Banos, CA
[7]. It  examined differences between conventional  fall  tillage practices  and a  Combined
Operations CMP.  This involved using an implement designed to  do the work of multiple
implements in one pass called the Optimizer1 on two fields that were adjacent and similar to one
another.  The conventional method consisted  of four different passes over the field in the
following order: 1)  disc pass, 2) chisel pass, 3) another disc pass, and 4) land plane pass. The
CMP  was applied in two steps in the following order:  1) chisel pass, and 2) the Optimizer pass.
This study  found that the CMP used 51% less fuel per unit area and took 62% less time per unit
area. Using a lidar mass balance approach, it was determined that the CMP produced 71%,  40%,
and 76% as  much  PM2.s, PMio, and  total  suspended paniculate (TSP)  as the conventional
method. Table 1 presents the lidar-based PM2.5, PMio, and TSP emissions rates for each tillage
step. Comparisons using inverse modeling coupled with PM measurements were not complete
due to insufficient differences between upwind and downwind PM2.5 concentrations during  some
measurement periods; however,  operation-specific  emission rates  calculated using  inverse
modeling were similar to those reported by the calibrated lidar measurements.
Table 1. Emission rates ± 95% confidence intervals found by SDL [7] using lidar measurements for a
conventional fall tillage sequence and a Combined Operations CMP sequence.
Operation
Combined: Chisel
Combined: Optimizer
Sum for Combined
Operation Method
PM25
(mg/m2)
45. 3 ± 13.1
32.5 ±5.1
77.8 ± 14.0
PM10
(mg/m2)
69.0 ± 19.9
42.7 ±6.6
111.6 ±20.9
TSP
(mg/m2)
265.9 ±76.6
169.9 ±26.2
435.8 ± 80.9

Conventional: Disc 1
Conventional: Chisel
Conventional: Disc 2
Conventional: Land plane
Sum for Conventional
Method
20.4 ±2.6
35.8 ±5.9
39.5 ±11.2
13.8 ±3.9
109.5 ± 13.5
99.7 ± 12.5
79.5 ± 13.1
80.7 ± 20.5
21. 9 ±6.2
281.9 ± 28.0
159.8 ±20.0
235.1 ±38.8
149.3 ±40.3
33.4 ±9.4
577.6 ± 60.1
The  use of an elastic lidar system by the University of California at Davis (UC Davis) to
examine dust plumes resulting from  tillage  activities was  presented  by Holmen  et al.[8].
Qualitatively,  the constructed system was able to track the  plume emitted from the  moving
source and provide a 2D vertical, downwind map of the plume. It was observed that the plume
1 Mention of a specific tradename or manufacturer does not imply endorsement or preferential treatment by the
USD A-ARS or Space Dynamics Laboratory or Utah State University
SDL/08-556
California 2008 Tillage Campaign

-------
 heights were often above the point samplers located at 10 m along the downwind plane. The
authors suggested that the best fugitive dust sampling procedures would include a combination
of elastic lidar and strategically placed point samplers.
Two papers by Holmen et al., both published in 2001, further discuss tillage PMio emission rate
investigations by UC Davis using filter-based mass  concentration samplers and qualitative
measurements from the previously mentioned  elastic  lidar system [10][11]. The 24 samples
labeled as valid were collected from Fall 1996 to Winter 1998 in the San Joaquin Valley during a
wide range of environmental  (temperature   7-35 °C, relative humidity   20-90%, and from
prior to the season's first precipitation to periods between winter storms) and soil moisture
conditions (1.5-20%). Tillage operations examined were discing,  listing, root  cutting, and
ripping. Calculated PMio emission rates ranged from 0 to 800 mg/m  (0 to 6.9 Ib/acre), the mean
± one standard deviation was 152 ± 240 mg/m (1.4 ±2.1 Ib/acre), and the median was 43  mg/m
(0.4 Ib/acre). One point made by Holmen et al. [11] is that several environmental conditions
(e.g., temperature profile, relative humidity, soil moisture) can significantly affect PM emissions
and should be monitored and accounted for in  emission rate  measurement and reporting. As a
result,  the reliability of direct  comparisons  of emission  rates  measured  under different
environmental conditions must be carefully examined.
The studies published by researchers at UC Davis and herein previously discussed were part of a
much larger investigation of  agricultural  PMio emission  rates in the San Joaquin Valley  as
funded by the U.S. Department of Agriculture Special Research Grant Program. Findings of this
broad study are published in  Flocchini et al. (2001)  [12]. Table  2 presents emission  factors
published in this study for different types of agricultural tillage along with the crop and time of
year. As seen in results measured by Holmen et al., the emission rates reported by Flocchini et al.
for agricultural tillage were influenced more by environmental conditions, such as the near-
ground temperature profile, relative humidity,  and soil moisture, than by the type of crop  or
equipment used for tilling [11][12].
The  California  Air Resources Board  (CARB)  developed  area  source emission inventory
calculation methodologies for  agricultural  tillage and harvesting operations based on the report
by Flocchini et al. [12][13][14]. A summary of the resulting emission factors appears in Table 3.
The given unit for the emission factor can be explained as the mass of PMio particles released
per acre for each operation, or pass. Specific tillage operations were assigned to one of these five
categories and the displayed emission factor was used for all operations in each category.
The U.S. EPA (2001)  uses the empirically derived equation shown below to estimate the quantity
of particulate matter emitted from all agricultural tilling  processes [15].

                                 E  cExkxs°'6 xpxa                               (1)

where E   PM emission in  Ibs, CE    constant 4.8 Ib/acre-pass, k   dimensionless particle size
multiplier (TSP = 1.0, PMio  0.21, PM2.5   0.042), s   silt content of surface soil (%),/? =
number of passes or tillage operations in a year, and a  acres of land tilled. The above equation
was developed to estimate TSP emissions  (k    1.0) and has since been scaled to estimate PMio
and PM2.5 emissions by using the respective k value. Average values of s are tabulated in Table
4.8-6 in U.S. EPA (2001) as a function of soil type on the soil texture classification triangle [15].
Based on the k values, PM released by tillage operations is dominated by large particles (>  10
jam).
SDL/08-556                      California 2008 Tillage Campaign

-------
Table 2. Emission factors and uncertainties for land preparation as reported by Flocchini et al. (2001) [12].
Date
Emission Factor
(mg/m2)
Uncertainty

Date
Emission Factor
(mg/m2)
Uncertainty
Stubble Disc
10/27/1995
11/3/1995
11/3/1995
11/3/1995
11/3/1995
11/3/1995
11/3/1995
11/15/1995
11/15/1995
6/24/1997
257.7
49.3
27.4
231.0
136.7
140.8
286.1
537.9
542.2
430.0
NC
9%
470%
4%
7%
6%
5%
9%
125%
17%










11/6/1998
11/6/1998
11/6/1998
11/6/1998
11/6/1998
11/6/1998
11/6/1998
11/6/1998
11/6/1998

50.0
28.4
35.0
28.0
117.0
32.4
58.9
93.5
74.2

146%
145%
NC
10%
18%
9%
8%
9%
8%

Finish disc
11/26/1996
11/26/1996
11/26/1996
12/2/1996
124.3
142.4
97.5
91.0
3%
4%
5%
9%




12/4/1996
12/4/1996
12/4/1996
12/5/1996
9.2
0.6
3.5
-0.5
NC
NC
NC
NC
Ripping/chisel
6/24/1997
6/26/1997
6/26/1997
765.0
112.0
776.0
5%
5%
3%



6/25/1997
6/25/1997

331.0
577.0

5%
6%

Root cutting
11/16/1996
30.0
12%

11/16/1996
36.0
8%
Table 3. Emission factors used by the California Air Resources Board in estimating agricultural tilling PM10
emissions [13].
Agricultural Tilling Operation
Root cutting
Discing, Tilling, Chiseling
Ripping, Subsoiling
Land Planing Floating
Weeding
Emission Factor
(Ib PM10/acre-pass)
0.3
1.2
4.6
12.5
0.8
(mg PM10/m2-pass)
33.6
134.5
515.6
1401.0
89.7
A comparison between standard tilling practices and  conservation tilling  (strip-till) in dairy
forage production on two farms in the San Joaquin Valley is given by Madden et al. (2008) [16].
Both strip-till and standard till  operations were monitored for PMio emissions  over two tillage
cycles at both farms.  Results  show that conservation tillage practices reduce  PMio  emissions
from one  farm  by 86%  and  52%  for 2004 and  2005,  respectively.  At the  second farm,
conservation tillage emissions were reduced by 85% and 93% for 2004 and 2005, respectively.
Derived emission rates are presented in Table 4. Madden et al. attribute these reductions, in part,
to a reduced total number of passes (from 3-6 passes in standard tillage to 1 pass in conservation
tillage) and the ability for conservation tillage operations to be done under a higher soil moisture
content than standard operations.
SDL/08-556
California 2008 Tillage Campaign

-------
Table 4. Conventional and conservation tillage emission rates reported by Madden et al. (2008) for tillage in a
dairy forage crop rotation [16]. ST= standard tillage method, CT = conservation tillage method.
Season/Year

Spring 2004
Sweet Haven Dairy
Operation
ST: 1st discing
ST: 2nd discing (w/ roller)
ST: 3rd discing (w/roller)
ST: Planting
CT: Strip-tilling
CT: Planting
Avg Emission
Factor (mg/m2)
198
1035
114
103
181
26
Barcellos Farms
Operation
ST: 1st discing
ST: 2nd discing
ST: Listing
ST: Bed discing
ST: Bed mulching
ST: Ring roller
ST: Planting
ST: Ring roller
CT: Planting
Avg Emission
Factor (mg/m2)
259
917
615
25
89
566
96
104
394

Spring 2005
ST: 1st discing
ST: 2nd discing (w/ roller)
ST: 3rd discing (w/roller)
ST: Planting
CT: Strip-tilling
CT: Planting
139
375
404
263
180
385
ST: 1st discing
ST: 2nd discing
ST: Circle narrow
ST: Listing
ST: Bed discing
ST: Bed mulching
ST: Planting
CT: Planting
51
123
337
466
109
384
481
130
Conventional tillage emissions factors were reported by Wang et al. (2010) and Kasumba et al.
(2011) from different field experiments conducted in cotton fields on the same farm in New
Mexico [17][18].  Kasumba et al.  2011 reported emission rates from  six  different  discing
operation  tests, resulting in averages of 154.6 ± 6.9 mg/m2, 78.3 ±  2.9 mg/m2, 238.7 ± 8.8
mg/m2, 89.2  ± 3.7 mg/m2, 8.4  ± 2.5 mg/m2, and 68.4 ±  2.3  mg/m2 where the uncertainty
represents the standard deviation (o) about the average. Wang et al. reported operation average
emissions ± la of 12.1 ± 9.7 mg/m2  for a rolling operation, 210.7 ±  115.3 mg/m2 for listing,
176.7 ± 77.0 mg/m2 for planting,  10.4 ± 6.5 mg/m2 for harvesting, and 44.8 ± 29.3 mg/m2 and
202.8 ± 79.8 mg/m for discing operations during different years. Wang et al. also developed an
empirical relationship relating emissions to soil moisture, silt fraction, crop type, and operation.
Dust concentrations produced by agricultural implements used at a UC Davis research farm were
reported by Clausnitzer and Singer (1996) [19]. Personal exposure samplers measuring respirable
dust (RD) concentrations—particles that may reach  the alveolar region  of the lungs when
breathed in (with  a 50% cut-point diameter of 4 (im)--were mounted on implements in 22
different operations over a seven month period in 1994; replicate samples were collected during
18 operations. Average RD concentrations measured on the implement ranged  from 0.33 mg/m3
for discing corn stubble to  10.3 mg/m3 for both land planing and ripping operations. While RD
concentration was heavily influenced by operation, other factors determined to be significant in
dust production were relative humidity, air temperature, soil moisture,  wind speed, and tractor
speed.
SDL/08-556
California 2008 Tillage Campaign

-------
Further investigation of the data set presented by Clausnitzer and Singer (1996) [19]  and of
another data set collected on a different UC Davis research farm was reported by Clausnitzer and
Singer (2000) [20]. Both sets of data focus on RD concentrations as measured on the agricultural
implement. The analysis examined environmental influences on the measured  concentration.
Again, soil moisture and air temperature were found to be significant factors in RD production.
The RD production with respect to soil moisture was well fit by a power function, with the curve
predicting RD concentrations becoming  significantly steeper below 5%. Air temperature was
hypothesized to be significant in that it was a surrogate measurement of atmospheric instability:
as temperature increases  near the surface, the atmosphere becomes less stable and may carry
greater quantities of dust upwards.
Baker  et al. (2005) examined differences between dust concentrations resulting  from standard
and conservation tillage practices in the San Joaquin Valley over a two year cotton/tomato crop
rotation, each under two  different cover  crop scenarios:  1) no cover crop and 2) a cover crop
forage mixture [21]. Total dust (TD),  particles  < 100 jim in aerodynamic diameter, and RD
samplers were stationed on the implements to collect samples in the plume. For both TD and RD,
the presence or lack of a cover crop in the standard till  treatment  did  not seem to  affect
concentrations. Summed concentrations for conservation tillage without a cover crop were about
one-third of standard tillage; for conservation  tillage with a cover crop, they were about two-
thirds for both dust fractions measured. Reductions in summed concentrations with conservation
tillage  were attributed to fewer operations, including the elimination of the dustiest (discing and
power incorporation).  When comparing  operations common to  all four treatments, tomato
planting and harvesting in conservation  till produced  higher concentrations than  standard till
(thought to be due to increased organic matter on the surface) and concentrations during cotton
harvesting, which does not disturb the soil,  were equivalent for all treatments. This study was
part  of a larger effort to quantify the effects  of conservation tillage in California on crop
production, soil quality, and time and resources dedicated to production as outlined by Mitchell
et al. (2008) and Veenstra et al. (2006) [22][23].
Upadhyaya et  al. (2001)  compared the Incorpramaster, a one-pass tillage instrument against a
conventional combination of discing twice and land planing twice based on fuel consumption,
timeliness, and effect on  soil [24]. Studies on four experimental fields  at UC Davis showed no
statistical difference between  resulting soil conditions  (bulk density changes, soil moisture
changes, and aggregate size), but the Incorpramaster used between 19%  and 81% less fuel with a
mean of 50%.  The time savings ranged from 67% to 83% with a mean of 72%.  In most cases,
two passes with the Incorpramaster were required to achieve the same soil conditions as the four
passes in the conventional till.
Three conservation tillage methods were compared in Mitchell et al. (2006) against the standard
tillage  method in cotton  production  in terms  of yield, yield quality, tractor passes, fuel, and
production [25]. A single field near Fresno, CA was  divided  in area among seven tillage
treatments: 1) standard, 2) no till/chop, 3) no till,  4) ridge till/chop, 5) ridge till, 6) strip-till/chop,
and 7) strip-till. Prior to both cotton growing seasons examined, a small grain wheat was planted
in the field to enhance soil properties; this crop was sprayed with herbicide, and in treatments 2,
4, and  6, it was chopped with a mower prior to tillage activities. In the other treatments (1, 3, 5,
and 7), the dead wheat was either incorporated  by the tilling or left standing. Yield and  yield
quality were statistically the same for both years for all treatments, though the standard treatment
was  numerically higher  in both yield and yield quality the 2nd year.  Conservation tillage
SDL/08-556                      California 2008 Tillage Campaign

-------
treatments reduced tractor passes by 41% to 53% over the standard method and estimated fuel
reduction was 48% to 62% for the conservation practices. The estimated overall production costs
of the conservation  tillage systems were  14% to 18% lower than the conventional system.
Mitchell et al. estimated, by extrapolation from other work, that whole-tillage process paniculate
matter emissions would also be decreased.
Particulate  matter  is released during  agricultural tillage activities from both the  operational
activity of the tillage implement, as well as the tractor in use. Emissions from the tractor mainly
originate from the  tires kicking up dust from the soil  and the combustion  engine. Attempts to
quantify the PM emitted in agricultural tractor exhaust have been made by the U.S. EPA and the
CARB in software  designed to estimate off-road engine emissions on a county or regional scale.
The U.S. EPA developed the NONROAD software program,  with the latest version distributed in
2005  [26]. Emission  factors from compression ignition (diesel) engines used in the NONROAD
model are calculated  by adjusting a zero-hour (ZH), steady-state measured emission factor (EFSS)
for engine deterioration  with  operation time (DF) and a transient adjustment factor (TAP) that
accounts for variations from steady-state engine  loading and speed, as shown in the following
equation [27].

                             EFadj(PM)   EFssxTAFxDF-SPMadj                          (2)

where EFadj(pM) is the adjusted PM emission factor and SpMadj is the emission factor adjustment
accounting for the  use  of a diesel  fuel  with a sulfur  content different than  the default
concentrations;  fuel sulfur level is known to affect PM emissions. The units for EFadj(pM) , EFSS,
and SpMadj are g/hp-hr, where hp stands for horsepower, and TAP and DF are both unitless. All
four variables  on the right side of Eq. 2 vary with model  year and engine size, expressed in
horsepower (hp), according to measured values and/or the emission standards  each  model year
and engine size was  designed to meet. The  selection of values for steady-state emission factors
and all the adjustment variables given in U.S. EPA (2004) [27] was performed using  a variety of
tests and resources, including the Nonroad Engine and Vehicle Emission Study (NEVES) Report
[28],  or by setting the values such that the adjusted PM emissions were equal to model year-
specific emission standards.
The CARB has also  developed a model to forecast and backcast daily exhaust emissions from
off-road engines, including agricultural tractors, herein called OFFROAD.  Similarly to the U.S.
EPA's NONROAD  model,  emission  factors (EF) for each engine size  and model year are
calculated based on a ZH emission rate with a deterioration factor (dr) applied to  account for
engine wear with use, as in Eq. 3. The  derived emission factor is then multiplied by the load
factor (Load),  the maximum  rated average horsepower (AvgHp), and the  amount  of time the
engine is active through the year (Activity) in hr/yr.
                                   EF   ZH+dr*CHrs                                (3)

                       Emissions  EF* AvgHp* Load* Activity *CF                    (4)

where CHrs is the cumulative engine operation hours,  Emissions is the amount of pollutant
released in tons/day, and CF is the conversion factor from units of grams per year to tons per day
and has a value of 3.02 x 10"9. The values for the EF and DR are derived from measured values
or they are calculated based on requirements to  meet  the proposed emissions limits for  future
years  [29].
SDL/08-556                      California 2008 Tillage Campaign

-------
Kean et al. (2000) estimated off-road diesel engine, locomotive, and marine vehicle emissions of
NOX and PMio for 1996 based on fuel sales [30]. Diesel engine exhaust emission factors were
developed based on information provided in the development of the U.S. EPA NONROAD off-
road vehicle emissions model with  supplemental  information in order to  calculate emissions
based on fuel consumption. A fleet-wide average PMio emission factor was determined for farm
diesel equipment to be 3.8 g/kg of fuel used, at an average mass per volume of 0.85  kg/L of
diesel fuel. Fuel sales surveys from 1996 were used to calculate regional and national emissions.
In the off-road category,  which includes farm equipment, the U.S. EPA NONROAD model
calculated on average 2.3 times higher emissions, which was attributed to higher engine activity
assumed in the EPA model than represented in the reported fuel sales data.
An uncertainty  analysis  was conducted  to  determine the statistics of the  preceding  PMio
emissions factors reported  from measurements.  This analysis was performed  following  the
emissions factor dataset analysis  technique  used by RTI International in "Emission Factor
Uncertainty Assessment,  Review Draft"  [31]. Data  points were categorized according to  the
following tillage operations, with the number of values given in parenthesis: chisel (2), disc (67),
land planing (1),  listing (8), ripping (5), root cutting (3), standard tillage planting (11),  strip-till
planting (9), strip-tilling (6), and weeding (15). In  cases where a report/paper only provided an
average emissions factor, the average value was used only once. RTI International found that two
parametric models, the lognormal and Weibull distributions, best fit the analyzed datasets. These
same two parametric models were fitted to the tillage operation datasets with eight or more data
points. The  estimated parameters for these models  and the corresponding goodness-of-fit to the
available data points, expressed as the root mean square error (RMSE), are given in Table 5 with
the better fit model values in bold. Note that a smaller RMSE represents a better fit to the data.
The Weibull distribution proved to be a better fit for four out of the five examined datasets. The
fits to the disc and weeding datasets were better than that for the remaining three, both  visually
and based on the RMSE values. Figure 1 presents the histogram for the disc operation dataset
and cumulative density functions developed from the data and the Weibull distribution fit to the
data. The mean and median values of the fitted Weibull distribution are shown on the  cumulative
density function line, along with the emissions factor value given by  ARE for a disc operation.
The ARB emissions  factor of 134.5  mg/m2 is very close to the Weibull distribution median of
136.3 mg/m . The emissions factor values corresponding to the 5%, 25%, median, 75%, and 95%
levels along the  cumulative distribution  curve, as well as the  average emissions factor, were
calculated for the five operations and are  presented in Table 6. The 95% level emissions factors
for the three operations with poorer fits  (i.e., higher RMSE) seem very high; this is likely an
artifact of fitting  the  distributions to  a limited number of data points.  In this analysis, the better
fits were obtained for datasets with n > 15.
SDL/08-556                      California 2008 Tillage Campaign                              10

-------
Table 5. Estimated distribution parameters and goodness-of-fit, as root mean square error (RMSE), for the
lognormal and Weibull distributions fit to tillage operation datasets with > 8 data points. Values in bold
represent model with a better fit.
Operation
Disc
Weeding
Standard-till Planting
Strip-till Planting
Listing
n
67
15
11
9
8
Lognormal
Mean
4.61
3.90
3.92
3.21
4.21
G
2.18
1.22
4.45
4.99
5.44
RMSE
0.10
0.06
0.19
0.17
0.23
Weibull
Scale
214.1
89.3
197.0
143.3
366.0
Shape
0.82
0.91
0.58
0.43
0.50
RMSE
0.04
0.07
0.13
0.15
0.22
                                                                                   Empirical CDF
                                                                                   Weibull CDF
                                                                    500
                                           1000
1500
                       EF (mg rrT2)
                                    EF (mg rrT2)
Figure 1. Histogram of disc operation emissions factors and both the empirical and the Weibull cumulative
distribution functions (CDFs).
Table 6. Emissions factor values corresponding to statistical measures of interest for the distributions fitted to
the various tillage operation datasets.
Operation
Disc
Weeding
Standard-till Planting
Strip-till Planting
Listing
Emissions Factor (mg/m )
5%
5.5
6.5
1.2
0.2
1.0
25%
46.1
21.6
23.4
7.9
30.7
Median
136.3
49.2
105.2
61.0
176.6
Mean
240.6
104.1
306.8
396.0
724.8
75%
320.1
112.4
344.4
306.5
701.0
95%
827.1
368.5
1,286.6
1,843.4
3,246.4
SDL/08-556
California 2008 Tillage Campaign
        11

-------
                           2.    EXPERIMENT DESIGN

2.1    SITE DESCRIPTION
After discussion among stakeholders, regulatory agencies, and researchers, it was collectively
determined that the appropriate type of tillage for this experiment would be the tillage sequence
following the harvest of a winter wheat crop in preparation for planting of corn. An appropriate
site was identified in the San Joaquin Valley of California (see Figure  2) consisting of two
adjacent fields. Both fields were cultivated in winter wheat in late 2007 and were to be planted in
corn for the 2008  summer growing season.  The focus of this study was comparing emissions
resulting from the tillage operations  in each field. Therefore, no data were collected to make
comparisons of other potentially varying characteristics, such as crop yield, soil organic matter,
etc., between the  two management  practices.  Tillage management  practices are often crop
specific and are not appropriate for use in all crop production activities.  The effectiveness of
CMPs used in other crop systems at reducing PM emissions should be investigated.
An aerial photograph and soil classification map of the experimental site is shown in Figure 3.
These data were obtained from the Web Soil Survey (WSS), a website maintained by the USDA
Natural Resources Conservation Service (NRCS) [32].  The photo was slightly modified from its
original format  to reflect current land use,  to delineate and label study fields, and to make soil
type boundaries and labels more visible. The study fields are outlined and labeled in yellow. The
soil type boundaries and labels are in white. The two fields that were used for this experiment are
labeled as Field 4 and Field 5.  The fields are referred to by these  names  throughout this
experiment and reflect the designation given to them by the landowner. Both Field 4 and Field 5
are rectangular  in  shape, contain an area of approximately 25 acres  and  are  completely
dominated by soil type  130 (Kimberlina fine sandy loam, saline-alkali).
The area immediately surrounding the experiment fields is dominated by agricultural practices.
Field  4 has  adjacent cultivated cropland on three sides and the fourth side is bounded  by  an
active dairy. Field 5 has adjacent cultivated cropland on two and a half sides and the other side
and a half are adjacent to active dairies. Both fields are surrounded on all sides by roads.   These
roads, with the  exception of one, are dirt roads used for field access  by farm  machinery. The
paved road is a heavily travelled, two-lane asphalt road that was downwind of the fields during
all measurements. Additionally, railroad tracks are located to the north of this site and two to
three  trains pass by per day with varying numbers of cars. The crops observed in the area at the
time of the experiment were grain, corn, almond orchards and grape vineyards.
The terrain surrounding the fields was relatively flat in all directions. Figure 4 is a photo taken on
the southern edge of Field 5 looking across Field 5 and Field 4. The main form of topographical
terrain relief was provided by the drainage and irrigation ditches and canals.

2.2    OPERATION DESCRIPTION
The purpose of this field study was to measure and compare the quantity of particulate emissions
between the conventional method (of spring tillage between a winter wheat crop and a summer
crop  of feed corn)  and  a  Conservation Tillage CMP.  As described  in the Conservation
Management Practices  Program Report (2006), the conservation tillage CMP "involves using a
system in which the soil is being tilled or cultivated to a lesser extent compared to a conventional
system" and it is "intended to  reduce primary soil disturbance operation such  as plowing,
discing, ripping, and chiseling".
SDL/08-556                      California 2008 Tillage Campaign                             12

-------
                t KM 100 Ml«
Figure 2. Shaded relief map of the State of California, USA, with the location of the selected sample site
shown by the white star. Image from geology.com [33].
The Conservation Tillage CMP under study is a strip-till  method which combines multiple
operations to reduce the number of passes required and disturbs the soil only in 8"-wide strips
centered every 30" instead of disturbing the entire surface like the plowing, discing, and listing
operations of a conventional method. This strip-till method therefore reduces the tilled surface by
up to 75% while leaving a lot of the wheat stubble as ground cover in addition to reducing the
number of passes.  The implement used in this study was the  Orthman  1-tRIPr, shown  in Figure
5. The level  of precision and repeatability required by the strip-till method makes the use of
accurate GPS systems a necessity. The cooperating farm has been using the Orthman 1-tRTPr for
seedbed preparation on all of its fields for several years with  the exception of Field 4 which has
continued to be prepared by conventional tillage methods.
SDL/08-556
California 2008 Tillage Campaign
13

-------
Figure 3. Satellite image of the study location with soil types shown in white. The study fields are outlined
within the two rectangles. Soil type 130 represents Kimberlina fine sandy loam, saline-alkali [34].
The conservation tillage  CMP applied in this study consisted of three operations totaling three
passes  across the field, with all three being measured in separate sample periods. In comparison,
the conventional tillage method as applied here had nine different operations totaling 13 passes,
excluding the building and removal of a ditch and in-field borders, with 12 of  them being
measured in nine sample  periods.
The conventional tillage method was employed in Field 4, and the conservation tillage CMP was
used in Field 5. The operations that were performed in each method are shown in order in Table
7, with their corresponding dates. In both the CMP and conventional tillage sequences, ditches
and field-edge borders were built and then broken down between 5/20/2008 and  6/5/2008 to
allow for flood irrigation of the field prior to planting. Drainage ditches on the east  side of both
fields returned excess water to the wastewater lagoon of the adjacent dairy. As the ditch and
field-edge border construction and removal were not measured in Field 5, the corresponding step
for the conventional tillage method was not considered in the total emissions per method. Also,
the in-field borders in Field 5 were not broken down and smoothed out, but instead were used for
the summer corn crop. The term in-field borders as used here applies to low ridges of soil that
separate the field into smaller areas for flood irrigation. Prior to any spring tillage activities, both
Fields 4 and 5 had in-field borders running in roughly an east/west direction, with the irrigation
water sources located on the west side of the fields. In Field 5 the borders were not broken down,
but in Field 4  they were removed and the irrigation water moved from west to east between the
ridges in which the corn was planted.
SDL/08-556
California 2008 Tillage Campaign
14

-------
Figure 4. Photo taken standing near the southern edge of Field 5 looking north across Field 5 towards Field 4.
Note the relative flatness of the terrain.
The cultivator passes in the conventional tillage sequence function as mechanical weed control,
whereas  a chemical weed control (herbicide) is used in the CMP sequence. Additionally, the
cultivator pass  4 was carried out the day after the cultivator pass 3, but unmovable schedules
prevented  its measurement. It is assumed, in  calculating  the  total PM emissions,  that the
emission rates of both passes 3  and 4 were equal. In general, two cultivator passes are done right
after the other in opposite directions down the rows to ensure adequate weed control.
The tractors and  implements used during all the tillage operations are listed in Table 8 by date
and operation. In cases where multiple tractors and implements were used, they are listed in the
order of use. A single, handheld GPS unit was used  to actively log the tractors' positions over
time for each run. During the first part of the lister operation, unharvested plant material along
the border lines  caused clogging of the lister  blades, decreasing its  effectiveness. A second
tractor with the disc set was brought in to go over the border lines again to further reduce the size
of residual plant material. Note that the cultivator passes 1 and 2 and  the roller pass were not
finished when the planter arrived to begin planting Field 4. Therefore, the first sample period on
6/5/2008 was stopped and the second sample period was  started shortly thereafter with the
cultivator and roller still operating in  the north end of the field - the implement locations in the
north end of the  field and meteorological conditions likely prevented significant impacts from
the cultivator and roller operations on  downwind samples.
Field personnel observed operations continually  and  recorded notes on tractor operation times,
potential contamination  issues  due to traffic on surrounding dirt roads  and wind-blown dust,
general meteorological observations, etc.
SDL/08-556
California 2008 Tillage Campaign
15

-------
Figure 5. Orthman 1-tRJPr in operation during this field experiment.
Table 7. Tillage operations and dates performed for the comparison study.
Sequence
Operation
Date
Conservation Tillage (Field 5)
1
2
o
J
Strip-Till
Plant and Fertilize
Herbicide Application
5/17/2008
6/7/2008
6/11/2008
Conventional Tillage (Field 4)
1
2
o
J
4
5
6
7
8
9
10
Break down in-field irrigation borders
Chisel
Disc 1
Disc 2
Lister
Build up ditch and field-edge borders
Break down ditch and field-edge borders,
Cultivator passes 1 and 2, and Roller
Plant
Fertilize
Cultivator pass 3
5/17/2008
5/18/2008
5/19/2008
5/19/2008
5/20/2008
5/20/2008
6/5/2008
6/5/2008
6/18/2008
6/25/2008
SDL/08-556
California 2008 Tillage Campaign
16

-------
Table 8. Agricultural equipment used to perform the tillage operations.
Operation
Strip-till
Break down in-field
borders
Chisel
Disc 1
Disc 2
Lister
Building borders and
ditches
Break borders and
ditches, cultivate,
and roll
Planting
Planting
Herbicide
Application
Fertilizer
Application
Cultivator
Field
#*
5
4
4
4
4
4
4
4
4
5
5
4
4
Date
5/17/08
5/17/08
5/18/08
5/19/08
5/19/08
5/20/08
5/20/08
6/5/08
6/5/08
6/7/08
6/11/08
6/18/08
6/25/08
Tractor
Case MX255
Case Puma 195
Case MX255
Case Puma 195
Case Puma 195
1) Case MX255
2) Case Puma 195
1) Kubota
M8030DT
2) Case 870
1) Kubota
M8030DT
2) Case 870
3) Case Puma 195
4) Case 2290
1) Case Puma 195
2) Case 2290
3) John Deere 4055
Case MX255
Kubota B-Series
Case 2290
Case 1370
Implement (1 per tractor)
Orthman 1-tRIPr, 6 row, 30 in. spacing
Custom border buster (2 sets of 3 discs that move
soil from center to edges)
Custom chisel, 13 ft. wide, 22 in. depth, w/ edged
roller
International Offset Disc, 19 ft. wide, pulling a
single axle (2 smooth road tires), pulling a 19 ft.
wide spiked roller
International Offset Disc, 19 ft. wide, pulling a
single axle (2 smooth road tires), pulling a 19 ft.
wide spiked roller
Custom lister, 6 row, 38 in. spacing
International Offset Disc, 19 ft. wide
Custom 1-way ditch digger (2 sets of 3 discs that
move soil toward center)
Custom border ridger (2 sets of 3 discs that move
soil from center to edges)
Custom 1-way disc (1 set of 3 discs that move soil
from one side to the other)
Custom border buster (2 sets of 3 discs that move
soil from center to edges)
Lilliston Rolling Cultivator, 6 rows wide, 38 in.
spacing
Flat roller, 6 rows wide
Lilliston Rolling Cultivator, 6 rows wide, 38 in.
spacing
Flat roller, 6 rows wide
John Deere MaxEmerge 2 Row Planter, single
row, 6 rows wide, 38 in. spacing
Monosem Twin-Row Planter Model 6x2, 6 rows,
30 in. spacing
Hardi ATV Sprayer, 40 ft. boom
Custom side-dress fertilizer, 6 rows wide, 38 in.
spacing, pulling a fertilizer tank (1 axle, 2 small
smooth tractor tires)
Lilliston Rolling Cultivator, 6 rows wide, 38 in.
spacing
        * Note: Field 4 = conventional tillage practice, Field 5 = conservation tillage CMP
SDL/08-556
California 2008 Tillage Campaign
17

-------
2.3    TILLAGE OPERATION DATA
Based on field notes and GPS data points, the total tractor run time in tractor hours (hrtractor), or
the sum  of individual tractor operation times, and the total area tilled per day were calculated.
This tractor run  time only includes times when the  tractor was moving and performing the
specified operation in the field. The tractor hours we report therefore do not  include time the
tractor spent motionless in the field with or without an idling engine. Operation delays observed
were  refueling, implement adjustments,  changing equipment, and corrections to the on-board
GPS  system.  The  tillage rate  (hectares/hrfractor)  and  the  operation  rate  of the  tractors
(hrtractor/hectares)  were calculated from  the tractor run time and total area tilled. While both fields
were 25 acres  (10.1 hectares) in area, measurement equipment had to be placed inside of the field
edge in Field  5 used for conservation tillage, reducing the monitored area to  about 22.4 acres
(9.05  hectares). The tractor operation rates were summed to provide the total amount of time per
hectare spent preparing the ground for the next crop. All of these data are presented in Table 9.
In this study, the tillage rate of the conservation tillage  operation was 0.86 hrtractor/hectare and the
conventional tillage rate was about five times that amount at 4.90
Table 9. Operation data for both the conventional and conservation tillage studies as recorded by field
personnel.
Operation
Date
Total
Tractor Time
(hrtractor)
Total Area
worked
(hectares)
Tillage Rate
(hectares/hrtractor)
Tractor Rate
(hrtractor/hectare)
Conservation Tillage (study area = 9. 05 hectares)
Strip till
Plant
Herbicide application

5/17
6/7
6/11

3.05
3.82
0.93

9.05
9.05
9.05

2.97
2.37
9.73
Sum
0.337
0.422
0.103
0.862
Conventional Tillage (study area = 10.05 hectares)
Break down borders
Chisel
Disc 1
Disc 2
Lister
Build ditches
Break down ditches, Cultivator
passes 1 and 2, & Roller
Plant
Fertilizer application
Cultivator pass 3

5/17
5/18
5/19
5/19
5/20
5/20
6/5
6/5
6/18
6/25

0.92
6.18
4.83
4.73
5.07
0.83
7.43
3.82
1.08
4.02

2.02
8.51
10.05
10.05
12.48
0.77
23.86
13.21
3.81
10.05

2.20
1.38
2.08
2.12
2.46
0.93
3.21
3.46
3.53
2.50
Sum
0.455
0.726
0.481
0.471
0.406
1.078
0.311
0.289
0.283
0.400
4.90
Due to the breaks in tractor run time and variation in the presence of other tractors, the ratio of
the sample period length and total tractor operation time was slightly different for each run, as
shown in Table 10 below. The difference between total tractor operation time and sample period
time is important because the source strength also varies based on how many tractors, if any, are
operating  at a given time. All calculations of emission rates herein have accounted  for these
differences in source strength with time, with final emission rates based on time reported as the
emission rate of a single tractor.
SDL/08-556
California 2008 Tillage Campaign
18

-------
Table 10. Sample period, total tractor operation time, and the sample period-to-tractor operation time ratio
for all sample periods.
Sample
5/17 Run 1
5/17 Run 2
5/18
5/19 Run 1
5/19 Run 2
5/20 Run 1
5/20 Run 2
6/5 Run 1
6/5 Run 2
6/7
6/11
6/18
6/25
Sample Time (hr)
3.92
0.92
6.58
4.92
5.25
3.83
1.07
7.25
2.00
5.33
1.58
2.17
4.25
Total Tractor Time (hr)
3.05
0.92
6.18
4.83
4.73
5.07
0.83
7.43
3.82
3.82
0.93
1.08
4.02
Sample time/Tractor time
1.29
1.00
1.06
1.02
1.11
0.76
1.29
0.98
0.52
1.40
1.70
2.00
1.06
SDL/08-556
California 2008 Tillage Campaign
19

-------
                   3.     MEASUREMENTS AND METHODS

3.1    MEASUREMENT OVERVIEW
A wide variety of air quality and meteorological sampling equipment was employed during this
field study. These instruments are described in the following sections along with their functions,
data analysis procedures, and calibration verification procedures which were performed before,
during, and after the campaign to ensure the accuracy of data collected.
As wind direction and wind speed are important factors in obtaining accurate and representative
data from  the tillage operations, it was necessary to determine the dominant wind direction for
this period of year in the area. Meteorological data from the California Irrigation Management
Information System (CEVIIS) database were downloaded for May and June of 2005 through 2007
for the Stratford station (#15) [35]. Based on these data, the dominant  wind direction for the
months of May and June in the area was determined to be from the north to the northwest, as
shown in Figure 6.
                      ,.<• X
              sv/
                      >..
                                       -•-.
                                      "•-,  .'
                                                              i
                                           ''10.0%
                                               /I 5.0%
            .'^\   /
                                                                E
'25.0%
    /30.0%
    /
                                             ,-'
                       ,  SF
                      ,   oc:


Figure 6. Wind rose for May and June of 2005 - 2007 as recorded by the CIMIS Station # 15 (Stratford). No
calm periods were recorded.
SDL/08-556
California 2008 Tillage Campaign
                            20

-------
Due to the layout of the fields and adjacent dairy, the preferred wind direction for sampling was
from the north to northwest. Assuming a northwest wind, instrument deployment was such that
samplers meant to measure background aerosol parameters were to the west and northwest of the
fields of interest. Samplers meant to measure background plus plume parameters were to the
south and  east.  Similar  sample layouts were arranged  around  each field  using  the same
instruments. Some equipment, such  as the lidar and air quality trailers, remained in the same
place for samples in both fields while more portable samplers were moved between fields.
The first layout, shown in Figure 7, presents the sample layout for monitoring conventional
tillage practices in Field 4. Table 11  summarizes  the types of instruments that were located at
each site and the dates certain instruments were used if they were not used during the entire
study. The greatest deviations in the daily layout occurred during the runs on June 18*  and 25*,
which  are  due to the  following situations:  1) a separate study of particulate  and gaseous
emissions from the adjacent dairy was carried out  from June 13th to the 20th during the break in
tillage operations that required the relocation of most samplers, including both the Aglite lidar
and Air Quality trailers; and 2) a failure in the Aglite lidar system on June 19  prevented the
lidar from being used in subsequent sample periods. The lidar system was located at position L2
throughout  the dairy investigation and for the June 18th sample of the fertilizer application  in
Field 4.
The sample array arranged to monitor the conservation tillage activities in Field 5 is shown  in
Figure 8. Table 12 summarizes the type of instruments that were located at each site. In contrast
to the sampling changes to measure  operations in  Field 4, there were not significant deviations
from the original sample layout used around this field as all sampling occurred prior to moving
equipment for the dairy emissions study.

3.1.1     Meteorological Measurements
A Vantage  Pro2  Plus weather station from Davis Instruments, Inc. (Hayward, CA) was used  to
monitor wind speed, wind direction, temperature, humidity, precipitation, barometric pressure,
and solar radiation. It was located on top of the Air Quality sampling trailer, at an  approximate
height of 5 m above ground  level. It was wired to a datalogger and a display panel inside the
trailer, which was connected to a computer for data storage.
Two 15.2 m towers were instrumented with 3-cup anemometers (model 12102) at five heights
(2.5,  3.9,  6.2,  9.7  and  15.3  m)  to  measure  the vertical  wind speed  profile.  Relative
humidity/temperature sensors  (Vaisala HMP45C)  from Campbell Scientific, Inc.  (Logan,  UT)
were  also  stationed  at five  heights (1.5,  2.5, 3.9, 6.2,  and 9.7  m) to provide profiles  of
temperature and relative humidity. A Met One Instruments, Inc. (model 024A, Grants Pass,  OR)
Wind Vane was stationed at a height  of 15.3 m on each tower. Data from each tower were stored
as one minute averages on the Campbell Scientific CR23X dataloggers and were downloaded
daily.
The eddy covariance (EC) instrumentation was  comprised  of four Campbell Scientific Inc.
(Logan, UT) 3D sonic anemometers (CSAT)  and four LiCOR 7500  infrared  gas  analyzers
(IRGA). The sensor separation for all four EC systems was 10 cm. All data were stored on to a
Campbell Scientific data logger (CR5000). Together the  CSAT and LiCOR measured water
vapor (q) and carbon dioxide (c) concentrations, and velocity components of the  wind flow  in
three spatial dimensions: x, y,  and z.  These measurements were made at a scan rate of 20 Hz, 20
measurements per second for w, v, w, q and c. Each EC instrumentation group was visited daily
SDL/08-556                      California 2008 Tillage Campaign                             21

-------
between  06:00-07:00  hours  for maintenance. Maintenance performed  included  exchanging
compact  flash cards for data  storage  on the CR5000, sensor interrogation at the  data logger
screen to evaluate measurement status, cleaning dust from the surface of the IRGA lens with de-
ionized water, and removing spider webs from the transducer arms.
     I
      Legend
          Met Tower
        A  Tower
        •  Tripod
          AQ Trailer
      °:
                                                                 11.0
                                                                   I
                                                       I   I  I
                                                        Held 4
                                                    onventional Tillage
                                                       25 Acres

                                                                              r

                                                              Fields
                                                         Conservation Tillage
                                                              25 Acres
                                      Meters
                                     400
Figure 7. Sample layout used for Field 4.

Table 11. Summary of instruments located at each position for the conventional tillage study of Field 4. All
heights given as above ground level (agl).
Instrument
Location
Description
Tl
1-10 m tower
1 - OPC: 1 @ 9 m
3 - MiniVols: 3 @ 9 m (TSP, PM10 andPM25)
T2
1-10 m tower
1 - OPC: 1 @ 9 m
3 - MiniVols: 3 @ 9 m (TSP, PM10 and PM25)
WM
1-15 m tower
5 - cup anemometers: 1 each @ 2.5, 3.9, 6.2, 9.7 and 15.3 m
1 - wind vane: 1 @ 15.3 m
5 - temp/RH sensors: 1 each @ 1.5, 2.5, 3.9, 6.2, and 9.7 m
2 - Campbell Scientific dataloggers
1 - sonic anemometers: 1 @ 11.3 m
1 - energy balance systems: 1 @ 2  m	
SDL/08-556
                   California 2008 Tillage Campaign
22

-------
Instrument
Location
EM
LI
L2
EC1
EC2
ECS
2.4
6.4
7.4
8.4
9.4
10.0
11.0
12.0
LARS
AQT
Description
1-15 m tower
5 - cup anemometers: 1 each @ 2.5, 3.9, 6.2, 9.7 and 15.3 m
1 - wind vane: 1 @ 15.3 m
5 - temp/RH sensors: 1 each @ 1.5, 2.5, 3.9, 6.2, and 9.7 m
2 - Campbell Scientific dataloggers
1 - sonic anemometers: 1 @ 11.3 m
1 - energy balance systems: 1 @ 2 m
2 - OPCs: 1 @ 9 m, 1 @ 14.5 m
1 - Lidar data collection system (5/14 - 6/11)
1 - Davis met station for lidar operator's reference
1 - Lidar data collection system (6/18)
1 - Davis met station for lidar operator's reference
1 - Campbell Scientific CSAT @ 1.6 m
l-LiCORIRGA@1.6m
1 - Campbell Scientific CSAT @ 2.9 m
1 - LiCOR IRGA @ 2.9 m
1 - Campbell Scientific CSAT @ 1.6 m
1 - LiCOR IRGA @ 1.6m
1 - 2 m tripod
1 - OPC: 1 @ 2 m
3 - MiniVols: 3 @ 2 m (TSP, PM10 and PM25)
1 - 2 m tripod
2 - MiniVols: 2 @ 2 m (PM10 and PM25)
1 - 2 m tripod
1 - OPC: 1 @ 2 m
2 - MiniVols: 2 @ 2 m (PM10 and PM25)
1 - 2 m tripod
2 - MiniVols: 2 @ 2 m (PM10 and PM25)
1 - 2 m tripod
2 - MiniVols: 2 @ 2 m (PM10 and PM25)
l-2mtripod(5/18-6/ll)
1 - OPC: 1 @ 2 m
3 - MiniVols: 3 @ 2 m (TSP, PM10 and PM25)
l-2mtripod(6/18, 6/25)
1 - OPC: 1 @ 2 m
3 - MiniVols: 3 @ 2 m (TSP, PM10 and PM25)
l-2mtripod(6/25)
1 - OPC: 1 @ 2 m
3 - MiniVols: 3 @ 2 m (TSP, PM10 and PM25)
1 - OPC: 1 @ 2.8 m
1 - OPC: 1 @ 5 m
3 - MiniVols: 3 @ 5 m (TSP, PM10 and PM25)
1 - Davis met station: 1 @ 5 m
1 - OC/EC Analyzer: 1 inlet @ 4.5 m
1 - AMS: 1 inlet @ 4 m
1 - radio and laptop for OPC Data collection
Note: The Air Quality Trailer was in the position indicated on the map for the tillage
practices that occurred from 5/1 7/2008 through 6/11/2008. For the two remaining runs on
6/18/2008 and 6/25/2008 the Air Quality Trailer was in a different location and the
analysis equipment were either not used or located in a different downwind position.
SDL/08-556
California 2008 Tillage Campaign
23

-------
       Legend
        _  Met Tower
        A  Tower
        •  Tripod
        •  AQ Trailer
        •  Lidar
           Field Outline
Figure 8. Sample layout used for Field 5.
Table 12. Summary of instruments located at each position for the Conservation tillage study of Field 5. All
heights given as above ground level (agl).
Instrument
Location
Tl


T2

T3


WM



Description
1
1
3
1
1
1
1
3
1
5
1
5
2
1
1
-10 m tower
- OPC: 1 @ 9 m
- MiniVols: 3 @ 9 m (TSP, PM10 and PM25)
-10 m tower
- OPC: 1 @ 9 m
-10 m tower
- OPC: 1 @ 9 m
- MiniVols: 3 @ 9 m (TSP, PM10 and PM25)
-15 m tower
- cup anemometers: 1 each @ 2.5, 3.9, 6.2, 9.7 and 15.3 m
- wind vanes: 1 @ 15.3 m
- temp/RH sensors: 1 each @ 1.5, 2.5, 3.9, 6.2, and 9.7 m
- Campbell Scientific dataloggers
- sonic anemometers: 1 @ 11.3 m
- energy balance systems: 1 @ 2 m
SDL/08-556
California 2008 Tillage Campaign
24

-------
EM
LI
EC1
EC2
ECS
2.5
6.5
7.5
8.5
9.5
10.0
LARS
AQT
1-15 m tower
5 - cup anemometers: 1 each @ 2.5, 3.9, 6.2, 9.7 and 15.3 m
1 - wind vanes: 1 @ 15.3 m
5 - temp/RH sensors: 1 each @ 1.5, 2.5, 3.9, 6.2, and 9.7 m
2 - Campbell Scientific dataloggers
1 - sonic anemometers: 1 @ 11.3 m
1 - energy balance systems: 1 @ 2 m
2 - OPCs: 1 @ 9 m, 1 @ 14.5 m
1 - Lidar data collection system
1 - Davis met station for lidar operator's reference
1 - Campbell Scientific CSAT @ 1.6 m
l-LiCORIRGA@1.6m
1 - Campbell Scientific CSAT @ 2.9 m
1 - LiCOR IRGA @ 2.9 m
1 - Campbell Scientific CSAT @ 1.6 m
1 - LiCOR IRGA @ 1.6m
1 - 2 m tripod
1 - OPC: 1 @ 2 m
3 - MiniVols: 3 @ 2 m (TSP, PM10 and PM25)
1 - 2 m tripod
2 - MiniVols: 2 @ 2 m (PM10 and PM25)
1 - 2 m tripod
1 - OPC: 1 @ 2 m
2 - MiniVols: 2 @ 2 m (PM10 and PM25)
1 - 2 m tripod
2 - MiniVols: 2 @ 2 m (PM10 and PM25)
1 - 2 m tripod
2 - MiniVols: 2 @ 2 m (PM10 and PM25)
l-2mtripod(6/7-6/ll)
1 - OPC: 1 @ 2 m
3 - MiniVols: 3 @ 2 m (TSP, PM10 and PM25)
1 - OPC: 1 @ 2.8 m
2 - OPC: 2 @ 5 m
3 - MiniVols: 3 @ 5 m (TSP, PM10 and PM25)
1 - Davis met station: 1 @ 5 m
1 - OC/EC Analyzer: 1 inlet @ 4.5 m
1 - AMS: 1 inlet @ 4 m
1 - radio and laptop for OPC Data collection
3.1.2     Wind Profile Calculations
Wind profiles near the ground surface were calculated based on one-minute averaged wind speed
data from the logarithmically spaced cup anemometers on the 15m meteorological tower at the
WM location, which was the meteorological  tower located upwind/crosswind of both tillage
sites. A power law wind speed profile fit to the tower  data was extrapolated up to 250 m to
estimate the wind speed at higher elevations. The power law for calculating wind speed (ui) at
height 22 based on a measured wind speed (iij) at height zj  is shown in Eq. 5 from Cooper and
Alley (2002) [36].
SDL/08-556
California 2008 Tillage Campaign
25

-------
                                                                                     (5)

where wind speeds are in m/s, height is in meters, andp is a unitless coefficient that varies based
on atmospheric stability and surface roughness. The approximate value ofp is 0.5 for very stable
conditions and 0.15 for very unstable conditions.
An error in the code used to calculate the one-minute average wind  directions from the wind
vane at the top of the meteorological towers (15.3 m) was discovered in post-processing of the
data taken during the 2007 fall tillage CMP study [7]. For both the 2007 fall tillage and the 2008
spring tillage datasets, one-minute averaged horizontal wind directions calculated from the sonic
anemometer data were used in subsequent analyses. The wind direction measured by the sonic
anemometer was applied to the entire wind speed profile. For most sample periods, wind profiles
above 150 m were not required for the lidar emission rate calculations. Averaged horizontal wind
speed data from  the sonic  anemometer (11.3 m  agl) on N Met were compared with the nearest
cup anemometer (9.7 m agl) wind speed measurements as a quality check. Both data sets showed
the same patterns  and recorded wind  speeds  were close (< 0.25 m/s difference), with  the
observed difference likely due to a combination of vertically  separated sample  heights and
instrument error.

3.1.3     Soil Characterization
Soil characterization involved collecting soil samples for analysis of bulk density, soil moisture,
and sand/silt/clay content. Bulk density samples were collected in each  field  prior to  tillage
operations. A manual device consisting of a 7.6 cm-diameter and 7.6 cm-deep cylinder was hand
driven  into the soil until the top of the cylinder was level with the soil surface. Samples were
removed and  placed into pre-weighed  cans. Post-weights were performed in the field  for
determination of wet weight. All weights were determined using a Mettler balance (Columbus,
OH), Model PM2000.
Samples for soil moisture were taken for each day of operation at random locations in the field
and collected in pre-weighed cans 7 cm in diameter and 4 cm  deep. Samples were collected
immediately prior to the tillage period or shortly  after commencement in areas that had not been
tilled. The can was pushed into the soil approximately 3cm then removed. The  can was  closed
and weighed in the  field for determination of wet weight.
All soil samples were dried at the Agricultural Research Service (ARS) National Laboratory for
Agriculture and  the Environment (NLAE) in Ames, IA  at  a temperature of  105 °C until a
constant weight was achieved (-60 hours).  Samples were then weighed to determine dry weight.
Calculations for soil moisture and  bulk  density were performed according to the following
equations as found  in Doran and Jones 1996 [37].
SDL/08-556                      California 2008 Tillage Campaign                            26

-------
    field water content (%) = weight of moist soil - weight of oven dried soil x 100         (6)
                                weight of oven dried soil

    % moisture = weight of moist soil - weight of oven dried soil x 100                    (7)
                                  weight of moist soil

           bulk density = weight of moist soil x (1 - field water content)                   (8)
                                 volume of soil collected
where "volume of soil collected" = TI x radius2 x length of cylinder = 71 x 3.812 x 7.62 = 347.3
cm3 and "field water content" is the value given by Eq. 6 expressed as a fraction.
A composite was made of all the samples collected.  It was analyzed for the percent of sand, silt,
and clay according to the  Hygrometer Procedure,  as given  in Soil  Sampling and Methods of
Analysis (1993) [38]. The percent of stable aggregates was also determined from  the composite
sample according to the Dry-Sieve Method, as given in the Soil Sampling and Methods of
Analysis (1993) [39].

3.1.4     Air Quality Point Samplers
The suite of air quality  point samplers deployed around the tillage plots to quantify both the
ambient and ambient plus operations emissions values were summarized in Table 11 and Table
12. Details of each of these sensors and their data processing methods are presented below.

3.1.4.1        MiniVol Portable Air Sampler
Twenty-four MiniVol Portable  Air Samplers (Airmetrics,  Eugene,  OR) were  distributed to
gravimetrically measure the time-averaged mass  concentrations of aerosols at multiple locations
surrounding the fields of interest. The MiniVol  is a battery  operated, ambient air sampler that
gives results that closely approximate air quality data collected by a Federal Reference Method
(FRM) PM sampler  [40][41]. The MiniVol is  neither designated as an FRM nor a Federal
Equivalency Method (FEM) by the EPA,  and  results should be  considered as approximate
measures of PM. The sampler draws air through  a particle size separator, or impactor head, and
then through a filter medium [42]. The photo in Figure 9 shows MiniVols mounted  on a
rechargeable battery pack  with  attached impactor  sample heads deployed during the reported
study.
Particulate concentrations in the PMio and PM2.5 size fractions were measured using impactor
heads for size separation based on  aerodynamic diameter, while Total  Suspended  Particulate
(TSP) was measured with the impactor head removed. Each PM sampler was assigned to sample
a specific size fraction at a  specific location throughout the study, with location changes made as
deemed necessary. Clusters of three PM samplers were assigned to four locations, two upwind
and  two  downwind,  in  order  to  provide size   fractionated,  mass-based  particle  loading
distributions.
SDL/08-556                      California 2008 Tillage Campaign                             27

-------
Figure 9. Two Airmetrics MiniVol Portable Air Samplers and a Met One Instruments Optical Particle
Counter (OPC) deployed for field sampling during Spring 2008 tillage study.
Filters used in the PM samplers were pre-conditioned according to the protocols outlined in 40
CFR 50 Appendix J before obtaining pre- and post-sample filter weights. Final average weights
were found using a Mettler Toledo Microbalance, Type MT5  located at the Utah Water Research
Laboratory (UWRL) in Logan, UT to determine three stable weights within ± 5 jig, measured on
different days. Filters were transported to and from the site and stored on-site in dessicators to
maintain filter conditioning. Flow calibrations on each MiniVol were performed using a slant
manometer prior to the study and the actual sample flow was adjusted daily on each instrument
in order to maintain the required 5.0 L/min for accurate particle size separation.
During this study it was found that the PM2.5 and PMio impactor heads were  being overloaded
with particulates  due to high wind  conditions and extremely  high ambient particle loading.
Therefore, for June sample periods, we installed a PMio impactor in series with a PM2.5 impactor
to serve as an additional filter.  These efforts are described more fully in Appendix B.

3.1.4.2       Optical Particle Counter
Ten Optical Particle Counters (OPCs),  Model  9722 from Met One Instruments, Inc. (Grants
Pass,  OR), were collocated with MiniVols in order to describe the particle  count and size
distribution  at locations  measuring background  aerosols  and  those  locations  measuring
background plus plume aerosols. Figure 9 shows a Met One  OPC collocated with MiniVols for
sampling in Field 5. The  9722 particle  counter uses scattered light to size and count airborne
particles. Particle counts are reported in eight, user-defined channels over a user-defined sample
time. For this study, the OPCs collected samples  continuously  at a sample time of 20 seconds
with the following channel sizes, in units of |im: (1) 0.3-0.5,  (2) 0.5-0.6, (3) 0.6-1.0, (4) 1.0-2.0,
SDL/08-556
California 2008 Tillage Campaign
28

-------
(5) 2.0-2.5, (6) 2.5-5.0, (7) 5.0-10.0, and (8) >10.0. The data from each OPC were relayed to a
single computer over a custom radio network for storage. Inlet flows for individual OPCs were
measured on-site before and after the experiment using a Gilian Gilibrator2 Calibration System,
a volumetric flow meter. The average of the averages from each flow measurement period was
used as the sample flow throughout the field study.
Calibration of OPC particle  counts  was  performed for  each sample day in post-campaign
analysis. For this  calibration, careful examination  of the number concentration (number of
particles/m3) time series yielded a time period before or after the tillage operation during which
no plumes were detected. Number concentration was chosen as the calibration point, as opposed
to the raw particle count data, because it normalizes the raw particle counts by sample flow (see
Eq. 10); sample flow varies up to 20% between OPCs. The average number concentration (Ny)
per bin (/') for  the designated calibration time for each OPC (/) was calculated,  and the mean of
the averages  (Nt~) was used  as the calibration concentration.  A scalar correction,  Counting
Correction Factor (CCFy), was applied to each bin (/') of each OPC (/) and was calculated for
each collocated run based on the calibration concentration for that bin according to the following
equation:

                                                                                    (9)
Number concentration (Ny) is a function of raw particle counts (py), the measured average flow
rate (qj), the sample time (tj), and the CCFy, as shown in Eq. 10.

                                         pa x CCFn
                                    N,,  ^ - L                                (10)
where the units for each variable are N   number/cm3, pt   number, qj = cm3/min, t   minutes,
and CCF is unitless. As in Eq. 9, the subscript y represents a specified OPC.
The volume concentration of aerosols based on a number concentration TV is calculated based on
the following assumptions:  1)  the particles are spheres, 2) the maximum particle diameter
measured is 20  jim,  and  3)  the  geometric  mean  particle  diameter per  bin  (GMD,)  is
representative of the particles in a given bin / with the assumption of a log-normal distribution of
particle numbers. The GMD, is calculated by Eq.  11.
                                 GMD,   ^upperxd,Jower                             (11)

where di:Upper and di:iower are the diameters of the upper and lower ranges for bin /'. The assumption
of a maximum measured particle diameter must be made in  order to calculate the GMD for
channel 8, which counts particles > 10 jim.
The cumulative volume concentration of aerosols (Vk) up to a particle diameter k may be
calculated using the following equation:
SDL/08-556                      California 2008 Tillage Campaign                             29

-------
                                          Vk  ~N(d)dedd                         (12)
where N(d) is the number concentration at diameter d. For application to the OPC data, Eq. 12 is
expressed in the following terms that have been previously defined:
                                  Vk    -     fiME?tf,                               (13)'
                                       o   i i
where GMD, is expressed in |im, N, is in number/cm3, and Vk is in units of |im3/cm3.  In this case,
the Vk definition is similar to PM& concentrations: the fraction of the total volume of particles
whose diameter, in jim, is < k = 1, 2.5, 10, and oo for TSP.
By collocating PM  samplers  and OPCs, the data provide information about the  relationship
between  optical  and  aerodynamic  measurements  and allow  direct  calibration  of optical
instruments  (both OPC and lidar) to mass concentration instruments by estimation  of the mass
conversion factor (MCF) for  each PMk fraction. The theoretical  conversion  from particulate
volume concentration to mass concentration is complex, and several simplifying assumptions
must be made.  These include a spherical particle shape approximation, a priori assumption of
the refractive index, and neglecting multiple  scattering effects. The time-resolved V^ data from
each OPC as calculated in Eq. 13 are then averaged over the corresponding PM sampler sample
time. The MCFs, in units of density (g/cm3), for each PM size  fraction k are calculated by
dividing the mass concentrations measured by the PM samplers (PMk) by the time-averaged V^.
These data are averaged over several locations or instrument clusters j, where Ey = «, and both a
daily mean value and an overall mean value of the MCF,t is calculated for each PM,t fraction
separately.
                                       1 '  "  PM
The aerosol number size distribution (dNld(ln(&))) is calculated as outlined in Hinds [43] and
expressed mathematically in Eq. 15.

                           dN/d(\n(d»   — - i- - -                       (15)'
where di:Upper and drawer are the diameters of the upper and lower ranges of bin /'.

3. 1 .4.3       Organic Carbon/Elemental Carbon Analyzer
An Organic Carbon/Elemental Carbon Analyzer (OC/EC), Model 5400 from  Rupprecht and
Patashnick Co.,  Inc. (Albany, NY), was located  in the  Air Quality  Trailer  (AQT) on the
downwind borders of Field  5. This instrument  provides sample-averaged organic carbon and
elemental carbon mass concentrations over a user-defined sample time, which  was set at one
hour for this study. In operation, the system alternately collects  particulate matter onto one of
two ceramic filters which, after the desired collection period, are heated within a closed-loop
system  to determine carbon content via direct thermal desorption and pyrolysis  techniques
developed  and  validated by Rupprecht and  Patashnick  [44]. As recommended  by the
manufacturer,  during the analysis  phases, an initial  temperature plateau of 250°  C  for 600
seconds was used for determination of the organic carbon (OC) fraction and a final temperature
SDL/08-556                      California 2008 Tillage Campaign                             30

-------
plateau of 750° C was used for quantification of the elemental carbon (EC) fraction. To account
for non-carbon components of the organic compounds' mass, the OC concentrations reported by
the 5400 analyzer were increased by the recommended multiplier of 1.7 [45]. A sharp-cut PM2.5
cyclonic separator was placed at the inlet, which was located directly above the instrument on
top of the trailer. A nominal flow rate of 16.7 Lpm was maintained through the system by on-
board mass flow controllers and integral temperature and pressure sensors.
The  5400 instrument successfully passed  the flow checks, leak checks, and CC>2 audits as
prescribed by its manual the week prior to departure for the field campaign [44]. The instrument
passed these  same  checks upon setup in  the field  on  May  13,  2008.  It passed CC>2 audits
administered in the field on June 3 and June 21, 2008.

3.1.4.4       Ion Chromatographic Analysis
To more fully chemically characterize the nature of the upwind and  downwind particulate matter,
ion chromatographic (1C) analysis was performed on filters collected via the MiniVol systems
that were collocated with the EC/OC inlet on top  of the Air Quality Trailer (AQT). After final
post-test weights  were  determined  from the filters  at the UWRL, the chosen  filters were
sonicated with triplicate  rinses in 10  mL  double-distilled, de-ionized water (DDI) and split into
two aliquots of approximately 15  mL each for separate anion and cation analysis. The anion
analysis occurred within 48 hours of sonication and the cation aliquots were stabilized with 10
uL of 0.5 M HC1 acid and analyzed  within 28 days of sonication.  The base 1C system (Dionex
Corporation) utilized an AS 40 (Automated  Sampler), a CD 20 (Conductivity Detector), a GP 40
(Gradient Pump), and a membrane suppressor. Cation quantification was accomplished using an
lonPac® CS12A cation column, a CG12A cation guard column, and a 500 jiL sample loop. The
system  eluent was 0.15 M H^SC^ with  a  1.0  mL/min  flow.  Anion concentrations were
determined using an lonPac® AS11HC anion column, a AG11HC anion guard column, and a 188
|iL sample loop. For anions, the system eluent was 30 mM NaOH with a 1.0 mL/min flow.
American Chemical Society (ACS) reagent grade materials were used to prepare stock standard
solutions for each of the target ions, from which concentrations of  0.5, 1.0, and 5.0 mg/L (ppm)
were mixed to make 1C  calibration curves. The ions quantified were fluoride (F~), chloride (Cl~),
nitrite (NCV), sulfate (SO/f2), nitrate  (NCV), sodium (Na+), ammonium (NH4+), potassium (K+),
magnesium (Mg+2), and calcium (Ca+2). Verifications of the system calibrations were performed
prior to each  analysis  run.  Continuing  calibration  verification standards (CCVs) and blank
samples (DDI water) were  analyzed roughly every 10  samples.  Peak identification and data
processing were executed using Dionex PeakNet Data Chromatography software (Version 2.0).

3.1.4.5      Aerosol Mass Spectrometer
An Aerosol Mass  Spectrometer (AMS)  from Aerodyne Research, Inc.  (Billerica,  MA) was
located in the sampling trailer, with a sample port on the upwind side of the trailer just below the
roof  level (~3  m).  The  AMS provides chemical composition and  particle size  information for
volatile and semi-volatile particle components in the 0.1 -  1.0  jim  size ranges in vacuum
aerodynamic diameter. Hence, it is nominally considered a PMi instrument for the fraction of
chemicals  detected.  An  AMS  size calibration  was  conducted  on-site  on May  14 using
polystyrene latex spheres (PSLs). Mass calibrations were also conducted on-site on May  14 and
June 4,  2008 using ammonium nitrate. The AMS vaporizer was operated at higher than normal
temperature (-800 °C vs. -600 °C) to detect some of the inorganic  components in dust particles.
SDL/08-556                      California 2008 Tillage Campaign                             31

-------
During  sampling, the AMS integrated and saved particle composition and size data every ten
minutes.

3.1.5     Lidar Aerosol Measurement and Tracking System
The Aglite lidar system is a monostatic laser transmitter and 28-cm receiver telescope (Figure
10). The laser is  a three-wavelength, 6W, Nd:YAG laser, emitting at 1.064 (3W), 0.532 (2W)
and 0.355 (1W)  um with a 10 kHz repetition rate.  The lidar utilizes  a turning mirror turret
mounted on the top of a  small trailer to direct the beam -10° to + 45° vertically and +  140°
horizontally.  Lidar scan rates from 0.5° - l°/s are used to develop the 3D map of the source(s),
dependent on range and concentration of the aerosol.
The process used to retrieve aerosol mass concentration from lidar data is illustrated in Figure
11. The details of Aglite lidar calibration and the aerosol retrieval process are discussed by
Marchant [46] and Zavyalov [47]. The retrieval is as follows:  first, the lidar data is preprocessed.
Then,  the  relationships   between  backscatter,  extinction,   volume concentration, and mass
concentration of the aerosol components are established using in-situ data measured by OPCs
and clusters of PM samplers with different separation heads  (P1VU). After that,  the inversion of
the lidar data is performed using a form of Klett's solution  [48] for two scatterers where
extinction is proportional  to backscatter. Finally,  a least-squares method is used to convert
backscatter values to aerosol mass concentration using the previously established relationships.
From the OPC channel counts, the particle size distribution at a single point as a function of time
may be calculated according to Eq. 15. The backscatter and extinction coefficients necessary for
solving the lidar equation are then calculated at the OPC reference point as a function of time.
Klett's inversion is used to convert the lidar signal to the optical parameters (backscatter values
in particular) of particulate emitted by the operation [48]. Having recovered backscatter values as
a function of range and wavelength using the Klett inversion, these must be converted to the
         Figure 10. The three wavelength Aglite lidar at dusk, scanning a harvested wheat field.
SDL/08-556
California 2008 Tillage Campaign
32

-------
                        f Lidar Signal 1  foPC Data]   fpM Samplers]
1

Preprocessing
i

Klett Inversion
{
Aerosol
Concentration
Retrieval
<

r Volume ^
Iponcentration;
4



^Particle Size^j 	
^Distribution^'
\
/^Boi
*~ Con
\Lida
(,


mdary^
dition +
• Ratiosy
i
Ma
onve
Fac


ss ^
rsion .
tor j



                            Mass
                        ^ConcentrationJ
Figure 11. The Aglite lidar retrieval algorithm flow chart, showing the input locations for the in situ data.
aerosol cumulative volume concentration. Expressing the particle normalized backscatter values
from the OPC (pv) and the lidar measured backscatter values (PE) in a vector form, and applying
the Moore-Penrose weighted minimum  least-squares solution results in the value for n(r), the
relative amplitude of the aerosol component to total atmospheric backscatter, at range r
                                    n(r]
                                                     (16)
which can be multiplied by the particle normalized volume concentration vector, resulting in the
                                     VK(r)  VEn(r)
                                                     (17)
The term W is a diagonal weighting matrix, whose diagonal elements are the expected variance
of the emission backscatter for the corresponding channel.
The retrieved aerosol volume concentration from the lidar return signal is multiplied by the
MCFfe which was previously calculated using in-situ data (Eq. 14). At this point, the Mi fraction
of the aerosol mass concentration of the emission component is known as a function of distance.
                                 PMK(r)
           MCFK
•VK(r)
(18)
The concept behind our flux measurement approach is shown in Figure 12A, where the facility
is treated as one would calculate the source strength in a bioreactor. In this simplified
mass-balance approach, the source strength is determined using the mean flow rate through the
reactor and the difference in reactive species concentration entering and leaving the vessel. The
scanning lidar samples the mass concentration fields  entering and leaving the facility and
standard cup anemometers provide the mean wind speed profile. If we define our box large enough
so that none of the emitted material escapes through the top or crosswind sides of the box,  and
the downwind side is far enough from the facility to minimize high frequency  fluctuations,
the same simple relationship found in a bioreactor applies.
SDL/08-556
California 2008 Tillage Campaign
                                     33

-------
    Wind Direction
  Flux In
                                         900
                                           800
                                          range 700
                                          (meters)  BOO
Figure 12. (A) Conceptual illustration of the method for using lidar to generate time resolved local area
particulate fluxes. (B) An example of a "staple" lidar scan over the facility showing aerosol concentration on
the three sides of the box.
An example of our lidar-derived concentration data is shown in Figure  12B. The concentration
plot pattern from  scanning up one side, across the top, and down the other looks like the common
office  staple  (the paper fastener), and will  be  referred to  as a'staple' scan. The data from
the top of the box is  regularly examined  to be  sure  that no significant particulate transport
is passing through the top. The data for the left side panel of the 'staple'  provides the background
concentration entering the box, while that on the right provides the background plus facility
concentrations leaving the box. The integrated mass concentration difference multiplied by the
wind speed during the scan completes the flux emission  calculation by yielding mass per unit
time emission from the facility.
The flux calculation in the integral form for calculating facility/operation  emissions (F) can be
expressed as following:
                             F
                                                      (19)
                                 r,h
where Vj/',   is the average wind speed component perpendicular to the lidar scanning plane,
CrKr, h) is the downwind PM& concentration  at position range r and height h, and Co is the
average PMg concentration in the upwind scanning plane. CD -  Cu form the mass concentration
difference upwind and downwind, integrated over the range (width) and height of the sides of the
staple. In our routine, Eq.  19 is discretized as:
                                   R  H
                                                                                      (20)
where RO and R are the near and far along beam edges of the box and H0 and H form the top and
bottom of the box. (In many cases, HO is set above eye level and concentration is extrapolated to
the ground to avoid illuminating personnel and animals.)  The Ar-A/z term is the individual area
element for which each flux component as calculated by each step in the double summation.
SDL/08-556
California 2008 Tillage Campaign
34

-------
Figure 12B shows an example of single-scan Cu and CD  mass concentration values for
measured along the vertical sides of the staple shape between the distances of 600 and 900 m
from the lidar. Single-scan differences, of course, do not account for accumulation or depletion in
the box due to wind speed variation during a scan, or from high or low concentration pulses that
may still exist in the downwind sample. For a meaningful estimate of the facility emission, many
scans are combined to achieve a time-averaged emission rate.
To perform the lidar calibration using the in-situ instruments, collocated samples  are needed for
the PM sampler, OPC, and lidar. In other words,  the lidar beam must directly sample the same
volume as an OPC/PM sampler location for at least the duration of one OPC sample collection
time.  When the lidar beam is held constant, usually pointing next to a sample tower, we refer to
this as the 'stare'  mode. The  'stare' sample mode provides not only calibration, but quality
assurance of the data set as well. After data processing and PM concentrations along the beam
have been calculated, successive 'stare' data are compared against data from the reference OPC
and PM samplers to verify that the concentrations calculated from both instruments are similar
within measurement errors.
For most of this study, the Aglite lidar was located at position LI and used the tower Tl, located
at 31° as the upwind reference point. Downwind scans for Field 4 were  made  along the line
between LI and T2 at about 73°; downwind scans for Field 5  were made along the line between
LI and AQT at about 91°. Scanning profiles were programmed for each field and included both
'staple'  scans  and 'stares', with some variations  as necessary according  to conditions at  the
sample time. An example  of a lidar scan profile used for each field is provided as a reference,
with Figure 13 and Figure  14 presenting graphical representation of the azimuth and elevation of
the beam  during  each sequential  data  averaging  and  archival  period (averaged and  archived
every 0.5-1.0 seconds).
Field 4 - The  scanning pattern  used during the breaking borders  operation on  May   17
commenced with a stare at 31° for 73 seconds followed by three upwind vertical  scans reaching
45° in elevation and lasting 90 seconds each. The  downwind portion of the profile consisted of
three vertical scans at an azimuth of 73.5°, each reaching 45° in elevation and lasting 90 seconds.
These vertical  scans were followed by a stare at the downwind tower for 67 seconds. The beam
was then returned to 31°  and  began again.   The total time per  profile including intermittent
delays to position azimuth was  758 sec. (-12.6 min.).
Field 5 - The scanning pattern used during the herbicide application operation on June 11
commenced with a stare  at 31° for 33 seconds followed by two upwind vertical scans reaching
15° in elevation and lasting 29 seconds each. The downwind portion of the profile  consisted of a
stare for 63 seconds at azimuth  angle 91° followed by four vertical scans each reaching 15° in
elevation and lasting 29 seconds. The profile then moved the beam to an azimuth of 73° for a
stare at T2 lasting 18 seconds. The beam was then returned to 31° and the profile was started
over.  The total time per  profile including intermittent delays to position azimuth was 395 sec.
(-6.6  min.).
SDL/08-556                      California 2008 Tillage Campaign                             35

-------
                                 500
            1000
           Data Point
1500
2000
Figure 13. Example of a lidar scan profile used to monitor PM concentrations around and emissions from
conventional tillage operations in Field 4. Each data point represents a 0.5 second averaging time, therefore
data point 1000 was taken at time = 500 seconds.
                 100

                  90

                  80

                  70
                  60
              "   50
              ff
              Q   40
                  30

                  20

                  10

                   0
                                           I
                                 500
            1000
           Data Point
1500
2000
Figure 14. Example of a lidar scan profile used to monitor PM concentrations around and emissions from
conservation tillage operations in Field 5. Each data point represents a 0.5 second averaging time, therefore
data point 1000 was taken at time = 500 seconds.
SDL/08-556
California 2008 Tillage Campaign
                                   36

-------
3.2    DISPERSION MODELING SOFTWARE
The  U.  S. EPA has approved a number  of air dispersion models  for  use in regulatory
applications. These are listed in Appendix W of 40 CFR Part 51 [49]; included are the Industrial
Source Complex Short-Term Model,  version  3  (ISCST3)  and the American  Meteorological
Society/Environmental  Protection  Agency  Regulatory  Model  (AERMOD),  which  as of
November  2005 is recommended for all regulatory applications [1][2]. Both models assume
steady-state conditions,  continuous emissions, and conservation of mass. The default, and
minimum, time-step available for both models is one hour.  Therefore, all meteorological input
data represent one-hour averages.
ISCST3 assumes a Gaussian distribution of vertical and crosswind pollutant concentrations [50].
The Gaussian  plume equation uses the Pasquill-Gifford horizontal and vertical plume spread
parameters, oy and erz, respectively, shown in Eq. 21.


            10            _         _

Cio is the ten-minute average concentration  (ug/m3),  Q is the emission rate (ug/s), UGM is the
average wind speed at release  height (m/s), y is the horizontal distance of the chosen receptor
from the  centerline of the plume (m), z is the height of the receptor above ground level (m) and H
is the effective stack height (m), which includes estimates  of plume rise due to buoyancy and/or
momentum [50].
ISCST3  assumes a Gaussian distribution  of pollutants based on time-averaged meteorological
data. It also uses stability classes to address pollution  dispersion due to atmospheric  mixing.
Stability  classes are typically determined by a combination of vertical temperature lapse rates
and incoming solar  radiation  or methods using vertical or horizontal  wind variance [51].
AERMOD requires more detailed meteorological and surface characteristic information. Because
of the additional input requirements for AERMOD and the  lack of an established database for
these inputs, many regulatory agencies continue to use ISCST3.
The suite of meteorological instruments employed  during this field study allowed us to use both
models   in  this  study.  AERMOD  uses   continuous  functions  for   atmospheric  stability
determinations, and based on  stability  determines the appropriate distribution:  a Gaussian
distribution for stable atmospheric conditions, and a bi-Gaussian distribution for  unstable, or
turbulent conditions.  AERMOD is also better at  accounting for terrain  features and building
downwash phenomena than ISCST3 [52].  The commercial interface used to run the models was
ISC-AERMOD View packaged by Lakes Environmental, Inc. (Waterloo, Ontario, Canada).
Final  emission rates were determined using inverse modeling coupled with observed facility-
derived pollutant concentrations. In  inverse  modeling,  the  downwind  impact on pollutant
concentrations by a source is known while the emission rate is unknown. To solve for the source
emission rate,  a model such as ISCST3 or AERMOD is run with the following inputs:  on-site
collected meteorological data, the facility layout, the locations of pollutant sources and receptors
(samplers), and an estimated or "seed" emission  rate, which can be obtained from  literature.
Observed facility-derived  concentrations  are calculated by subtracting measured  background
levels from concentrations measured at  locations impacted  by the source plume. Modeled
concentrations are then compared to the facility-derived  concentrations from both MiniVols and
OPCs  at each  sampler  location. The location-specific ratio of the measured concentration
SDL/08-556                      California 2008 Tillage Campaign                             37

-------
          to the modeled concentration (Cmodekd) is multiplied by the seed emission rate (Eseed)
and  an average  across  all valid locations is  calculated  to yield  the source emission  rate
corresponding to the measured facility-derived concentrations (Eestimated) as shown in Eq. 22.

                                               (c      \
 '                                F       F   \  measured                               (22)
                                 ^estimated  *-*' seed\ ^                                    \^^>
                                               \  modeled j

A seed emission rate of 8.6 ug/s-m2 was used in the dispersion models. This value was an
average of preliminary emission  rates found through  inverse modeling in the 2007 fall tillage
CMP study that is a companion to the current study [7].
It should be noted that evaluations of air dispersion model accuracy have shown that models are
better at predicting concentrations over longer averaging times than shorter time periods at a
specific location; such models were developed and optimized  to predict longer-term averages
and do not incorporate all of the many temporally and spatially variable factors that may affect
dispersion in the atmosphere.  Models  can predict the magnitude of the highest  concentration
reasonably well, with a typical range of error of ±10% to 40%, but  do not reliably predict the
exact location and time of the  highest value. Measured and modeled  concentrations at the same
location are usually poorly correlated, which is likely due to  a combination of uncertainties in the
input data and unquantifiable uncertainties within the model. The uncertainty of the input data
(e.g., meteorological data, emission rates, source and receptor locations, topography) potentially
may be reduced through careful collection and screening. In addition, the deposition and particle
removal  mechanisms are  limited in the ISCST3  and  AERMOD  models  herein employed.
Insufficient  correction within  a  given model  for these  and other  processes that  decrease
downwind pollutant concentrations will lead to an underestimation of emission rates  based on
measured downwind impacts.
An example  of errors due to  uncertainty in the input  parameters is that concentration  errors
between  20% and 70% can result from an uncertainty  of 5° to 10° in the wind  direction that
directly affects plume location, depending  on  atmospheric stability and  the sampler/receptor
location. Uncertainty within air dispersion models, called "inherent uncertainty," is mostly due to
the  simplification  of complex  and  highly variable  processes affecting  dispersion in  the
atmosphere. If atmospheric conditions that are used as inputs into the model  (e.g., wind speed,
wind direction, mixing height) are consistent across multiple sample periods, the model would
predict the same  concentration, while measured concentrations could vary significantly due to
variability in the complex processes that are  not directly  accounted for in  the model. These
inherent uncertainties can produce predicted ground level concentration errors of up to 50%. A
more detailed discussion of air  dispersion model uncertainty and accuracy  is presented in
Appendix W, 40 CFR 51  [49]. The report "Air Emissions from Animal Feeding Operations:
Current Knowledge, Future Needs" by the National Research Council of the National Academies
states that due to the assumptions required with Gaussian dispersion models, the uncertainty
associated with predicted concentrations are not  smaller than ±50%. Additional uncertainty is
introduced by stability classification and sample instruments [53]. However, the placement of
near ground-level receptors along  the downwind side of large, ground-level area sources, such as
the setup in this study, potentially may reduce the  inherent uncertainty of the  predicted
concentrations due  to both the vertical and horizontal proximity of the receptor to the source.
Unfortunately, no discussion  of  uncertainty in  dispersion models  under such  conditions is
available.
SDL/08-556                       California 2008 Tillage Campaign                             38

-------
Based on the issue of potentially reducible uncertainty, all of the input data for both models were
very carefully screened to reduce uncertainty in the output to the maximum extent possible, as is
subsequently described. Even with such efforts, the error in the model predicted concentrations
for this study is expected to be ±50% according to the above sources, which, when combined
with a ±20% sampling error of the  difference between upwind and downwind MiniVol PM
samplers (±10% each) in Eq. 22, yields a range of error about the calculated emission rate at a
single location between -46% and ±140%. Averaging valid emission rates per sample period may
reduce the potential error range to -33% to ±100% by removing instrument sampling error.
The meteorological data were carefully screened and corrected for errors prior to preprocessing
for the dispersion models. It was this  same screening process that uncovered the erroneous wind
direction averaging code for the meteorological towers in the analysis of the 2007 fall tillage
CMP study [7]. As previously described, we have instead employed the horizontal,  one-minute
averaged wind  directions from sonic anemometers.  The meteorological  inputs for  the models
generated by the preprocessing programs were also screened for consistency with input measured
values; adjustments were made as necessary. The effects of the uncertainty associated with the
meteorological input, however small, combined with the inherent uncertainty within the model
most greatly impacts model-predicted concentrations at receptor locations that are near the edges
of the  predicted plume, which in turn may greatly impact the  calculated emission  rates. Arya
(1998) suggests that the plume edge be defined as 10% of the maximum modeled concentration
to minimize these effects [54]. Therefore,  emission rates calculated at locations with predicted
concentrations  less  than 10%  of the maximum predicted  concentration  were not used  in
calculating the average  emission rate.
AERMET, the preprocessor of meteorological data for AERMOD, requires that the surrounding
land use and land cover be categorized to quantify the Bowen  ratio, surface roughness, and
midday albedo.  For this study, the land use on all sides of the  site was classified as cultivated
land and the default spring time values of midday albedo (0.14), Bowen ratio (0.30),  and surface
roughness (0.03 m) were used because the summer values for cultivated land assume vegetation
cover, which was not the situation for the majority of measurement periods. During  each of the
sample runs, the sky was clear of clouds, so the amount of cloud cover was set to 0.0 for all
hours. The mixing height for input into RAMMET and AERMET was set at 1000 m because all
samples were started at least two hours after sunrise and ended before sunset. In addition, all the
receptor locations of interest were on the  southern edge of the field so the  exact depth of the
mixing layer during daylight hours over such a short distance at ground level was not considered
to be a significant factor. Based on the measured incoming solar radiation, vertical temperature
lapse rates,  and surface wind speed, stability classes for ISC during all sample periods were
determined to be slightly unstable to very unstable. The Upper Air Estimator  in AERMET View
was used to calculate required upper air parameters based on observed surface conditions.
Digital elevation model files based on 7.5 minute topographical maps  were used to  calculate
receptor and source elevations. The terrain was not considered to be a significant factor in the
modeled concentrations as the change in elevation over the entire  modeled domain (~2 km x 2
km) was gradual and no greater than  2 meters. The areas tilled during the sample periods were
represented in the dispersion models by ground level area sources that varied in size and shape.
The readings of the handheld GPS tracking device located on Tractor 1  for each tillage operation,
mentioned in Section 2.2 were used to develop the source areas and shapes. The GPS  readings of
sample locations were used to specify discrete receptors for comparison between modeled and
SDL/08-556                      California 2008 Tillage Campaign                             39

-------
measured concentrations at specific locations.  In addition, a receptor grid with  15 m x 15 m
spacing and a flagpole height of 2.0 m above ground level was set up slightly upwind of, over,
and downwind of the area source to provide a visualization of the modeled particulate  matter
concentrations resulting  from the area  source  emissions. Elevations for receptors and  source
areas were calculated and assigned by AERMAP. Table 30 in Appendix A provides more  details
about the settings used in each model. All concentrations used for emission rate calculations and
presented in this report are averages over the modeled periods.

3.3    STATISTICAL ANALYSIS OF DATA
The goal of this investigation  is to determine if there are statistical differences between  the
conventional and conservation tillage practices.  Many different instruments and techniques were
used  during  this experiment;  some techniques  exhibit  Gaussian  distributions  and  some
instruments  exhibit Poisson distributions. Instruments that count  events (photons or particles)
exhibit Poisson statistics. The specific "Poisson" instruments are the OPC and Aglite. All other
instruments  exhibit Gaussian statistics. Regardless of the kind of statistics a particular instrument
might use,  the  same general principle  is  employed, which is  based on  comparison  of  the
maximum error (as computed through confidence intervals and T-values) for conservation and
non-conservation tillage processes. The respective maximum errors for two tillage processes are
compared using a simple T-test for the two time intervals. The null hypothesis is "there  are no
differences  in the particulate emissions between conventional  and  conservation management
practices."
Soil samples were collected over the entire field of study and analyzed using Gaussian statistics.
The respective averages  and confidence intervals were calculated for each region of the field.
Since no statistical difference was found among regions of the field, global  statistics for  the
whole  field were calculated and  used to describe the entire  field.  In  the case of the OPC
instruments,  the  OPCs generate a  particle size distribution every  20  seconds for a  tillage
operation lasting several hours. These data were first manually examined to remove data artifacts
such as  unrealistic high  particle  counts or partial  distributions.  Then, all of the particle size
distributions for  the entire tillage process were averaged and  the  variance and  confidence
intervals calculated according to their corresponding Poisson definitions.
The process of extracting particle volume and mass concentrations from the  Aglite lidar return
signal is an  iterative process that involves several linear least squares regressions. Our process
for retrieval is to average  the returns for an entire tillage process, then run the concentration
extraction regression. Typically, several million laser shots are averaged prior to calculation of
concentrations.
A challenge in dealing with highly variable discrete events, such as agricultural tillage operations
is that PM emissions from each tractor/implement pass can - and does - vary considerably from
the previous tractor pass  across the field.  This leads our raw data to  appear "noisy," though in
fact the error bars on any given measurement are typically much smaller, usually by a factor of
five or more, than the natural variability in the instantaneous emissions from the tillage process.
The aggregation of the particulate data, lidar and meteorology data provide  an estimate of the
particulate flux within a field. This data will be expressed  as cumulative particulate emission for
a specific portion of the tillage operation. Then, differences between systems will be compared
using Gaussian mean comparison methods.
SDL/08-556                       California 2008 Tillage Campaign                              40

-------
                       4.     RESULTS AND DECISIONS
4.1    GENERAL OBSERVATIONS

4.1.1     Soil Characteristics
The average bulk density and soil moisture values, with la and the number of samples collected
(«), are presented in Table 13. Figure 15 provides a map of soil sample locations in both fields.
The average bulk densities (± la) for Fields 4 and 5 were 1.57 ± 0.05  g/cm3  and 1.57 ± 0.08
g/cm3, respectively. Similarly, the soil moisture did not differ significantly between the two
fields.  Soil moisture generally did not change with increasing number of tillage operations. The
largest in-field differences expectedly were recorded after the irrigation event at the end of May
2008 (see Table 13).
Analysis performed on samples from Field 4 yielded an average soil composition of 62% sand,
29% silt, and 9% clay. For Field 5, the composition was 67%  sand, 23% silt, and  10% clay.
Stable aggregate results are presented in Table 14 and show that the mass of stable aggregates
decreased after the first tillage operation with the greatest decrease in the largest sieve. Since the
bulk density and the  soil moisture data from the two fields are essentially the  same within  the
error of the measurement, we expect these fields to have similar characteristics for aerosol/dust
generation.  Similarly, as has been shown in the literature [10][11][12][16], we expect the soil
moisture content to strongly influence the amount of aerosol/dust production from  any given
tillage operation.
Table 13. Statistics of soil characteristics measured for both fields.
Bulk Density Data Summary
Field 4 (Conventional Tillage)
Field 5 (Conservation Tillage)
Soil Moisture Data Summary (Date)
Field 5 (5/17/08)
Field 4 (5/17/08)
Field 4 (5/19/08)
Field 4 (5/20/08)
Field 4 (6/5/08)
Field 5 (6/7/08)
Mean (g/cm3)
1.57
1.57
Mean (%)
1.3
1.0
1.7
3.3
6.1
8.1
o (g/cm3)
0.05
0.08
o (%)
0.5
0.6
0.7
1.6
2.2
1.6
n
10
10
n
10
10
10
10
12
10
95% CI
0.03
0.05
95% CI
0.3
0.4
0.4
1.0
1.2
1.0
SDL/08-556
California 2008 Tillage Campaign
41

-------
Table 14. Stable aggregate analysis results for both fields.
Sample collection
Sieve Size (mm)
Mass per
sieve pan (g)
Total Stable Aggregate Mass
out of 100 g analyzed (g)
Conventional Tillage, Field 4
Prior to tillage
After Disc 1
After Disc 2
After flood irrigation
4
2
1
0.5
0.25
4
2
1
0.5
0.25
4
2
1
0.5
0.25
4
2
1
0.5
0.25
31.41
11.38
9.67
10.48
12.16
4.1
6.25
5.2
15.74
19.51
12.92
8.75
15.64
11.62
9.27
12.88
11.55
6.16
14.28
12.02




75.1




50.8




58.2




56.89
Conservation Tillage, Field 5
Prior to tillage
After flood irrigation
4
2
1
0.5
0.25
4
2
1
0.5
0.25
24.84
10.63
6.57
12.58
11.68
7.23
10.87
8.64
13.87
16.72




66.3




57.33
SDL/08-556
California 2008 Tillage Campaign
42

-------
Figure 15. Soil sample collection locations in fields under study.

4.1.2     Meteorological Measurements
A summary  of the meteorological characteristics recorded during the study period are presented
in this section with some examples.
                                                                               .th
4.1.2.1        Precipitation Event
One precipitation event occurred during the break in measurement periods from May 20m to June
5* to allow for flood irrigation of both fields. The proximity in time of the precipitation event to
flood irrigation of the fields suggested that the effects of flood irrigation on soil conditions and
PM emissions would mask any effect of the precipitation event. A minimum of ten days passed
between the completion of irrigation on the study fields and any tillage activity.

4.1.2.2       Hourly Wind Data
One-minute averaged wind data collected at the meteorological tower at the WM location from
May and June 2008 were averaged again to hourly values in order to create a wind rose of actual
conditions during the field experiment (see Figure 16) for comparison with the wind rose created
for May and June 2005-2007 (see Figure 6) from the CJJVIIS station located near Stratford,  CA.
The wind speed measured at 2.5 m agl on the tower was used because the wind measurement for
the CJJVIIS station  appeared to be at the 2-3 m agl  level.  The wind  rose in Figure 16 closely
resembles the  data collected by the CIMIS station for the previous three years. Wind conditions
were very favorable for the designed sample layout throughout the duration of the field study.
SDL/08-556
                               California 2008 Tillage Campaign
43

-------
                                                       NE
           SW
                                        ,                     .
                                   \ /10.0% :
                                             ,20.0%;
                                                    30.0% /'
                                                          •'400%
                        SE
Figure 16. Wind rose of wind speed and direction measured during the May-June 2008 campaign

4.1.2.3       Period-Average Meteorological Conditions
Meteorological characteristics were monitored on-site throughout the field study. Sample period
average conditions were calculated and are presented in Table 15 based on measurements taken
at the WM and EM locations. As can be seen from these data, all measurements were made
during warm and dry conditions. Winds were consistently out of the northwest  with  average
speeds between 2 and 6 m/s.
SDL/08-556
California 2008 Tillage Campaign
44

-------
Table 15. Average ± lo temperature, relative humidity, wind speed, and wind direction for each sample
period as measured at the WM tower. Temperature, relative humidity, and wind speed were measured at 9.7
m agl and wind direction was measured at 11.3 m agl.
Sample
5/17 Run 1
5/17 Run 2
5/18
5/19 Run 1
5/19 Run 2
5/20 Run 1
5/20 Run 2
6/5 Run 1
6/5 Run 2
6/7
6/11*
6/18
6/25*
Ambient Temperature (C)
32.3 ±2.1
36.8 ±0.2
33. 8 ±2.8
31.4 ±2.5
35.3 ±1.5
29.1 ±2.2
32.5 ±0.4
24.7 ±2.6
27.6 ±0.5
22.5 ±2.7
29.1 ±0.1
34.1 ±0.3*
30.2 ±2.5
Relative Humidity (%)
33 ±4
24 ±0.3
29 ±4
27 ±3
21±3
30 ±10
23 ±1
34 ±7
26 ±2
40 ±9
19 ±1
16 ±1*
29 ±5
Wind Speed (m/s)
3.6 ±0.6
4.3 ±0.6
4.3 ±1.2
2.9 ±0.8
3. 3 ±0.5
5.1±1.1
5. 9 ±0.8
3. 3 ±1.3
4.0 ±0.9
4.0 ±1.0
3. 8 ±0.6
5.6 ±0.7
1.9 ±0.8
Wind Direction (°)
321 ±15
321 ±8
325 ± 16
318 ±22
319 ±16
320 ± 10
317±8
320 ± 30
315±7
335 ±20
326 ± 17
326 ±4
328 ± 29
* Data taken from EM tower due to missing data at WM tower

4.1.2.4       Wind Profile Calculations
The wind profiles used in calculating the PM mass fluxes through the vertical lidar beam planes
were determined by fitting the power law wind speed equation (Eq. 5) to cup anemometer data
from the meteorological tower located at location WM. In fitting the measured data, calculated/?
values ranged from the minimum to the maximum designated limits of 0.1 and 0.6, respectively,
and followed a diurnal cycle similar to that of atmospheric stability class; the mean value of p
throughout the field study ± la was 0.25 ± 0.13.An example of a calculated wind speed profile
based on tower-mounted cup anemometer data  and a power-law fitting model is presented in
Figure 17.
SDL/08-556
California 2008 Tillage Campaign
45

-------
             250
             200
             150
             100
              50
                        Generated Wind Profile
                        Cup Anemometer Wind Speed
                                     »--«*
                        2.5
             3.5
      Wind Speed (m/s)
4.5
Figure 17. Cup anemometer measurements shown with the wind speed profiled calculated using the
measured wind speed.

4.2    AEROSOL CHARACTERIZATION DATA
The following section contains the aerosol data collected using both the filter and optical sampler
methods.

4.2.1     Minivol Filter Sampler Data
As mentioned in Section 3.1.4.1, an array of Airmetrics MiniVol samplers was positioned to
characterize the upwind/background and downwind PM concentrations. A total of 296 samples
were collected: 116 (39%) PM2.5, 116(39%) PMio, and 64 (22%) TSP. Calculated PM2.5
concentrations based on filter catch ranged from 23.2 to 3244.9 ug/m3; PMio concentrations
ranged from 38.1 to 1458.4 ug/m3; TSP concentrations ranged from 73.6 to 2276.9 ug/m3. The
MDL for each sample period was calculated based on the average run time of the MiniVols, the
targeted flow of 5.0 L/min, and the minimum filter catch that could be measured in the difference
between the pre- and post-test filter weights of 5 ug. The MDL for each run varied based on
different sample times; the average MDL ± 1 o (n = 13) was 6.6 ± 4.9 ug/m3 and the median was
4.3 |-ig/m3, with a range of 2.3 for a run length of 7.3 hours to  17.3 ug/m3 for a run length of 1.0
hour. The calculated PM concentrations measured at all sample locations during May 2008 may
be found in Table 31 and during June 2008 may be found in Table 32 in Appendix A.  Figure 18
presents a contour plot of the measured PMio concentrations at 2 m agl for the June 25 sample
period across the field of study. Sample locations are shown by (+) markers and the field edges
are approximated by the black lines. Note the background PMio concentrations were around 100
Hg/m3.
SDL/08-556
California 2008 Tillage Campaign
                      46

-------
          1200
                                                                      (Mg/mJ)
                                                                          180
                                                                          160
                                                                          140
                                                                          120
                                                                          100
           700	
                 900   950  1000  1050  1100 '1150  1200  1250  1300
                                      E(m)
Figure 18. Contour plot of measured PM10 concentration for the June 25 sample period.
Of the 296 filter samples collected, 165 did not pass quality  analysis/quality control (QA/QC)
steps  applied to the dataset, leaving  131  for use  in calculating emission rates using inverse
modeling. An investigation  into this high rate of failure was conducted and a detailed description
is provided in Appendix B.  In summary, filters that did not pass QA/QC were suspected to have
been contaminated either during sampling or during storage and handling. Evidence of "particle
bounce"  was found on many PIVb.s and PMio samples  collected during May  sample periods.
Particle bounce occurs when particles that collide with the impactor plate, the mechanism used
by the MiniVols to exclude particles larger than the design size, are re-entrained in the airstream
and collected on the filter downstream and result in  higher measured levels than actually existed.
This issue is most likely due to exposing the  MiniVol samplers to  dust plumes exceeding the
maximum recommended exposure level and  improper  instrument  maintenance and  cleaning
through the May sample periods. Corrective action was taken during the June sample periods and
no issues associated with particle bounce were observed in the second portion of the study.
Additionally, some particles were observed on top of and  imbedded into the plastic annular ring
around the Teflon filter material - the plastic ring is covered by the filter holder assembly during
deployment. This was likely due to contamination during  on-site filter storage or handling.
Efforts were made to minimize this issue throughout, especially during the June sample periods.
However, windblown dust did impact the handling and storage area during the last sample
SDL/08-556
California 2008 Tillage Campaign
47

-------
periods in May. The size fraction distribution of approved filters was nearly identical to the total
sample set: 51 (39%) were PM2.5, 50 (38%) were PMio, and 30 (23%) were TSP. These finalized
concentrations are given in Table 33 and Table 34 in Appendix A. See Appendix B for a detailed
discussion of the QA/QC steps, filter inspection failure, possible causes of the failures, and
preventative solutions for future sampling.

4.2.2     PM Chemical Analysis

4.2.2.1       Organic Carbon/Elemental Carbon Analyzer
As   mentioned  previously,  the  organic  carbon/elemental   carbon  analyzer   passed  the
manufacturer's suggested in-field audits and after completion of the field project the data were
manually screened for completeness and potential outliers.  During the  two distinct periods of
sampling, May  13-20 and June 3-21, 2008, the EC/OC instrument operated continually except
for brief periods for QA/QC checks, servicing of the  system generator, or significant breaks in
the producer operations. An unanticipated consequence of the one hour sample times, coupled
with  the  dual  channel  operation  of  the  R  P  5400  EC/OC  Analyzer,  was that the
sampling/analysis/cleaning cycles extended beyond the planned two hours.  The net result was a
sampling profile wherein every third hour of data was missing (66% sample collection efficiency
over the entire deployment).  However, because the actual farm practice periods varied from one
to eight hours, the actual, observed data periods  ranged  from one to eight hours,  the data
coverage was from 50-100%, averaging 78.2% ± 10.3%.
The  PM2.s  OC/EC time series data collected at the downwind Air Quality Trailer (AQT) are
shown in Figure 19, with  the shaded sections indicating the discrete sampling periods for the
observed tillage practices.  As can be seen, the PM2.s-associated EC was typically quite low,
ranging from 0.3  ng/m3 to 0.7 |-ig/m3  and  averaging 0.4 ng/m3  ±0.1  ng/m3  (at the 95%
confidence interval) during the concurrent tillage sampling periods and demonstrated relatively
little variability.  During these same sample periods, the total PM2.5 at the AQT, measured by the
filter-based MiniVol systems, ranged from 30.8 ng/m3 to 141.6 ng/m3 and averaging 64.1 |ag/m3
± 26.3 |-ig/m3 (at the 95% confidence interval).  For further comparisons, during  the non-test
periods the EC PM2.5  concentrations  averaged 0.8  ng/m3 ±0.1  ng/m3 (at the 95% confidence
interval).  This would  suggest that the EC fraction  of the PM2.5 at the  site was not greatly
impacted by the examined tillage practices or by typical EC sources such as biomass or diesel
combustion and is more likely of a regional phenomenon.
The  observed OC PM2.5 concentrations, derived by multiplication of the  raw OC concentrations
by 1.7 (refer back to Section 3.1.4.3), were approximately an order of magnitude greater than the
EC concentrations, ranging from 2.8 ng/m3 to 9.2 ng/m3 and averaging  5.6 ng/m3  ±1.3 ng/m3
(see  Figure 20). Although the PM2.5 organic component seemed to vary  more and occasionally
show greater concentration spikes than the EC, these episodes generally occurred during non-test
events, when the OC PM2.5 averaged 6.0 ng/m3 ±1.3  ng/rn3, statistically indistinguishable from
the tillage test periods.
SDL/08-556                      California 2008 Tillage Campaign                             48

-------
              -Elemental Carbon
              -Organic Carbon
              -Total Carbon
      o
     5/13/08    5/14/08    5/15/08    5/16/08     5/17/08    5/18/08     5/19/08    5/20/08
                                                                                  5/21/08    5/22/08
                                                                                                     5/23/08
     18
             -Elemental Carbon
             -OrganicCarbon
             -Total Carbon
      o
     6/3/2008   6/4/2008    6/5/2008   6/6/2008    6/7/2008   6/8/2008   6/9/2008   6/10/2008   6/11/2008  6/12/2008   6/13/2008
Figure 19. PM2.5 OC/EC time series concentrations as collected at the downwind AQT location. The shaded
sections indicate the observed agricultural practices. It should also be noted that the raw instrument OC
concentrations have been multiplied by 1.7 to account for potential non-carbon functional groups.
SDL/08-556
California 2008 Tillage Campaign
49

-------
I
a
3: 10
1/1
C
o
£ s
Material Co nee
4^ 01
C
O
(5 2
u
s"
a- 0

/Total ft Iter-b a se d P M? 5 fug/m5 )
- - 141 S
53,9 XHX.P

P
























^_











65.1 34.5







-













3S.2
-




























• Elemental Carbon
D Organic Carbon








52.3
PI






* -*** <^ <^
fc°

-------
18

16 -

M
•H
ns
s "-
rJ
S 12
Q. C
•a o
 c 8
•S3
5s g
D Organic Carbon .

















C
O
re 4
u
2













• Elemental Carbon

































••










































-r
• •
AvgField4 A vg Fie Id 3

























Figure 21. PM2.5 organic matter and elemental carbon concentrations during specific sampling periods
(parallel to filter-based sampling).
The cations observed included sodium (Na+), ammonium (NH4+), potassium (K+), magnesium
(Mg2+), and calcium (Ca2+). The resolved  anions included fluoride (F), chloride (Cl~), nitrite
(NCV), and nitrate (NCV). It should be noted here that sulfate (SC>42~) is often an expected anion
present in common ambient particulate; unfortunately, chromatographic failures prevented clear
resolution of the sulfate peak so no values for SC>42~ were obtained. However, similar studies by
the project investigators found central California tillage dust contained low levels of SC>42~,
averaging less than 4% of the total observed soluble ions and around 0.5% of the  total PM2.5
mass [7]. Aside from the sulfate, all of the  expected ions where observed with the exception of
fluoride, which was not detected (n.d.) in any of the  samples. As can be derived from Figure 22,
the observed soluble ions contributed around 10%  of the total filter PIVb.s mass. On average,
chloride was the most  common anion and ammonium and magnesium were the most common
cations. It  should be noted that these  dominances were not always significant at the 95%
confidence level. Table 16 presents the average ion concentrations across the seven analyzed
filters (uncertainty represents the 95% confidence interval).
Figure 23  shows the average  mass percentages for the individual  ions, along with the 95%
confidence interval. As can be seen, the  average individual ionic species concentrations rarely
approached 2% of the total PM2.5. Although not explicitly shown in either Figure 22 or Figure
23, the total observed ionic percentage can be derived and averaged 10.3% ± 3.8%.
As might be expected, if the presumed  sources  of the combustion and secondary particles are
more regional  in nature, the ionic species would  show very little mass  concentration  differences
across  the samples. When the ionic contributions observed within this study are combined with
the EC and OC contributions, the bulk  of the  particulate material, 81.8% ± 6.1%, was  still
contributed by unknown (unanalyzed) constituents (see Figure 24).   These unaccounted for
compounds are most likely insoluble  crustal elements associated with the background aerosol
and expected soil disturbance emissions during the tillage processes.
SDL/08-556
California 2008 Tillage Campaign
51

-------
       12
       10
   i  8
^
_3

O
in
                             Total filter-based PM2.5 (ng/m3)
                                                               52'3
        6 -       /      141.6       J"


               /       R      R
                                           38.2
                                        s
                                                     34.5
                                                                        I NO3-    • NO2-




                                                                        1 c-      D Ca+2




                                                                        ...i-     HNH4+
ICI-




IMg+2




INa+
                                                                         38.2
Figure 22. Soluble ionic mass concentrations of AQT downwind PM2.5 filters.
Table 16. PM2.S filter ion concentrations averaged over seven downwind samples collected at AQT

(uncertainty represents the 95% confidence interval).
Analyzed Ion
Mass
Concentration
F
(Hg/m3)
ad.
cr
(Hg/m3)
1.1±
0.5
NO2
(Hg/m3)
0.8 ±
0.2
NO3"
(Hg/m3)
0.2 ±
0.1
Na+
(Hg/m3)
0.7 ±
0.2
NH4+
(Hg/m3)
0.8 ±
0.1
K+
(Hg/m3)
0.4 ±
0.2
Mg+2
(Hg/m3)
0.8 ±
0.6
Ca+2
(Hg/m3)
0.5 ±
0.3
SDL/08-556
                                  California 2008 Tillage Campaign
          52

-------
             Na+
                      NH4+
                                 K+
                                         Mg+2
                                                   Ca+2
                                                                       Cl-
                                                                                NO2-
                                                                                          NO3-
Figure 23. Average soluble ionic mass percentage composition of the AQT downwind PM2.5 filters.  The error
bars represent the 95% confidence interval.
                                                                        DOC
                                                                        • EC
                                                                        DNa
                                                                        • NH4
                                                                        • K
                                                                        • Mg
                                                                        DCa
                                                                        • F
                                                                        • Cl
                                                                        • N02
                                                                        • N03
                                                                        D unkown
Figure 24. Average compositional mass percentage of the AQT downwind PM2.5 filters.
SDL/08-556
California 2008 Tillage Campaign

-------
4.2.3     Aerosol Mass Spectrometer
During the tillage experiment, the AMS acquired chemical composition data from May 14-May
19 with some significant gaps in the data due to mass spectrometer malfunctioning. This required
manual  screening of the AMS data. Approximately 35% of data over the time period was found
to be valid.  Large  gaps occurred throughout the sampling period with,  for example, no data
acquired on May 17. Similar to the AMS data acquired in the  companion experiment in Los
Banos the previous  year [7],  the OC  fraction  dominated the detected  particle  chemical
composition (Figure  25).  The mass  spectrum shows that PMi is  mostly dominated  by the
presence of photochemical  regional pollutants,  such  as  ammonium nitrate  (NFLiNOs) and
ammonium sulfate  ((ML^SO/O salts and OC, as  in  Figure  26. Particle size information was
obtained (Figure 27 a and  b)  and is indicative of photochemical production  of particles with the
size distributions showing  a mode at approximately 0.6 jam (600 nm).
                                                                   I Organic

                                                                   I Nitrate

                                                                   I Sulfate

                                                                   I Ammonium

                                                                   I Chloride
Figure 25. Average chemical composition of particles detected by the AMS from Mayl4-15.
        N
        CO
        (Z
        c
        o
                              30-	^Nitrate   Su,fate
                                                        5/14/2008 0:00-6:00 AM
                                                                120
                                        140
Figure 26. Representative AMS mass spectrum of particles detected during the study. Mass-to-charge (m/z)
ion assignments include nitrate (m/z 30 NO+ and m/z 46 NO2+), sulfate (m/z 48 SO+, 64 SO2+, 81 HSO3+, and 98
H2SO4+), and carbon (e.g. m/z 55 C4H7+, 57 C4H9+, 77 C6HS+, 91 C7H7+).
SDL/08-556
California 2008 Tillage Campaign
54

-------
Because of the major gaps in the AMS data, there were few and limited time periods where the
AMS and Met One OPC data overlap. A brief six-hour overlap period is shown in Figure 28
where AMS mass concentration data accounts for -80% of the converted mass concentrations
for the 1  jam  size bin of the OPC at  the  AQT sampling site.  Since the AMS does not detect
refractory components of the aerosol (e.g., inorganic oxides and black/elemental carbon) and it
measures  vacuum aerodynamic diameter versus the OPC that  measures optical diameter, it is
expected that  the AMS would detect less  mass concentration than the OPC.  The overall trend,
however, for the time period is consistent between the two instruments.
                  12:00 AM
                  5/14/2 DOS
     3:00 AM
                                Date and Time
6:00 AM
     (b)
                 5/14/2008 2:00-4:00 AM
                                                      3   456789
                                                                  1000
                             Vacuum Aerodynamic Diameter (nm)
Figure 27. (a) Image plot of nitrate particles on May 14,2008 using m/z 30 (NO+). This shows the formation
of a mode of nitrate particles in the early morning hours of 5/14/2008, with peak mass concentration at ~3:00
AM Pacific Standard Time, (b) Integrated size distribution of nitrate particles from 2:00-4:00 AM on May 14,
2008 showing the peak in the mass distribution at ~0.65 um.
SDL/08-556
California 2008 Tillage Campaign
                                 55

-------
           8 -
         C
         •2 6
         re
         	
         c
         0)
           4 -
         o
         U
         (O
         ro 2
                      T
                   8:00 AM
                   5/15/2008
       T
                                             -r
T
                                                                      - 10
                                          8   §>
                                             o
                                             o>
                                             3_
                                          6   3
                                             o'
                                             3

                                          4   5
                                             3
                                                                     i- 2
    10:00 AM         12:00 PM

      Date and Time
  1
2:00 PM
Figure 28. Comparison of AMS PMi and OPC PMt (assuming a MCF of 1.0 g/cm3) data for the morning of
May 15,2008

4.3    OPTICAL CHARACTERIZATION DATA
Several  different  optical  instruments were used to characterize the airborne particles, both
background and those emitted by the operations under study. Samples of the results from each
instrument are presented below, with emphasis on the lidar results.

4.3.1      Met One Optical Particle Counter
The collected OPC  data  were  analyzed  for particle size distribution, particle  volume
concentrations, and converted to particle mass concentration through multiplication with  the
MCF, as described in Section 3.1.4.2. Table 35 and Table 36 present the PM2.5, PMio, and TSP
concentrations  as  reported by the  OPCs. Three to four  OPCs were  in  positions immediately
downwind of the field under study in each sample period, with between one and four OPCs in
upwind  locations. Unlike the downwind MiniVol samplers,  the  downwind OPCs were  not
overwhelmed by the dust plumes from the tillage activities - the manufacturer specified range of
the OPC of 0 to 318,000,000 particles/m3 was never exceeded - and  thus provided usable data
throughout all sample periods. Upwind OPC time series data were examined for contamination
from activities upwind of the site, such  as unpaved road traffic. Contamination was found in six
of the 12 sample periods,  with five  of those occurring at the  10.0  sample site that was
immediately downwind  of an  unpaved road  (see  Figure 7).  Large spikes indicative  of
contamination were removed  from the upwind OPC time series data in these instances to
estimate the background aerosol concentration;  the estimated background  levels were in very
good agreement with those measured by an OPC at a different, uncontaminated upwind location
(see Table 37 and Table 38).
Data completeness for the OPC datasets was calculated as a ratio of the number of valid samples
per sample period  over the possible number of valid samples and expressed as a percentage. Data
completeness was less than 100% due to  communication errors between the OPCs  and  the
computer logging  the data, resulting in lost  packets of 20 second sample data. Communication
error frequency was variable between  OPCs  and  across time. Data  completeness per sample
period ranged from 81.8% to 100.0%, and  averaged (± la) 97.4 ± 3.7%.
SDL/08-556
California 2008 Tillage Campaign
                               56

-------
The distribution of Met One optical particle counters surrounding the fields of interest at two
heights  provided the  ability to  examine particle emissions by  number  and size  at  several
locations, as well as to monitor the PM concentration time series. Examples of particle volume
size distributions measured during tillage operations, in units of |im3/cm3/|im, are presented in
bar graph form in Figure 29 (dV/dd is the change in aerosol volume concentration normalized by
the change in particle diameter). Each graph shows the background particle volume distribution,
the volume distribution of aerosols downwind of the source, and the difference between the two
that is the volume distribution  of the emitted particles. Examining these data based on volume
concentration  causes  the large particles to  have  a  more  visible effect  than if number
concentrations were examined.  However, viewing the data in this way is advantageous because it
is analogous to mass: the relative shape of and  difference between the curves remains the  same
while the scale changes as the transition to mass concentration  is made. Therefore, the given
graphs show that the greatest volume (and mass) of emitted aerosols is in the large particle range
above a diameter of about 2.5 jim. As seen in the reported PM sampler levels, the greatest
contribution by tillage activities was in the PMio and TSP measurements.
Examples of PM time series are shown in Figure 30 from  both upwind and downwind elevated
locations during the May 19 R2 period. This example was chosen because it demonstrates that
downwind samplers were exposed to extremely high PM  concentrations in the tillage plumes,
supporting the  conclusion that downwind filter-based samplers adjacent to the activity  were
overloaded during this and other sample periods. OPC time series from other periods in which
the filter-based samplers were  determined to have been overloaded also show extremely high
concentrations over very short  durations of time. Applying contouring software to OPC number
and PMio concentrations measured around the field under  study yields the graphs in Figure 31.
Data from the June 25 sample period  were chosen for  direct comparison with the filter-based
PMio  concentrations (Figure 18).  The number concentration contour plot presents the number of
particles larger than 1 jam per  liter of air; this is the particle size range containing the greatest
volumetric contribution from agricultural tillage. The OPC PMio concentrations were calculated
by multiplying the Vw concentrations by the MCFio for that sample period. Sample locations are
shown by the (+) markers; there were fewer  OPC sample  locations than MiniVol locations,
leading  to a coarser contour map for OPCs. Field edges  are approximated by the solid black
lines.
SDL/08-556                      California 2008 Tillage Campaign                             57

-------
                      Conservation: Strip-till
                                                                        Conventional: Disc 2
                Diameter (urn)
                                  Diameter (urn)
                       Conservation: Plant
               2.5
                           10
I
                Diameter (urn)
I
                                                                        Conventional: Plant
                                                                            10
I
                                  15
                                  Diameter (urn)
                                                                                    15
Figure 29. Particle volume size distributions measured from upwind (background) and downwind
(background plus emissions) locations, with the difference being the aerosol emitted by the tillage activity: (a)
strip-till operation, conservation tillage method; (b) disc 2 operation, conventional tillage method, (c) plant
operation, conservation tillage method; and (d) plant operation, conventional tillage method.
SDL/08-556
       California 2008 Tillage Campaign
                                                                      58

-------
        500

        450

        400

        350

        soo
                                      Upwind OPC, z = 9m
      Ul
      o 250
      c
      o
        200
        150

        100

         50

          0
a)
                 14:24
                15:36
 16:48
Time
18:00
19:12
      60,000
      50,000
                                    Downwind OPC, z = 9m
                 14:24
                                                            19:12
Figure 30. OPC PM time series, created by multiplying the volume concentrations (Vk) by the daily MCF, as
measured at elevated locations in a) an upwind and b) a downwind positions during the May 19 R2 sample
period.
SDL/08-556
                California 2008 Tillage Campaign
                                             59

-------
                                                                               (#/L)

                                                                                 7600
                                                                                 7200
                                                                                 6800
                                                                                 6400
                                                                                 6000
                                                                                 5600
                                                                                 5200
                                                                                 4800
                                                                                 4400
                                                                                 4000
                            900  950  1000  1050 1100  1150  1200  1250  1300
                                                E(m)
                                                                              (H9/m3
                                                                                  200
                                                                                  190
                                                                                  180
                                                                                  170
                                                                                  160
                                                                                  150
                                                                                  140
                                                                                  130
                                                                                  120
                                                                                  110
                                                                                  100
                                                                                  90
                                                                                  80
                            900  950  1000 1050  1100  1150  1200  1250 1300
                                                E(m)
Figure 31. Contour plots of average OPC a) number concentration (#/L) for particles larger than 1 |im and b)
PM10 concentration (|ig/m3) across the field for the third cultivator pass on June 25.
SDL/08-556
California 2008 Tillage Campaign
60

-------
4.3.2     Optical To PM Mass Concentration Conversion
A key factor in converting the OPC and lidar data from number density (volume) data to mass
concentration is the derivation of the MCF, which was described in Eq.  14. The daily average
MCF from all locations with valid collocated OPC and MiniVol samples are shown in Figure 32
with the corresponding 95% confidence interval (CI) about the mean. As OPC data collected at
each site and during each sample  period passed QA/QC, the calculation of MCF values was
dependent solely on the presence of a valid filter-based PM measurement. While the near-source
downwind filter-based samples for sample periods May 18 through May 20 R2 were determined
to have been overloaded, and thus prevented emission rates being calculated for those operations,
the  upwind and  far-source downwind  samples  were not  compromised and  were  used in
calculating the daily MCF values.
Day-to-day variation  in the MCF  is not well understood. Potential factors leading to  this
variation include,  but are not limited to: 1) changes in background aerosol sources, composition,
and   particle   shape   that   affect  the   OPC/MiniVol    relationship   between   particle
detection/separation; 2) different treatments of particles larger than the sampled PM size k by the
two methods.  If a particle larger than the cut-off size k is present on the filter at the time of the
post-sample weighing, its mass is included and, with sufficient numbers of large particles, would
lead to higher-than-actual PM,t concentrations. The collocated OPC counts particles larger than k
in their respective size bins; therefore, they are not integrated into the V^ value like in the PM,t
concentration calculation. Thus, a higher-than-actual PM,t concentration would be divided by the
actual Vk, leading to a higher-than-actual MCF value. The second potential factor would be most
influential in cases where the PM samplers had been overloaded. However, previous efforts have
removed filter-based samples known or suspected to have been overloaded from emission rate
and MCF calculations.
     45
     40 -

     35 -

     30 -

     25 -

     20 -

     15 -

     10 -

      5 -

      0
                                 •PM2.5 	PM10 	TSP
           5/17    5/18    5/19    5/20    6/5     6/7     6/11    6/18    6/25    Overall

                                            Date
Figure 32. Average daily measured MCF values with error bars representing the 95% confidence intervals.
SDL/08-556
California 2008 Tillage Campaign
61

-------
Most of the PMio and TSP MCF values were within the  expected range of 1.0 to 2.5 g/cm  .
However, the PM2.5 MCF values were much larger than expected, with individual values ranging
from 3.2 g/cm3 to 28.2 g/cm3. The mean ± la was 14.6 ± 3.7 g/cm3 and the median was 10.1
g/cm3 with 25% and 75% quartiles of 6.1 and 17.7 g/cm3.  As previously stated, calculating the
MCF is a simplified method  to account  for  several complex and  possibly  interdependent
variables that affect how an aerosol mixture is measured/detected based on both optical and
aerodynamic properties. Some of the known aerosol property relationships that are incorporated
into the single MCF value are particle density (p), dynamic shape factor (%), index of refraction
(n), and porosity. In past field campaigns, PM2.5  MCF values have generally been slightly higher
than PMio and TSP values, but the PM2.5 MCF values calculated from the MiniVol and OPC data
collected during this field campaign are generally much higher than those seen before in our own
unpublished work. Due to the non-physically large MCF values for PM2.5, the calculated PM2.5
MCF values were not used to convert the lidar volume concentrations to mass concentrations.
By way of example, the density of Ni is 8.9 g/cm3  and Hg  is 13.5  g/cm3, while the average
observed PM2.5 MCF value  during this campaign is 14.6 g/cm3. There is no conceivable way
that this number is correct. Instead, the average soil particle density of 2.65 g/cm3 given in the
NRCS National Soil Survey Handbook was used as a constant PM2.5 MCF for all sample periods
[56]. Table 17 presents the MCF values (± 95% confidence interval where applicable) used to
convert OPC and lidar volume concentration data to mass concentration for this study.
Table 17. Mass conversion factors (g/cm3) used to convert optical particle measurements to mass
concentrations for each day and averaged for the whole campaign. Error values represent the 95%
confidence interval for n>3.

Date
May 17
May 18
May 19
May 20
June 5
June 7
June 11
June 18
June 25
All
PM
Avg
2.65
2.65
2.65
2.65
2.65
2.65
2.65
2.65
2.65
—
2.S
n
—
—
—
—
—
—
—
—
—
—
PM10
Avg ± 95% CI
2.6 ±1.3
1.6 ±0.3
1.7 ±0.3
1.6 ±0.5
1.8 ±0.3
1.5 ±0.3
4.3 ±1.2
1.8 ±0.5
2.0 ±0.3
2.1 ± 0.3

n
9
2
5
5
5
5
4
6
6
49
TSP
Avg±95%CI
4.4 ±4.0
.6±0.1
.6 ±0.3
.4 ±0.2
.5 ±0.2
.4 ±0.2
2.9 ±0.5
2.3 ±1.0
2.2 ±0.6
2.3 ±0.7

n
7
3
8
4
2
4
4
4
5
44
A possible correlation with the high PM2.s MCF values was found when sample period averages
were compared against the length of each sample period,  as  seen  in Figure 33. The highest
average MCFs for all size fractions were found during the shortest sample periods. It is unclear
what this correlation may mean, but it is not likely a cause as sampler operation is assumed to be
independent of sample period length.
SDL/08-556
California 2008 Tillage Campaign
62

-------
                         •PM2.5       PM10  — TSP  -• -SampleTime
      60
       
-------
                     Upwind PM-
          1J:09    14:24    13:38     14:52     15:07
                                                       200
                                                       150
                                                       100
                                                         60
                                     Upwind PMW,6/18
                                                             14:09    14:24    14:38    14:52     15:07
                                                                             Urn
                                               up*ind TSP. em
                                    14:09     14:24     13:38    14:52     15:07
Figure 34. PM2.S, PM10, and TSP mass concentrations retrieved from collocated lidar and OPC during the
'stare' time series for 6/18. Data acquisition time of the lidar data is 0.5 s while OPCs were set to 20 s
accumulation time. Measurements were done on the upwind side of facility (location Tl)
Table 18. Comparison of PM mass concentrations (jig/m3) as reported by MiniVol samplers and mean values
measured by collocated OPCs and lidar at Tl (upwind) and T2 (downwind) for 6/18/2008.
Upwind
PM sampler (Tl)
Upwind PM sampler
average ± lo
OPC (Tl) ± 95% CI
Lidar ± 95% CI
Downwind
PM sampler (T2)
Downwind PM sampler
average ± lo
OPC (T2) ± 95% CI
Lidar ± 95% CI
PM2.5 (jig/m3)
30.1
29.1 ± 1.3
(n = 2)
4.9 ±0.06
5.1 ±0.09

63.2
48.1 ±12.8
(n = 5)
6.4 ±0.5
6.3 ±0.1
PM10 (jig/m3)
56.5
62.6 ±8.5
(n = 2)
48.2 ±1.1
50.8 ±1.3

87.5
262.6 ±153.9
(n = 5)
93.0 ±15.1
68.1 ±2.1
TSP (jig/m3)
195.4
214.5 ±27.0
(n = 2)
185.4 ±8.5
200.0 ± 6.7

—
1696.0 ±284.1
(n = 2)
442.3 ±95.2
284.5 ±10.2
SDL/08-556
California 2008 Tillage Campaign
64

-------
A similar comparison of OPC and lidar time series data measured from downwind of the tillage
field is shown in Figure 35. The spikes of the concentrations (especially of PMio and TSP) are
due to the dust plume generated by tillage operations crossing this OPC.  Spikes are rarely
observed on the upwind side of the field (upwind spikes are associated with  road traffic). As a
quality check, comparison analysis of collocated lidar, OPC, and MiniVol data was done for all
days  of  the  campaign. In  general, the OPC and  lidar data averaged  for  the PM sampler
acquisition time are in agreement with  mass concentrations  measured by the  collocated PM
samplers. Discrepancies between point-source instruments (PM samplers and OPC) and lidar are
due  to  inherently  different measurement  techniques.  Point  instruments  capture  particle
concentration at a single point (a few cm3) with a small volumetric flow (1.6 x 10~5 to 8.3 x 10~5
m3/s). The lidar acquires information in a volume of ~ 3 m3 for each bin along the laser beam for
each  sample (1-2 samples  per second in  this experiment), and thus can capture  a spatially
variable plume while the plume may miss the point sensors. The  best agreement is observed
when the lidar is compared with PM sampler data averaged over several locations along the up-
or down-wind side of the tillage field. For similar reasons, using MCF values averaged over the
whole campaign yields larger discrepancies between collocated lidar and OPC/PM sampler PM
concentration  data  than daily averaged MCF values. Based on these observations,  a daily
averaged MCF is used for conversion of lidar 'staple' data  used for  flux  and emission rate
calculations.
    300


    250


  "g 200
  B

  o 150
  g 100
  E
  O
  O
    50
                  DownwindPM25,6718
                               Downwind PM10,6/18
     0 *++#
           ,U~*
        14:09
               14:24
                      14:38
                       Time
                             14:52
                                    15:07
3500
t-
B2500
o 2000
•
| 1500
u
° 1000
500
0
• Lidar


*
•

t
I ll
1J JJLijJjfci

t
•
i-lll *
i^JlottyiLk
ft ** * P • M ^ •flffi fciVP1'' -j- — — ,. ^^- ^^^,,
14:09 14:24 14:38 14:52 15:07
Time
                         20000
                         15000
                         10000
                          5000
                                         Downwind TSP, 6(18
                                                            •MM^f
                               14:09
                                      14:24
                                             14:38
                                             Time
                                                    14:52
                                                           15:07
Figure 35. PM2.S, PM10, and TSP mass concentrations retrieved from collocated lidar and OPC during
'stares' at downwind locations. Data acquisition time of the lidar data point is 0.5 sec while OPCs were set to
20 sec accumulation time. Measurements were done on the downwind side of field (location T2) on 6/18/2008.
SDL/08-556
California 2008 Tillage Campaign
65

-------
4.3.3
Lidar Aerosol Concentration Measurements
Lidar data was  collected throughout all  sample periods, except for the Cultivator 3  pass
monitored on June 25 due to an equipment failure after the previous  measurement. All  lidar
scans collected during times when no tillage activity was occurring, based on detailed field notes,
were removed from further calculations. The remaining scans were visually checked to remove
scans with data acquisition errors and to prevent the use of data that was contaminated by other
sources; sources of observed contamination were vehicular traffic on unpaved roads, agricultural
activities immediately upwind, activities associated with the adjacent dairy, and windblown dust.
Contaminated upwind scans, as well as the corresponding downwind scans, were removed from
further emission rate calculations. In most cases two or more downwind scans use the same
upwind scan as a reference because multiple downwind scans were made for each upwind scan.
Downwind scans were not used in further calculations if the corresponding wind direction was
outside of ± 70° from perpendicular to the downwind lidar beam path; if the lidar scan contained
apparent  plumes from an outside source (such as  from unpaved road traffic or the dairy); if no
plumes were detected; or if the tillage plume had a potentially significant portion crossing the
lidar beam closer than 500 meters  down beam. In addition, upwind scans were invalidated if
none of the associated downwind scans were usable. Light wind speeds (< 1 m/s) were recorded
during portions of some measurement periods.  Light winds challenge the measurement system
and resulted in additional invalidation of some lidar scans. The total number of lidar scans made
along both upwind and downwind  planes is presented in Table  19, along with the  number of
valid scans, i.e. those that passed all quality checks, and the percent of valid scans compared to
the total number of scans.
Table 19. Total number of upwind and downwind vertical lidar scans and the number of those scans
determined to be valid for emission rate calculations. No lidar data exists for the 6/25 run because of
instrument problems after the 6/18 run.
Date
5/17 Run 1
5/17 Run 2
5/18
5/19 Run 1
5/19 Run 2
5/20 Run 1
5/20 Run 2
6/5 Run 1
6/5 Run 2
6/7
6/11
6/18
6/25
Operation
Strip-till
Break down borders
Chisel
Disc 1
Disc 2
Lister
Build ditches
Break down ditches,
Cultivator, Roller
Plant
Plant
Herbicide Application
Fertilizer Application
Cultivator
Field
5
4
4
4
4
4
4
4
4
5
5
4
4
Upwind Scans
Total
62
20
122
103
150
126
42
90
36
58
24
20
—
Valid
22
6
92
20
19
17
5
56
25
33
0
10
—
% Valid
35.5
30.0
75.4
19.4
12.7
13.5
11.9
62.2
69.4
56.9
0.0
50.0
—
Downwind Scans
Total
118
60
366
130
148
126
42
259
114
162
48
40
—
Valid
86
14
276
56
31
53
15
169
73
86
0
14
—
% Valid
72.9
23.3
75.4
43.1
20.9
42.1
35.7
65.3
64.0
53.1
0.0
35.0
—
In several cases,  the percent of valid  scans was very low. Although wind and background
conditions were good during the herbicide application on 6/11, none of the scans taken were
deemed valid because the mass difference between the upwind and downwind scans was below
the minimum  detection level (MDL). This means that the operation didn't produce plumes
significant enough to be detected by the  lidar; this operation  was performed by a very small
SDL/08-556
                     California 2008 Tillage Campaign
66

-------
tractor pulling a small spraying apparatus - the only disturbance of the ground was  due to
moving tires. Plumes  of insufficient concentration  differences from  background levels led to
downwind scan invalidation in many other instances  also. The dairy pen areas adjacent to Field 4
proved to be sufficient PM sources such that lidar scans showing dust plumes passing over the
pens  prior to crossing the scanning  plane  were nearly always  invalidated.  Additionally,
windblown dust was entrained off of both Field 4 and upwind field surfaces during both the May
20 sample periods. All of these factors combined to significantly decrease the number of valid
lidar scans available for emissions estimation from most operations.
Examples of the lidar-derived PM concentration have been presented in Figure 34, Figure 35,
and Table 18. Additionally, upwind and downwind  plume area average volume concentrations
used in the flux calculations are shown in Figure 36  and Figure 37. The two top panels show the
profile-averaged wind  speed and direction values used in the flux calculation, with the third and
fourth panels showing the area-averaged volume concentrations measured upwind (Cu) and
downwind (Co) in |j,g3/cm3 (see Eq. 17).  It was  assumed that the upwind PM concentrations
would be more uniform than the downwind PM, therefore, more downwind scans were made
than those  of the upwind conditions. In Figure 36 and Figure  37, the gaps in the  upwind
concentration are due to this sampling plan.

4.4    FLUXES AND EMISSIONS RATES
Emission rates for the  observed tillage operations were calculated using lidar measurements and
measured PM concentrations coupled with model-predicted concentrations via inverse modeling.
During the May 20 R2 sample  period, strong ground-level winds were measured throughout the
entire 1-hr sample time; wind-blown dust plumes originating from the field of study were visible,
often impacting  both  ground-level and elevated downwind  samplers,  and recorded by field
personnel. Due to the size and frequency of these wind-blown dust plumes in comparison to the
strength of the  monitored operation (building  irrigation ditches and field edge  borders), any
emission rate derived from data collected during this  sample period does not accurately represent
the operation's actual impact on downwind PM concentrations. In addition, the same operations
of building  and breaking  down of the  ditch  and field-edge borders  around Field  5,  the
conservation tillage field, were carried out but not  measured. Therefore, the emission  rate for
building ditches and field-edge  borders before irrigation and breaking them down after irrigation
and before planting was omitted from summations of the emission rates within each investigated
tillage method.
SDL/08-556                      California 2008 Tillage Campaign                             67

-------
ID
Wind Speed 10
(m/s)
D
Wind Dir. 50
-50
Upwind 1000
Cone.
3 , 3 1 00
10
Downwind 1000
Cone.
, 3, 3l 100
(iimj/cmj)
10
C
Figure 36. Wind speed, \
concentrations, for the IV
10
Wind Speed
(m/s) 5
0
60
Wind Dir
(cleg) 0
-60
Upwind 1000
Cone 10Q
(iimj/cmj) 1Q
Downwind 1000
c°nc- 100
(u^/cm3) 1Q
r
Figure 37. Wind speed, \
concentrations for the 1V1
TSP *
O PM10
, DM * **J
**< 2.5 *
i i


* * **%*%**** *** * *
*•+•+•• •»•••*••• +*•%*>•+

ijWx^xxxx*?^^— 	
^5£5jj®a^s£%!5£&3;5jj
II 14-

••*v*x-x-xx 	
$*$8S&s
. ~h_rl=l — 1- i
bL^TJrr~1rhi" ^r' ',' 'X 	 J
H- i - r-l-. ' i i^-i i i
"r^ ^-|- -|-I ' i-rH 1 1 1

A.
***+**%+


v*******1
••*•••
. 	 1



=Hlllllllh

XX
3^J%«^
-Hf^mn-
i
4.
i.
+
+**+++


****+****'
+
1 	 1
^x^xxx^tf
w^HHHWGB

I-H+H+H+

'i'ir"0*'
XjXV ^
'^idgS
. . .i-i-O+K
hfpttmth
+
+
	 + •
+ *
+
> 	
*

+
4.


!•** ******
1 	 1

™ST^'

III M-l

'^A^
t*#+W-
+* • ;* :
+
: * * * :
**** 	 **** • • •*++:• -*+-**+-«*+t*+ »


w«»+++++i+**********+*****/+*«+
i 	 1 	 1 	 1 	
	 > 	 i 	 i. . . .

^C^vi£^XXVvii-^l • V -j


l+llllillllll||||-HllUi+|i|iii-
:«^ . . ,jJS;;<: " : 1
K !S^:WX!^W< x^^-^-trf- j
^%^^^
'• "tu1_+;^^
W^^F^^rV^!

) 10 20 30 40 50 60 70 80 90
Sample Number
dnd direction, upwind and downwind plume area average particulate volume
lay 17, 2008 strip-till pass of the conservation tillage method.
TSP
* + PM2.5 *"**
i

* *„* +*++*+


xxx xxx

1+1 1 1 1 I-H-H-

^£
^4--f++++i-'1~r-


)-+++ ++-
i

* **


* +


l+H4444^++4+f4]
X
^^Q^c^^o1
^nj^

) 10 20 30 40 50 60
Sample Number
dnd direction, upwind and downwind plume area averaged particulate volume
ay 19, 2008 first disc pass of the conventional tillage operation.
SDL/08-556
California 2008 Tillage Campaign
68

-------
4.4.1
Lidar Based Fluxes and Emission Rates
The combination of 'staple' and 'stare' measurements from the upwind and downwind sides of
the field, as described in Section 3.1.5, were performed continuously  during  each tillage
operation of the field campaign except the last one. A  critical component  failure prevented us
from further using the lidar during the June 25 sample period. Figure 38  and Figure 39 show
calculated net fluxes  for sequential lidar measurements taken during the  strip-till  pass in the
conservation tillage method on 5/17/2008 and the initial discing pass of the conventional tillage
operation on 5/19/2008, respectively. The  net mass flux through the lidar's vertical scan is the
product of the plume  area averaged volume concentration difference (Co-Cu) multiplied by the
daily average MCF and the component of the wind  velocity that is perpendicular  to the lidar
beam. It is then normalized by the total area tilled and the ratio of the sample time to  total tractor
time to yield the emission rate per unit area tilled per unit of operation time.
Net fluxes were calculated using up- and downwind concentration measurements averaged over
each vertical scan with average wind information for the time of the individual scan.  Single-scan
differences do not account for accumulation or depletion in the measurement box due to wind
speed variation during a scan, incoming background variation, or  storage in/flushing of the flux
box due to the  existing large  scale wind eddy structure (i.e. we do not attempt to measure the
same air mass at the upwind and downwind scans). The process is assumed to be a continuous
emission source, requiring several scans to achieve a meaningful mean estimate of the facility
emission. For calculation efficiency, the net flux was calculated through the downwind surface
first and then the  upwind flux, using  the difference in upwind/downwind fluxes  rather than
difference in concentrations.
       2000
     •s
                        20
                     30
   40     50
Sample Number
60
70
80
Figure 38. Lidar-derived fluxes Og/m2/s) of PM2 s, PM10, and TSP for the May 17,2008 strip-till pass of the
conservation tillage method.
SDL/08-556
                     California 2008 Tillage Campaign
                                                   69

-------
           4000
                                    20
                30
        Sample Number
40
50
Figure 39. Lidar derived fluxes (jig/m2/s) of PM2 s, PM10, and TSP for the May 19,2008 first disc pass of the
conventional tillage operation.
The plume area was chosen manually by observing each downwind scan and highlighting the
area to be included in the calculation. Choosing an area that fully includes the source plume but
not a lot of extra area has two advantages: the primary is that portions of the upwind/downwind
scans that would  clutter the plume can be  removed; secondarily, it minimizes the required
computational resources used by not calculating flux values for pixels which do not contribute
significantly to the overall tillage process.
Using the series of flux measurements collected during each tillage operation, similar to those in
Figure 38 and Figure 39, the mean emissions of PM per unit area tilled per second of tillage
operation were calculated for all days and are shown in Table 20 with respective 95% confidence
intervals.  Emission rates reported for multiple  operations occurring  in the same run are the
average amount of PM emitted per tillage pass and does not represent measured a emission rate
from individual operations.  The breaking down of borders and ditches operation in the June 5 Rl
sample tilled the same 4% of the total field area each pass, so the effects of the greater number of
passes is  spread over a much smaller area than the majority of other tillage operations. The
summed emission rate  for the  conventional tillage method includes  the following  thirteen
implement passes listed in order of occurrence: two break down in-field borders passes (emission
rate spread over entire field area, not just the area tilled), a chisel pass, two disc passes, a lister
pass, two cultivator  passes, a roller pass, a  planter pass, a  fertilizer injection  pass, and two
additional cultivator passes. The summed emissions do not include the emissions calculated for
the building of ditches and borders on May 20. While the last two cultivator passes were not
monitored using the  lidar, they have been included in the summed  emissions because they are
standard weed control operations in the conventional tillage sequence at this production site. The
measured emission rate from the June 5 Rl sample period with the  plant, cultivator 1   2, and
roller operations was used  for the emission rate of the  cultivator 3 and 4  passes. Each of the
cultivator and roller passes were assumed to have the  same emissions and variance as was
SDL/08-556
California 2008 Tillage Campaign
                            70

-------
reported for the June 5 Rl sample period. Likewise, the planter and lister were assumed to have
the same emissions as reported for the June 5 R2 and May 20 Rl periods, respectively.
Table 20. Mean fluxes (jig/m2/s) ± 95% confidence interval from quality controlled samples for each tillage
operation.
Operation (date)
Valid lidar
samples with
detectable
Amass («)
PM25
Oig/m2/s)
PM10
Oig/m2/s)
TSP
Oig/m2/s)
Conservation Tillage Method
Strip-till
Plant
Herbicide Application
86
86
0
Summed Emissions
2.5 ±0.6
3.6 ±1.1

-------
technique, conventional or CMP, was calculated as the sum of the variances of each operation.
The 95% confidence interval was then calculated, again assuming a Gaussian distribution.
The flux data presented in Table 20 were multiplied by the total tractor operation time to yield a
total mass emitted  per unit  area  of the  field for each  operation  to calculate emission rates
presented in Table  21. The  same thirteen passes in an identical  configuration were used  to
calculate the summed emissions for the conventional tillage method.
Table 21. Aerosol mass transfer (± 95% confidence interval) from each field (flux normalized by operation
duration and area tilled) as calculated from lidar data for all tillage operations.
Sample
Operation
#of
passes
PM25
(mg/m2/
operation)
PM10
(mg/m2/
operation)
TSP
(mg/m2/
operation)
Conservation Tillage Method
May 17 Rl
June 7
June 11
Strip-till
Plant
Herbicide Applications
Summed Emissions
1
1
1
3
26.9 ±6.1
50.2 ±15.0

-------
4.4.2     Inverse Modeling Calculations

4.4.2.1        ISC/AERMOD dispersion models
The ground level area source dimensions used in the dispersion models were based on the GPS
readings of field perimeter measurements and from the tractor while performing the operation.
The seed emission rate of 8.6 |ig/s-m2 per operation, calculated by averaging across preliminary
emission rates for  all operations derived in the 2007 tillage CMP study [7], was applied to all
sources  on a per pass basis. In some measurement periods there were multiple passes over the
same  area by either the  same implement or different implements. The seed emission rate was
therefore adjusted by multiplying by the number of passes to account for the  actual total
emissions per operation and sample period. Table  22 lists the tillage operations carried out
during each measurement period, the number of passes performed per operation, the area tilled
per pass per  sample period, the  percent of the  field  that was tilled per pass during the
measurement period, and the resulting seed emission rate used in  the dispersion  models. To
compare the  modeled concentrations  to  field measurements,  the measured background  PM
concentration must be subtracted from the measured downwind concentration results to yield the
concentration resulting from the source activity.
Note that the area worked by some of the operations was much smaller than the entire area of the
field, with the building and breaking down of borders and ditches on May 20 and June 5 working
less than 4%  of the field each. However, these operations required more passes over the same
area to  accomplish the task. The total area tilled (area/pass times  the number of passes) for
breaking down ditches and borders passes of the June 5 Rl period was 31,008 m2, just  13% of
the total area tilled during the sample period. Therefore, the contribution of this operation to the
average emission rate per pass is thought to be small compared to the other two operations.
In situations in which multiple passes were monitored in the same sample period, the emission
rates were calculated on a  single-pass  basis and all  applications of the emission rates herein
derived  should take into account the total number of passes performed over the same area. While
the seed emission rates for operations with multiple passes over the  same area are significantly
higher than other operations in the same measurement period that involve just one pass, as found
in the first June 5 sample, the low percentage of the total area tilled combined with lower active
time for such sources,  results in  the  period-averaged,  model  predicted concentration being
dominated by the longer term, more ubiquitous operation. As the individual effects of each of the
three operations in the June 5 Rl  sample on measured concentrations could not be distinguished
in the period-averaged filter sample, the derived emission rates for each measurement  period,
using the three seed emission rates shown, represents a conglomerate emission rate. However,
for the June 5 R2 sample period, the derived emission rates are for planting only; the  cultivator
and rolling operations were performed in the northern  third of the field and, coupled with
meteorological conditions,  did not impact most downwind samplers. Therefore, source impacts
at all but one downwind location are due strictly to the planting operation.
SDL/08-556                      California 2008 Tillage Campaign                             73

-------
Table 22. Tillage operations, number of passes, area tilled per pass during monitoring period, and the seed
emission rate used for modeling particle dispersion using ISCST3 and AERMOD for each sample period.
Sample
Period
May 17 Rl
May 17 R2
May 18
May 19 Rl
May 19 R2
May 20 Rl
May 20 R2
June 5 Rl
June5R2
June 7
June 11
June 18
June 25
Tillage
Method
Conservation
Conventional
Conventional
Conventional
Conventional
Conventional
Conventional
Conventional
Conventional
Conservation
Conservation
Conventional
Conventional
Tillage
Operation
Strip-till
Breaking down
borders
Chisel
Disc
Disc
Lister
Disc
Build ditches and
borders
Break down
ditches/borders
Cultivate
Roll
Plant
Cultivate
Roll
Plant
Herbicide
application
Fertilizer
application
Cultivate
# of passes
1
2
1
1
1
1
3
N,W,S sides
- 2; E side -
4
W-4;
E-12
2
1
1
2
1
1
1
1
1
Area
tilled/pass
(m2)
90,484
10,112
85,116.1
100,482.5
100,482.5
100,482.5
8,121
N,W,S- 1,795;
E- 1,034
W- 1,548;
E - 2,068
69,953.5
67,644
48,175
25,741
32,404
90,484
90,484
38,126.5
100,482.5
Field portion
tilled/pass
(%)
100.0
10.0
84.7
100.0
100.0
100.0
8.1
N,W,S-1.8;
E-1.0
W-1.5;
E-2.1
69.6
6.5
47.9
25.6
32.2
100.0
100.0
37.9
100.0
Seed
emission
rate (jig/s-
m2)
8.6
17.2
8.6
8.6
8.6
8.6
25.8
N,W,S sides
-17.2;E
side - 34.4
W-34.4;
E- 103.2
17.2
8.6
8.6
17.2
8.6
8.6
8.6
8.6
8.6
ISCST3 concentrations for modeled sample periods ranged from 0.0 to 194.8 ug/m3, with the
highest concentrations typically modeled at a height of 2 m on the southern and western edges of
the tillage area, although the  exact location varied slightly with differing wind directions. Figure
40 shows an example of ISCSTS-modeled concentrations for the third pass with the cultivator on
June 25, 2008 of the conventional tillage method with northwest winds.  It should be noted that a
10 m x 10m grid of receptors  at 2 m agl were employed throughout the near-field modeling
domain, which  produces  a much smoother and more detailed contour plot than the measured
PMio concentration contour  plot seen in Figure 18 and the OPC PMio concentration plot in
Figure 31.
AERMOD modeled concentrations for sample periods besides the May 20 Rl and June 5 R2
samples ranged from 0.0 to 205.1 ug/m3, with the highest concentrations typically modeled at a
height of 2 m on the  southern and western edges of the tillage  area and  at different locations
depending on wind direction. Figure 41 shows an example of AERMOD modeled concentrations
for the same cultivator third  pass shown in Figure 40 for ISCST3, with the same  10 m x 10 m
grid spacing.
SDL/08-556
California 2008 Tillage Campaign
74

-------
                                                                           ua in'-:
                                                                              100
                                                                              90
                                                                              80
                                                                              70
                                                                              60
                                                                              $0
                                                                              30
                                                                              10
Figure 40. ISCST3-modeled results for the third cultivator pass of the conventional tillage operations on June
25,2008 with northwest winds. The area of operations and sampler locations are denoted in black and
contour line numerical values are in jig/m3.

4.4.2.2       ISCSTS-Based Emission Rates
The MiniVol PM2.s, PMio, and TSP concentrations were evaluated for use in calculating
emission rates based on inverse modeling as described in Appendix B, with the verified PM
concentrations shown in Table 33 and Table 34 in Appendix A. OPC mass concentrations were
calculated by multiplying the Vk by the MCF,t for each sample period for use in emission rate
calculations and are presented in Table 35 and Table 36. The OPC data were evaluated for
impacts to upwind concentrations and whether intense plumes from the tillage operations
overwhelmed the downwind OPCs or not for the May 18 through May 20 R2 sample periods.
Manufacturer specifications state the range of the OPC is 0 to 318,000,000 particles/m3.
SDL/08-556
California 2008 Tillage Campaign
75

-------
                                                                               100
                                                                               80
                                                                               70
                                                                               30
                                                                               20
Figure 41. AERMOD modeled results for third cultivator pass of the conventional tillage operations on June
25,2008 with northwest winds. The area of operations and sampler locations are denoted in black and
contour line numerical values are in jig/m3.
Examination  of the downwind  May 18 through May 20  OPC data found that there were no
significant problems  with  overloading, allowing  these  data to be  used in  emission  rate
calculations.  Potential contamination of both OPC  and MiniVol upwind samples from nearby
activities was investigated using  OPC time series  data.  Some significant  events were  found.
Upwind  filter samples  affected by  these  events  had already been removed in the QA/QC
investigation  detailed in Appendix B. Measured upwind OPC PM concentrations were adjusted
to represent solely background  concentrations by removing high  values identified as potential
contamination from  background concentration average calculations. OPC PM concentrations
used in emission rate calculations are given in Table 37 and Table 38.
Sample period average upwind MiniVol and OPC PM concentrations in each  size fraction for
were subtracted from downwind concentrations to calculate PM concentrations  produced by the
tillage  activity, with only downwind concentrations greater than  the average upwind plus the
SDL/08-556
California 2008 Tillage Campaign
76

-------
calculated 67% confidence interval used to calculate emission rates. These operation-produced
PM concentrations were then compared with the model-predicted concentrations as shown in Eq.
22 and multiplied by the seed  emission  rate to  calculate  the observed emission rate at each
location.  Emission rates were then averaged across all valid sample locations for each sample
period and are presented in Table 23 in units of mass per  unit area tilled  per tractor operation
time with 95% confidence intervals when three or more samples were available. Emission rates
presented in Table 24 are in units of mass per unit area tilled, representing the total emissions per
unit  area per operation. The 95% confidence intervals reported  for the summed emissions per
tillage method were calculated the same way as those for the lidar emission rates.
Emission rates were not calculated using filter data for the  following sample periods due to the
near-source downwind samplers being overloaded: May 18,  May 19 Rl, May 19 R2, May 20 Rl,
and May 20 R2. Additionally, PM2.5 and PMio filter-based emission rates  for May  17 R2 and
PMio emission rates for June  5 R2 were not calculated  because the  upwind  samples were
compromised. All of these sample  periods were operations in  the conventional tillage field;
therefore, a summation of emissions  based on  MiniVol  data across all  conventional tillage
operations was not possible. In addition, some emission rates based on the  MiniVol data set are
based on one or two data points, preventing further statistical calculations. However,  emission
rates were able to be calculated using at least three OPC PM data points for all of these sample
periods. The emission rates reported for the June 5  R2 sample in the conventional field are for
planting only: the first and second cultivator passes and the roller pass that were being performed
at the same time were  shown  to  impact  only  one downwind sampler location through
examination of tractor location, average  wind direction  (320°  ± 4), OPC data, and model
predicted concentrations. Data at this location were removed from emission rate calculations.

4.4.2.3       AERMOD-based Emission Rates
Sample period averaged emission rates based on concentrations predicted by AERMOD were
determined using the same techniques described for the calculations of ISCST3-based emission
rates. Emission rates on a mass per unit area tilled  per unit time of operation are found in Table
25 and emission rates on a mass per unit area tilled are  found in Table 26, with the 95%
confidence intervals calculated the same way as those for the lidar emission rates for periods
with three or more valid downwind samples of each size.
SDL/08-556                      California 2008 Tillage Campaign                             77

-------
Table 23. Mean emission rates per unit area per unit time (± 95% CI for n > 3) for each operation as
determined by inverse modeling using ISCST3.
Sample
Operation
#of
passes
PM25
(Hg/m2/s)
MiniVol
OPC
PM10
(Hg/m2/s)
MiniVol
OPC
TSP
(Hg/m2/s)
MiniVol
OPC
Conservation Tillage Method
May 17 Rl
June 7
June 11
Strip-till
Plant
Herbicide
Application
Summed Emissions
1
1
1
3
2.7 ±0.5
6.5
12.1
21.3
1.0 ±
0.1
0.3 ±
0.2
0.2 ±
0.1
1.5 ±
0.2
21.8±
4.9
6.7
16.5
45.0
30.7 ±
4.5
8.5 ±
1.8
10.4 ±
4.1
49.7 ±
6.3
108.3 ±
54.1
41.9±
25.7
31.6±
14.9
181.7 ±
61.7
147.9 ±
33.0
43. 3 ±
11.4
30.0 ±
16.9
221.2 ±
38.8
Conventional Tillage Method
May 17 R2
May 18
May 19 Rl
May 19 R2
May 20 Rl
May 20 R2
June 5 Rl
June 5 R2
June 18
June 25
Break down
in-field
borders
(spread over
entire field
area)
Chisel
Disc 1
Disc 2
Lister
Build up ditch
and borders
Break down
ditch, break
down borders,
Cultivator 1 &
2, Roller
Plant
Fertilizer
Injection
Cultivator 3
Summed Emissions
2
1
1
1
1
4,2
12,4,
2,1
1
1
1
13
—
—
—
—
—
—
1.5 ±0.6
6.1 ±0.6
12.6 ±
11.4
0.9 ±0.1
—
0.3 ±
0.1
0.4 ±
0.1
0.8 ±
0.3
1.1±
0.3
1.0 ±
0.1
—
0.2 ±
0.1
0.4 ±
0.2
3.9±
2.9
0.3 ±
0.1
9.4 ±
2.9
—
—
—
—
—
—
3.1 ±1.6
—
102.9 ±
63.1
6.6±3.1
—
24.0 ±
5.3
12.7 ±
6.3
30.7 ±
12.4
53. 1±
18.1
42.5 ±
4.8
—
3.7 ±
1.6
10.0 ±
3.9
116.8
±61.0
9.9 ±
3.8
344.8
±66.1
3.2
—
—
—
—
—
—
33.1
682.7
55.1 ±
8.1
—
288.8 ±
51.6
79.5 ±
27.2
203.2 ±
86.1
310.7 ±
75.8
245.2 ±
42.0
—
18.7 ±
11.3
52.7 ±
19.3
841.5 ±
387.2
65. 9 ±
22.1
2498.2
± 415.5
SDL/08-556
California 2008 Tillage Campaign
78

-------
Table 24. Mean emission rates per unit area (± 95% CI for n > 3) for each operation as determined by inverse
modeling using ISCST3.
Sample
Operation
#of
passes
PM25
(mg/m2/
operation)
MiniVol
OPC
PM10
(mg/m2/
operation)
MiniVol
OPC
TSP
(mg/m2/ operation)
MiniVol
OPC
Conservation Tillage Method
May 17 Rl
June 7
June 11
Strip-till
Plant
Herbicide
Application
Summed Emissions
1
1
1
3
29.9 ±
5.9
89.6
40.5
159.9
10.7
± 1.6
4.5 ±
2.1
0.6 ±
0.3
15.9
±2.7
239.2 ±
54.1
92.5
55.3
386.9
337.6
±49.0
117.2
±24.8
34.8 ±
13.7
489.6
±56.6
1189.1 ±
593.9
575.6 ±
353.3
105. 8 ±
50.0
1870.4 ±
692.8
1623.7
± 362.0
595.9 ±
157.4
100.4 ±
56.7
2320.0
± 398.7
Conventional Tillage Method
May 17 R2
May 18
May 19 Rl
May 19 R2
May 20 Rl
May 20 R2
June 5 Rl
June 5 R2
June 18
June 25
Break down in-
field borders
(spread over
entire field
area)
Chisel
Disc 1
Disc 2
Lister
Build up ditch
and borders
Break down
ditch, break
down borders,
Cultivator 1 &
2, Roller
Plant
Fertilizer
Injection
Cultivator 3
Summed Emissions
2
1
1
1
1
4,2
12,4,
2, 1
1
1
1
13
—
—
—
—
—
—
40.2 ±
16.4
84.0 ±
8.6
48.8 ±
44.3
12.7 ±
2.0
—
1.1±
0.3
9.5 ±
2.3
13.7
±5.6
18.3
±5.5
17.7
±1.2
—
5.6 ±
3.3
5.1±
2.8
15.1
±
11.2
4.4 ±
1.7
107.1
±
15.5
—
—
—
—
—
—
84. 1±
42.8
—
400. 1±
245.2
94.8 ±
45.5
—
79.5 ±
17.7
282.9
±
140.4
533.3
±
215.9
904.5
±
308.3
775.1
±87.1
—
100.2
±42.2
137.6
±53.5
454.3
±
237.0
142.9
±54.8
3833.0
±
489.9
10.5
—
—
—
—
—
—
454.6
2654.3
796.7 ±
117.6
—
956.6 ±
170.8
1767.8
± 605.0
3533.4
±
1497.6
5290.4
±
1291.0
4475.3
± 766.8
—
498.9 ±
301.7
724.6 ±
265.0
3271.6
±
1505.5
954.4 ±
320.3
24381.5
±
2781.5
SDL/08-556
California 2008 Tillage Campaign
79

-------
Table 25. Mean emission rates per unit area per unit time (± 95% CI for n > 3) for each operation as
calculated by inverse modeling using AERMOD.
Sample
Operation
#of
passes
PM25
(Hg/m2/s)
MiniVol
OPC
PM10
(Hg/m2/s)
MiniVol
OPC
TSP
(Hg/m2/s)
MiniVol
OPC
Conservation Tillage Method
May 17 Rl
June 7
June 11
Strip-till
Plant
Herbicide Application
Summed Emissions
1
1
1
3
2.7 ±0.5
6.5
12.1
21.3
1.2 ±
0.2
0.4 ±
0.2
0.2 ±
0.1
1.7 ±
0.3
25.1 ±
7.0
6.8
15.6
47.5
36.2 ±
7.3
9.2 ±
1.8
10.2 ±
3.1
55.6 ±
8.1
131.9±
74.0
46.3 ±
27.1
33. 8±
19.8
212.1 ±
81.3
174.6 ±
49.8
47.0 ±
12.1
28.3 ±
14.2
249.9 ±
53.2
Conventional Tillage Method
May 17 R2
May 18
May 19 Rl
May 19 R2
May 20 Rl
May 20 R2
June 5 Rl
June 5 R2
June 18
June 25
Break borders
(spread over entire
field area)
Chisel
Disc 1
Disc 2
Lister
Build up ditch and
borders
Break down ditch,
break down borders,
Cultivator 1 & 2,
Roller
Plant
Fertilizer Injections
Cultivator 3
Summed Emissions
2
1
1
1
1
4,2
12,4,
2, 1
1
1
1
13
—
—
—
—
—
—
1.5 ±0.6
6.1 ±0.6
12.6 ±
11.4
0.9 ±0.1
—
0.4 ±
0.1
0.5 ±
0.2
1.0 ±
0.6
1.2 ±
0.6
0.9 ±
0.1
—
0.2 ±
0.2
0.4 ±
0.2
3.9±
2.9
0.5 ±
0.2
10.4 ±
3.0
—
—
—
—
—
—
3.4 ±1.7
—
102.7 ±
64.5
10.7 ±
4.9
—
25.0 ±
4.6
18.4 ±
6.9
40.5 ±
21.4
56.2 ±
33.1
39.2 ±
4.7
—
4.4 ±
1.7
10.4 ±
3.0
115. 9±
60.9
16.5 ±
5.7
376.8 ±
73.9
4.0
—
—
—
—
—
—
40.4
692.4
90.3 ±
13.7
—
301.7 ±
42.0
93.7 ±
36.2
269.0 ±
143.4
325.7 ±
179.1
226.3 ±
41.9
—
21.6 ±
12.0
54.9 ±
14.1
829.6 ±
376.4
110.5 ±
32.0
2688.5
± 451.2
SDL/08-556
California 2008 Tillage Campaign
80

-------
Table 26. Mean emission rates per unit area (± 95% CI for n > 3) for each operation as calculated by inverse
modeling using AERMOD.
Sample
Operation
#of
passes
PM25
(mg/m2/
operation)
MiniVol
OPC
PM10
(mg/m2/
operation)
MiniVol
OPC
TSP
(mg/m2/ operation)
MiniVol
OPC
Conservation Tillage Method
May 17 Rl
June 7
June 11
Strip-till
Plant
Herbicide
Application
Summed Emissions for
Conservation Tillage
1
1
1
3
31.7±
5.6
91.5
36.7
159.9
12.7 ±
2.8
4.9 ±
2.2
0.6 ±
0.3
18.1 ±
3.5
275. 8 ±
76.7
93.5
52.1
421.4
397.2 ±
80.4
126.9 ±
25.1
34.1 ±
10.3
558.2
±84.9
1448.3 ±
812.5
636.8 ±
372.8
113.3 ±
66.2
2198.4 ±
896.4
1917.1 ±
547.0
645.9 ±
166.2
94.7 ±
47.5
2657.7 ±
573.7
Conventional Tillage Method
May 17 R2
May 18
May 19 Rl
May 19 R2
May 20 Rl
May 20 R2
June 5 Rl
June 5 R2
June 18
June 25
Break borders
(spread over entire
field area)
Chisel
Disc 1
Disc 2
Lister
Build up ditch and
borders
Break down ditch,
break down borders,
Cultivator 1 & 2,
Roller
Plant
Fertilizer Injections
Cultivator 3
Summed Emissions for
Conventional Tillage
2
1
1
1
1
4,2
12,4,
2,1
1
1
1
13
—
—
—
—
—
—
43.8 ±
11.1
86.9 ±
13.5
50.3 ±
46.7
21.0 ±
3.4
—
1.2 ±
0.3
11. 3±
4.2
18.0 ±
9.9
20.2 ±
9.9
16.3 ±
1.7
—
6.6 ±
4.1
5.2 ±
2.5
15.0 ±
11.2
7.5 ±
2.8
123.1
±20.4
—
—
—
—
—
—
90.3 ±
45.6
—
399.4 ±
250.7
155. 5 ±
71.6
—
83.0 ±
15.4
409.5 ±
154.2
704.7 ±
372.4
957.7 ±
564.2
714.8 ±
86.3
—
116.5 ±
46.0
142.6 ±
40.7
450.5 ±
236.9
239.3 ±
82.3
4374.0
± 752.6
130.6
—
—
—
—
—
—
555.6
2692.1
1307.3 ±
198.0
—
999.1 ±
139.2
2085.4 ±
804.5
4677.8 ±
2492.6
5545.3 ±
3049.5
4131. 1±
764.0
—
578.0 ±
320.8
754.5 ±
193.6
3225.5
±1493.4
1599.8 ±
462.7
27,351.4
± 4438.3
4.5    DERIVED EMISSION RATE COMPARISON
Two emission rate determination approaches were employed to calculate three different sets of
emission rates in order to quantify differences between conventional  tillage methods  and a
conservation  tillage method  using  strip-till  technology.  Problems  with  filter-based  PM
measurements prevented emission rate calculations for all size fractions over five consecutive
sample periods and  some size fractions during two other periods, preventing the summation of
ISCST3- and AERMOD-based emission rates for MiniVol data across the conventional tillage
SDL/08-556
California 2008 Tillage Campaign
81

-------
method operations. The OPC-based PM measurements passed  QA/QC  and were used to
calculate emission rates through inverse modeling for all size fractions for all sample periods.
The lidar system effectively sampled the vertical downwind plane and measured time-resolved
plume characteristics for each operation at each paniculate size  fraction,  except for when a
critical component failure in the lidar system prevented it from collecting data for the last tillage
operation. Sample period average emission rates, with their respective error estimates, are shown
in Figure 42 for comparison when values from all methods are available.
A summary table of the PMio emission rates calculated from each approach is given in Table 27
for comparison. The calculated PMio emission rates demonstrate that for total mass of PMio per
unit area per  unit time  of operation (|ag/m2/s), the conservation tillage method produced about
10% as much as the conventional  method based on lidar measurements. PMio emission rates
calculated by the lidar and the inverse modeling method are within the extent of their 95%
confidence intervals for the May 19 R2, June 5 Rl, and June  18 samples, while differences
between them are statistically significant for the May 17 Rl, May 18, May 19 Rl, and June 5 R2
samples.  Another  difference is that the filter  and OPC-based inverse modeling did report
emission rates for the  herbicide application in the conservation  tillage field while the lidar
reported  upwind  and  downwind  concentration differences  less than the  MDL.  A potential
explanation for this is that the lidar beam was only able  to come down to 6-8 m agl  for safety
reasons, while most filter and OPC samplers were  located below this level. If a plume stayed
close to the ground, the  lidar may have been unable to detect it.
The largest difference in derived emission rates is between the model-derived  values and the
lidar. This is likely due to the fact that ISCST3 and AERMOD are very similar numerical models
and the lidar  measures  actual fluxes. The models are limited to the types and configuration of
sources that can be used (point source, volume source, area source,  or line source), which means
that a compromise must be  made when  dealing with a small, moving area source such as is the
case in agricultural tillage.  Also, the temporal and spatial resolution of the lidar  allows it to see
micro-scale variations in plume characteristics and movement such as plume strength, frequency,
lofting and detachment from the surface, wind direction and wind speed effects.
The models must assume a constant emission from the entire emitting surface  area,  while the
lidar sees the  plume movement as the tractor moves across the field. This kind of microstructure
cannot be  captured by the long-term sampling required for implementation of ISCST3 and
AERMOD. Consequently, ISCST3 and AERMOD are  incapable  of generating fine levels of
spatial and temporal detail.  As discussed previously  [7],  examples of such differences in
technique are evident when comparing single-scan and sample period average PM concentrations
measured by the lidar with average  modeled concentrations in a vertical plane. In this work, it is
obvious that  the  plumes measured by lidar  were  detected  at much higher elevations  than
predicted by the models, as  observed by the greater than background concentrations measured at
heights up to  and exceeding 100 m. Since the point sensors were deployed near the surface (at 2
and 9 m) downwind of the source, these  higher plumes were only sampled by the lidar.
Another factor to consider is the exhaust emissions from the tractor engines. The  report detailing
the comparison of fall CMP and conventional tillage method measurements and findings [7], the
companion study to the study herein described, calculated exhaust PMio emissions based on fuel
usage information  provided by the cooperating producer and a fuel use-based PMio  exhaust
emission rate  given by Kean et al. [30]. Exhaust emissions in the fall tillage report were
SDL/08-556                      California 2008 Tillage Campaign                             82

-------
             30

             25

           c 20
           re
          •£ 15
          fVI  •*—>
I ISCSTS-MiniVol

IISCST3-OPC

I Lidar
                                   PM2.5
                          AERMOD-MiniVol
            180
            160
            140
            120
            100
             80
             60
             40
             20
              0
1000
 900
 800
 700
 600
 500
 400
 300
 200
 100
   0
 I ISCSTS-MiniVol
 IISCST3-OPC
 I Lidar
                                  PM10
                            AERMOD-MiniVol
                           IAERMOD-OPC

                                      COV
                     I ISCSTS-MiniVol
                     IISCST3-OPC
                     I Lid
                           TSP
                    AERMOD-MiniVol
                   IAERMOD-OPC
Figure 42. A comparison of PM2.5, PM10, and TSP emission rates ± 95% confidence intervals derived through
lidar scanning techniques and inverse modeling using ISCST3 and AERMOD dispersion models and
measured PM concentrations.
SDL/08-556
               California 2008 Tillage Campaign
                                                                                    83

-------
Table 27. Calculated PM10 emission rates (± 95% confidence interval) from the lidar and inverse modeling
using two dispersion models.
Sample
Operation
Emission Rates (|ig/m2/s)
Lidar
ISCST3
MiniVol
OPC
AERMOD
MiniVol
OPC
Conservation Tillage Method
May 17 Rl
June 7
June 11
Strip-till
Plant
Herbicide Application
Sum for Conservation Tillage
(mg/m2)
47.7 ±
10.7
12.8 ±
3.8

-------
                    5.     SUMMARY AND CONCLUSIONS
A project to determine the control effectiveness of Conservation Management Practices (CMPs)
for agricultural tillage was funded by the San Joaquin Valleywide Air Pollution Study Agency
and carried out in the San Joaquin Valley of California in May and June of 2008. The study was
conducted  to  quantify  particulate  emissions (PM2.5, PMio, and  TSP) from conventional
agricultural tillage methods and a conservation tillage CMP utilizing an Orthman 1-tRIPr, a strip-
till  implement.   The  Orthman 1-tRIPr is  a tillage  implement that incorporates multiple
conventional tillage implements into one piece of equipment and applies these to narrow strips
across the field, as opposed to the conventional tillage method which applies more operations to
the entire surface area of the field. The objective of this study was to address three fundamental
research questions:
       1)       What are the magnitude, flux,  and transport of PM emissions produced by
                agricultural practices for row crops where tillage CMPs are being implemented
                vs.  the magnitude,  flux,  and  transport  of  PM  emissions produced  by
                agricultural practices where CMPs are not being implemented?
       2)       What are the control efficiencies  of equipment being used to  implement the
                "conservation tillage" CMP? Jf resources allow assessing  additional CMPs,
                what  are the  control  efficiencies of the  "equipment  change/technological
                improvements" CMP?
       3)       Can these CMPs for a specific crop be quantitatively compared, controlling for
                soil type, soil moisture,  and meteorological conditions?
The  study was carried out from May  15 until June 25, 2008  in the San Joaquin Valley of
California on a commercial production farm. Two adjacent fields with the same crop  and similar
soil properties  were selected for observation. The  study investigated conventional 13-pass spring
tillage sequence for a  field  going from winter wheat to corn was:  in-field irrigation border
breakdown (x2), chisel, disc (x2), lister, cultivate (x2), top roll,  plant, fertilizer injection,  and
cultivate  (x2). The conservation  tillage  CMP sequence consisted of three passes:  strip-till,
plant/fertilize,  and herbicide spray. Both fields also required passes over very small areas  (<5%
of total field area) to build and break down ditches and field-edge borders, but this step was only
measured in the conventional tillage application. Particulate emissions were determined  using
arrayed filter and optical-based sampling coupled with inverse modeling, using both ISCST3  and
AERMOD, as  well as advanced scanning lidar techniques.  Supporting operation characteristics
(operation time, number of tractors in  operation, potential contamination issues,  etc.) were
recorded   and   meteorological,   soil  characteristic,   and   particle   chemical  composition
measurements  were made and are reported in this document.
Overloading of downwind filter-based samplers and contaminated upwind samplers prevented
emission rates  from being calculated through inverse modeling for some or all PM size fractions
during seven sample periods. The collocated OPCs were found to have not been overloaded, thus
allowing  emission rates to be calculated  from  mass converted OPC data for all sample periods.
The  scanning  lidar technology employed  was  able to calculate emissions for all but  one
measurement period.
Table 28 summarizes PMio aerosol emission values found  during this study along with results
from previous  studies found in the literature in units of mass emitted per unit area tilled. The
summed  emissions for the conventional tillage method based on lidar data includes the following
SDL/08-556                      California 2008 Tillage Campaign                             85

-------
13 passes in order: two break down in-field border passes, chisel, two disc passes, lister, two
cultivator passes, roll,  plant,  fertilizer injection, and two more cultivator passes. The summed
conservation  tillage sequence  consists of  the  following  three  passes:  strip-till,  plant, and
herbicide application.
Some of the  values herein reported are in  agreement with those reported by Flocchini et  al.
(2001) and Madden et  al. (2008), as well as the PMio emission factors used by CARB, such as
the strip-till and plant passes in the conservation tillage method and the cultivate and roll passes
in conventional tillage  [12][16][13]. Other emission rates are different and larger than previous
values reported in the literature, especially the discing 1 and 2, chisel, and lister passes of the
conventional  tillage, though those derived through inverse modeling are  below the estimated
95% level  for their respective distributions  as presented in the uncertainty calculations in this
report. The lidar-derived emission rates for the disc 1, disc 2, chisel, and lister passes are high
when compared to values found by inverse modeling coupled with OPC PM data, values in the
literature for the same operations, and values reported in the 2007 fall CMP tillage study which
used the same lidar methodology. These relatively high emission  rates provide indirect support
for the  conclusion that downwind PM samplers were  likely  overloaded  during those sample
periods by high aerosol concentrations.  While the  values from listed published studies  are
generally not in  close  agreement,  they  are  relatively well fit by lognormal or Weibull
distributions,  which  have previously been  shown to represent emissions  factors datasets well
[31]. In addition, they are within the range of the variability expected from measurements made
under different meteorological and soil conditions, as demonstrated by the wide range of values
from Flocchini et al. (2001) [12] summarized in Table 2. The results from this campaign are not
in as good agreement as previous results have been [7]. In general our results are larger than their
corresponding literature values.
Emission rates for PM2.5, PMio, and TSP from both lidar data and inverse modeling coupled with
OPC data by operation are presented in Table 29 in units of mass emitted per unit area tilled per
operation pass, along with the average tillage rate in hours per hectare where the hour represents
total tractor operation time. In the case where two tractors  were operating at the same time for
the duration of the sample period, the total tractor operation  time would be twice that of the
elapsed time  of the operation. Comparisons were made between the conventional  and CMP
tillage practices based  on these variables  and are shown in the same table.  The conservation
tillage method produced 9.5% as much PM2.5, 6.3% as much PMio, and 9.1% as much TSP as the
conventional method according to the lidar data and 14.7% as much PM2.5, 12.8%  as much PMio,
and  9.7% as  much TSP  as  the  conventional  method  according to the AERMOD-OPC
combination.  Therefore, the control efficiency  of the CMP  for particulate emissions from these
three  data sets  was as follows: lidar - 0.91,  0.94,  and 0.91  for PM2.5,  PMio,  and TSP,
respectively; and AERMOD-OPC - 0.85, 0.87, and 0.90 for PM2.5, PMio, and TSP, respectively.
The control efficiency (r\) was calculated according to Eq. 23, which was based on a collection
efficiency equation found in Cooper and Alley (2002)

                                      1=  ECT~£ST                                   (23)
                                            hCT

where ECT is the calculated emission rate  for the conventional tillage  method and EST is the
calculated emission rate for the strip-till conservation tillage method [58]. These differences in
SDL/08-556                      California 2008 Tillage Campaign                              86

-------
total emissions per tillage treatment are significant at the 99.99% level for PM2.5, PMio, and TSP
for both emissions calculation methodologies employed.

Table 28. A comparison of PM10 emission rates herein derived and found in literature.
Operation
Flocchini
etal.
(2001)
[12]*
(mg/m2)
Madden
etal.
(2008)
[16]*
(mg/m2)
CARB
(2003a)
[13]
(mg/m2)
Previous
SDL
CMP
Study:
Lidar
m*
(mg/m2)
This study
Lidar
(mg/m2)
ISCST3
MiniVol
(mg/m2)
OPC
(mg/m2)
AERMOD
MiniVol
(mg/m2)
OPC
(mg/m2)
Conservation Tillage Method
Strip-till
Plant
Herbicide
Application
Conservation
Method Sum




180.5
233.8










523.8 ±
117.8
175.6 ±
52.6

-------
Table 29. Participate emissions, from Lidar data and inverse modeling with OPC data, and tillage rate
comparison between conventional and conservation tillage.

Average Emission Rates ± 95% CI (mg/m2)
PM2.5
Lidar
Conservation Tillage Method
Strip-till
Plant
Herbicide
Application
Sum
26.9 ±
6.1
50.2 ±
15.0

-------
The time per hectare required to perform the CMP work was 16.3% of the conventional method,
similar to the percent difference between PM emissions of the two treatments. It is interesting to
note that while the absolute values  of calculated emissions are different between the two
techniques, the control efficiencies are very close to each other. Also, the control efficiencies are
similar across the three  size fractions. It should be noted that  other reductions in  emissions
between  the two tillage management practices are likely  to have occurred, mainly due  to
decreased fuel usage in tractor engines. Unlike the previously conducted companion study, fuel
usage was not quantified in this study and prevented accounting for the associated reductions in
emissions.  However, these reductions are expected to be similar to the reduction  in tractor
operation time (-85%).
Our lidar measurements indicate comparable levels  of suspended particulate as is predicted by
our modeling.  Previous results supported  the  use of lidar  measurements as an  important
complement to ground-based sensors because ground-based sensors cannot measure elevated
plumes and ISCST3 and AERMOD would realize significant benefit if lidar-derived information
could be  incorporated into their calculations. The present results are consistent with that claim as
during this campaign we were able to generate a more consistent picture of the  emission rates
using both technologies.
SDL/08-556                      California 2008 Tillage Campaign                              89

-------
                           6.     LESSONS LEARNED
The sampling difficulties encountered during this field study, specifically concerning filter-based
measurements, prompt us to provide a list of practices and procedures that can mitigate these
problems in future field measurements. There are several lessons to be learned from the
experiences of this study.
   1)  Each kind of instrument has a specific operational range in terms of concentration, mass,
       temperature, etc. Therefore,  adequate accommodations must be made during the planning
       stages of the field  exercise such that each instrument can operate within its specified
       range. In this study, PM2.5 and PMio MiniVol samplers were very likely overwhelmed by
       concentrated plumes from the tillage  operations  during five  sample periods. To  prevent
       overwhelming the  samplers in future field measurements, they can be moved further
       away from  the  source,  thereby allowing more time and distance for dispersion of the
       plume and a decrease in concentration.

   2)  Frequent (e.g. daily or even immediately prior  to  each sampler  period) inspection of
       instrument condition is important and may provide early warnings of potential  problems,
       especially in high concentration conditions. During this experiment large crustal particles
       and organic matter were found in  the sample head assembly,  pointing to potential
       problems with  the impactor,  overloading of the  sampler, and/or  contamination. In
       addition, very heavy cumulative particle loading of the silicon vacuum grease used in the
       impactor assembly  to prevent particle bounce  and/or re-entrainment was observed after
       the  May sample periods. Heavy loading of the  grease may increase the probability of
       particle bounce and/or re-entrainment into the air stream. These two observations lead us
       to the conclusion that extreme vigilance regarding instrument condition is required at all
       times - the MiniVols shall not be considered "set and forget" instruments. In the present
       case, had we performed a more thorough daily  inspection of the sampler heads  we would
       have been able to take corrective action earlier.  One example of corrective action is, in
       the  case of PIVb.s samplers,  is to install a PMio impactor upstream of the PIVb.s impactor
       head to  act as an additional large particle filter and decrease the overall loading on the
       PM2.5 impactor. Another preventative  action is to  clean  and regrease  the MiniVol
       impactor assemblies after each use if exposed to high aerosol concentrations.

   3)  Filter  handling and storage  must  occur  in  a  clean environment, with  potential
       contamination factors minimized. During this experiment, filter handling was performed
       onsite in a trailer. While it was maintained as clean as possible, there were high winds
       with blowing dust on May 20 that may have been a source of contamination on the filters
       being  handled.  If  an  adequately clean  onsite filter processing location cannot be
       identified, an appropriate solution might be to transport the impactor heads to an offsite
       location for processing (such as a hotel room).

Careful application of these lessons learned will help prevent future problems similar to  those
encountered during  this field campaign. However, constant evaluation  of actual conditions for
potential problems is strongly encouraged, with  proper mitigation promptly employed.
In addition to the above lessons learned, peer review of this report also brought to our attention
that measurements of additional parameters should be considered in future studies. Specifically,
SDL/08-556                      California 2008 Tillage Campaign                             90

-------
it was suggested that soil surface temperature measurements be made. Differences in soil surface
temperature may  result in  different  vertical  and  horizontal  PM dispersion (with all  other
conditions being equal) as surface conditions are known to affect turbulence and dispersion.
SDL/08-556                      California 2008 Tillage Campaign                              91

-------
                          7.    ACKNOWLEDGMENTS
The Space Dynamics Laboratory would like to thank the individuals and groups whose efforts
made this study and subsequent analysis possible. Cooperators include the USDA-ARS National
Laboratory for Agriculture and the Environment (Dr. Jerry Hatfield, Dr. John Prueger, and Dr.
Richard Pfeiffer), Utah State University (Mark Erupe, Dr. Randy Martin, Derek Price, Emyrei
Reese,  Dr. Phil  Silva), U.S. EPA (Sona Chilingaryan, Kerry Drake, Ron Myers, Dr.  Robert
Vanderpool,  and Dr.  David J.  Williams), the  San Joaquin Valleywide Air Pollution  Study
Agency, the San Joaquin Valley Ag  Technical Group, the San Joaquin Valley Air Pollution
Control District (Jessi Fiero, Sheraz Gill, Ramon Norman, Samir Sheikh, Patia Siong, James
Sweet), California Air Resources Board (Kevin Eslinger,  Shelby Livingston, Karen Magliano)
and the cooperative agricultural producers and industry representatives. Individuals from SDL
who participated in the field measurement campaign or contributed to the report include Dr. Gail
Bingham, Bill Bradford, Jennifer Bowman, Eve Day, Carrie Farmer, Eva Gillespie,  Cassi Going,
Dr. Allen Howard, Everett  Ito, Derek Jones, Tanner Jones, Spencer Kitchen, Richard Larsen,
Christian Marchant,  Kori Moore,  Brad Peterson,  Andrew Pound,  Shane Topham,  Nathan
Whipple, Dr. Tom Wilkerson, Dr. Michael Wojcik, Cordell Wright and Dr. Vladimir Zavyalov.
SDL/08-556                     California 2008 Tillage Campaign                             92

-------
                               8.     PUBLICATIONS
G. E. Bingham, C. C. Marchant, V. V. Zavyalov, D. J. Ahlstrom, K. D. Moore, D. S. Jones, T. D.
       Wilkerson, L. E. Hipps, R. S. Martin, J. L. Hatfield, J. H. Prueger, R. L. Pfeiffer. 2009.
       Lidar based emissions measurement at the whole facility scale: Method and error
       analysis. Journal of Applied Remote Sensing 3(1):033510 [doi: 10.1117/12.829411].
C. C. Marchant, T. D. Wilkerson, G. E. Bingham, V. V. Zavyalov, J. M. Andersen, C. B. Wright,
       S. S. Cornelsen, R. S. Martin, P. J. Silva, J. L. Hatfield. 2009. Aglite lidar: A portable
       elastic lidar system for investigating aerosol and wind motions at or around agricultural
       production facilities. Journal of Applied Remote Sensing 3(1):033511 [doi:
       10.1117/12.829412].
C. C. Marchant, K. D. Moore, M. D. Wojcik, R. S. Martin, R. L. Pfeiffer, J. H. Prueger, J. L.
       Hatfield. 2011. Estimation of dairy particulate matter emission rates by lidar and inverse
       modeling. Transactions of ASABE 54:1453-1463.
K. D. Moore, M. D. Wojcik, C. C. Marchant, R. S. Martin, R. L. Pfeiffer, J. H. Prueger, J. L.
       Hatfield. 2011. "Comparisons of measurements and predictions of PM concentrations and
       emission rates from a wind erosion event," in Proc. International Symposium on Erosion
       and Landscape Evolution  (ISELE) 2011, D. C. Flanagan, J. C. Ascough II, and J. L.
       Nieber, eds., American Society of Agricultural and Biological Engineers, St. Joseph.
K. D. Moore, M. D. Wojcik, R. S. Martin, C. C. Marchant, G. E. Bingham, R. L. Pfeiffer, J. H.
       Prueger, J. L.  Hatfield. 2013. Parti culate emissions calculations from fall tillage
       operations using point and remote sensors. Journal of Environmental Quality [doi:
       10.2134/jeq2013.01.0009].
M. D. Wojcik, G. E. Bingham, C. C. Marchant, V. V. Zavyalov, D. J. Ahlstrom, K. D. Moore, T.
       D. Wilkerson, L. E. Hipps, R. S. Martin, J. L. Hatfield, J. H.  Prueger. "Lidar based
       particulate flux measurements of agricultural field  operations," in IGARSS 2008, Boston,
       Massachusetts, July 2008.
M. D. Wojcik, K. D. Moore, "Laboratory for Atmospheric and Remote Sensing (LARS)", Wind
       Erosion Workshop organized by the Agriculture Research Service - Wind Erosion
       Research Unit, Manhattan, Kansas, April 2008.
M. D. Wojcik, R. S. Martin, J. L Hatfield. 2012. "Using lidar to characterize particles from point
       and diffuse sources in an agricultural field," in Environmental Remote Sensing and
       Systems Analysis, ed. N.-B. Chang, CRC Press, Taylor and Francis Group, Boca Raton,
       pp. 299-331.
V. V. Zavyalov, C. C. Marchant, G. E. Bingham, T. D. Wilkerson, J. L. Hatfield, R. S. Martin, P.
       J. Silva, K. D. Moore, J. Swasey, D. J. Ahlstrom, T. L. Jones. 2009. Aglite lidar:
       Calibration and retrievals of well characterized aerosols from agricultural operations
       using a three-wavelength elastic lidar. Journal of Applied Remote Sensing 3(1):033522
       [doi: 10.1117/12.833365].
SDL/08-556                      California 2008 Tillage Campaign                             93

-------
                               9.     REFERENCES

[1]     U.S.  EPA.  1995. User's guide for the Industrial  Source Complex (ISC3) Dispersion
       Models. Research Triangle Park, NC: U.S. Environmental Protection Agency, Office of
       Air Quality Planning and Standards Emissions. Monitoring, Analysis Division. January,
       2008. http://www.epa.gov/scram001/userg/regmod/isc3v2.pdf.

[2]     U.S.  EPA.  2005. Revision to  the  Guideline on Air Quality Models: Adoption of a
       Preferred General Purpose  (Flat  and Complex  Terrain)  Dispersion Model and Other
       Revisions; Final Rule. 40 CFRPart 51. Washington, D.C.,  U.S. Environmental Protection
       Agency. January 2008. http://www.epa.gov/ttn/scram/guidance/guide/appw 05.pdf.

[3]     Pope, C.A.  1991. Respiratory hospital admissions associated with PM10 pollution in
       Utah, Salt Lake, and Cache Valleys. Archives of Environmental Health 46(2):90-97.

[4]     Hinds,  W.  C.  1999. Aerosol Technology:  Properties,  Behavior, and  Measurement of
       Airborne Particles, 2nd Edition. John Wiley & Sons, New York. 233-242.

[5]     40 CFR 50.6. National primary and secondary ambient air quality standards for PMi0.

[6]     40 CFR 50.7. National primary and secondary ambient air quality standards for PM2.5.

[7]     Williams, D.J.,  Chilingaryan, S.,  Hatfield,  J. 2012. Los  Banos, CA Fall 2007  Tillage
       Campaign:  Data Analysis.  U.S. Environmental  Protection Agency Report EPA/600/R-
       12/734,   U.S.   Government   Printing   Office,   Washington,   D.C.  Available:
       http://cfpub.epa.gov/si/si_public_record_report.cfm?dirEntryId=248752.  Date accessed:
       March 1,2013.

[8]     Holmen, B.A., Eichinger, W.E., Flocchini, R.G.  1998. Application of elastic LIDAR to
       PMio emissions from agricultural  nonpoint  sources.  Environmental  Science  and
       Technology 32:3068-3076.

[9]     Conservation Management Practices Program Report. January 2006. San Joaquin Valley
       Air Pollution Control District for 2005.

[10]    Holmen,  B.A.,  James, T.A., Ashbaugh,  J.L., Flocchini,  R.G. 2001a. LIDAR-assisted
       measurement  of PMio emissions  from agricultural tilling in California's San  Joaquin
       Valley—Part I. LIDAR. Atmos. Env. 35:3251-2364.

[11]    Holmen,  B.A.,  James, T.A., Ashbaugh,  J.L., Flocchini,  R.G. 2001b. LIDAR-assisted
       measurement  of PMIO emissions from agricultural tilling in California's San  Joaquin
       Valley—Part II: Emission factors.  Atmos. Env. 35:3265-3277.

[12]    Flocchini, R.G., James, T.A., Ashbaugh, L.L., Brown, M.S., Carvacho,  O.F., Holmen,
       B.A., Matsumura,  R.T.,  Trezpla-Nabalgo,  K.,  Tsubamoto,  C. 2001. Interim  Report:
       Sources and sinks of PMio in the  San Joaquin Valley. Crocker Nuclear Laboratory, UC-
       Davis, CA.
SDL/08-556                      California 2008 Tillage Campaign                             94

-------
[13]    California Air Resources Board (CARB). 2003 a. Area Source Methods Manual, Section
       7.4: Agricultural Land Preparation.

[14]    CARB.  2003b.  Area  Source Methods  Manual,  Section  7.5:  Agricultural  Harvest
       Operations.

[15]    U.S. Environmental Protection Agency (EPA).  2001. Procedures document for national
       emission inventory. Criteria Air Pollutants 1985-1999. EPA-454/R-0 1-006.

[16]    Madden, N.M., Southard, R.J., Mitchell, J.P. 2008. Conservation tillage reduces PM10
       emissions in dairy forage rotations. Atmospheric Environment 42:3795-3808.

[17]    Wang, I, Miller, D.R.,  Sammis, T.W., Hiscox, A.L., Yang, W.,  Holmen, B.A.  2010.
       Local dust emission factors for agricultural tilling operations.  Soil Science 175:194-200.
[18]    Kasumba, J., Holmen, B.A., Hiscox, A., Wang, I, Miller, D. 2011. Agricultural
       emissions from cotton field disking in Las  Cruces,  NM. Atmospheric Environment
       45:1668-1674.

[19]    Clausnitzer,  H.,  Singer,  MJ.   1996.  Respirable-dust  production from agricultural
       operations in the  Sacramento Valley, California. Journal of Environmental  Quality
       25:877-884.

[20]    Clausnitzer,  H.,  Singer,  MJ.  2000.  Environmental  influences  on respirable  dust
       production from   agricultural  operations  in  California.  Atmospheric  Environment
       34:1739-1745.

[21]    Baker, J.B., Southard, R.J., Mitchell, J.P. 2005. Agricultural dust production in standard
       and conservation tillage systems in  the San Joaquin Valley. Journal of Environmental
       Quality 34: 1260-1269.

[22]    Mitchell, J.P., Southard, R.J., Madden, N.M., Klonsky, K.M., Baker, J.B., De Moura,
       R.L., Horwath, W.R., Munk, D.S., Wroble, J.F., Hembree, K.J., Wallender, W.W. 2008.
       Transition to conservation tillage evaluated in San Joaquin Valley cotton and tomato
       rotations. Journal of California Agriculture 62(2): 74-79.

[23]    Veenstra, J.J., Horwath, W.R., Mitchell, J.P., Munk, D.S. Conservation tillage and cover
       cropping influence soil  properties in San Joaquin Valley cotton-tomato crop.  Journal of
       California Agriculture 60(3):146-153.

[24]    Upadhyaya, S.K.,  Lancas, K.P.,  Santos-Filno, A.G., Raghuwanshi, N.S. 2001. One-pass
       tillage equipment outstrips conventional tillage method. Journal of California Agriculture
       55(5);44_47.

[25]    Mitchell, J.P., Munk, D.S., Prys, B., Klonsky, K.K., Wroble, J.F., De Moura, R.L. 2006.
       Conservation tillage production systems compared in San Joaquin Valley cotton. Journal
       of California Agriculture 60(3): 140-145.
SDL/08-556                      California 2008 Tillage Campaign                             95

-------
[26]    U.S.  EPA.  March  17,  2008.  NONROAD Model (nonroad  engines, equipment, and
       vehicles).        Accessed:        August        21,        2008.        Available:
       http://www.epa.gov/otaq/nonrdmdl.htm#techrept.

[27]    U.S. EPA. 2004. Exhaust and Crankcase Emission Factors for Nonroad Engine Modeling
       - Compression-Ignition. EPA420-P-04-009. April 2004.

[28]    U.S. EPA. 1991. Nonroad Engine and Vehicle Emission Study - Report. EPA 460/3-91-
       02. November 1991.

[29]    CARB. 1999. Emissions inventory of off-road large compression-ignited engines (>25hp)
       using the new OFFROAD Emissions Model. Mail-Out #MSC99-32. December 1999.

[30]    Kean, A.J.,  Sawyer, R.F.,  Harley, R.A. 2000. A fuel-based assessment of off-road diesel
       engine emissions. Journal of the Air and Waste Management Association 50:1929-1939.

[31]    RTI International. 2007. Emissions Factor Uncertainty Assessment, Review Draft. 29
       March 2013. http://www.epa.gov/ttn/chief/efpac/uncertainty.html.

[32]    USDA National Resource Conservation Service (NRCS). 2009. Web Soil Survey 2.0. 2
       January 2009. http://websoilsurvey.nrcs.usda.gov/app/.

[33]    Geology.com. May  2009. http://geology.com/state-map/california.shtml.

[34]    Live Search Maps. December 2008. http://maps.live.com.

[35]    California Irrigation Management Information  System (CEVIIS). 2009. Data for  Station
       #15   (Stratford)   for  May   and  June   of  2005   through  2007.  July  2009.
       http://wwwcimis.water.ca.gov/cimis/data.jsp.

[36]    Cooper,  D.C., Alley,  F.C. 2002. Air pollution control: A design approach. Waveland
       Press Inc. Prospect Heights, Illinois. 575.

[37]    Doran, J.W.,  Jones,  A.  1996. Methods for assessing soil  quality.  SSSA Special
       Publication Number 49. Soil Science Society of America. Madison, Wisconsin.

[38]    Soil Sampling and  Methods of Analysis,  ed. by M.R. Carter,  Canadian  Society  of Soil
       Science. Lewis Publishers, 1993:508-509.

[39]    Ibid. 659-662.

[40]    Chen, F.-L., Williams, R., Svendsen, E., Yeatts, K.,  Creason,  J., Scott, J., Terrell, D.,
       Case,  M. 2007. Coarse  particulate matter concentrations from residential  outdoor  sites
       associated with the North Carolina Asthma  and  Children's Environment  Studies (NC-
       ACES). Atmospheric Environment 41:1200-1208.
SDL/08-556                      California 2008 Tillage Campaign                             96

-------
[41]    Chow, J. C., Watson, J. G., Lowenthal, D. H., Chen, L.-W A., Tropp, R. I, Park, K.,
       Magliano, K. A. 2006. PM2.5 and PM10 Mass Measurements in California's San Joaquin
       Valley. Aerosol Science and Technology 40(10):796-810.

[42]    Airmetrics MiniVol Portable Air Sampler Operation Manual v. 5.

[43]    Hinds, W. C.  1999. Aerosol  Technology: Properties, Behavior,  and Measurement of
       Airborne Particles, 2nd Edition. John Wiley & Sons, New York. 75-82.

[44]    Rupprecht & Patashnick, n.d.  Series 5400 Elemental  Carbon/Organic Carbon Analyzer
       Instrument Manual.

[45]    Malm, W.C. and J.L. Hand. 2007. An examination of the physical  and optical properties
       of aerosols collected in  the IMPROVE program. Atmospheric Environment, 41, 3407-
       3427.

[46]    Marchant, C.  2008. Algorithm Development of the AGLITE-LIDAR Instrument, MS
       Thesis, Utah State University.

[47]    Zavyalov, V.  V., Marchant, C. C., Bingham,  G.  E., Wilkerson, T. D., Hatfield, J. L.,
       Martin, R. S., Silva, P. J., Moore, K. D., Swasey,  J., Ahlstrom, D. J., Jones, T.  L. 2009.
       Aglite lidar: Calibration and retrievals of well  characterized aerosols from agricultural
       operations using a  three-wavelength elastic lidar. Journal of Applied Remote Sensing
       3(1):033522 [doi: 10.1117/12.833365].

[48]    Klett, J.D. 1985. LIDAR inversion with variable backscatter/extinction ratio. Appl. Opt.
       24: 1638-83.

[49]    U.S. EPA. 2009. Guideline on Air Quality Models.  Appendix W. U.S.  Code of Fed.
       Regulations,        40        CFR        Part        51.        May        2009.
       http ://www. epa. gov/scramOO 1/guidance/guide/appw_05. pdf.

[50]    Cooper,  D.C.,  Alley,  F.C.  2002. Air pollution control: A design  approach. Waveland
       Press Inc. Prospect Heights, Illinois. 611-626.

[51]    Turner, D.B. 1970.  Workbook of Atmospheric Dispersion  Estimates. Washington, D.C.,
       U.S. Environmental Protection Agency.

[52]    Paine, R.J., Lee, R.F., Erode, R., Wilson,  R.B., Cimorelli, A.J., Perry, S.G, Weil, J.C.,
       Venkatram, A., Peters,  W.D.  1998. Model evaluation results for  AERMOD. Research
       Triangle  Park,  NC:U.S. Environmental  Protection  Agency, Office  of Air Quality
       Planning and  Standards  Emissions,  Monitoring,  Analysis Division.  March  2008.
       http://www.epa.gov/scram001/7thconf/aermod/evalrep.pdf.

[53]    National Research Council, National Academies of Science. 2003. Air emissions from
       animal feeding operations: Current knowledge, future needs. The National Academies
       Press. Washington, D.C.  95.
SDL/08-556                      California 2008 Tillage Campaign                             97

-------
[54]    Arya, S.P. 1998. Air Pollution Meteorology and Dispersion. Oxford University Press.

[55]    Lakes Environmental. 2009. Terrain Data:  7.5-Min DEM Native Format - United States.
       July 2009. http://www.webgis.com/terr us75m.html.

[56]    U.S. Department of Agriculture, Natural Resources Conservation Service. 2007. National
       Soil Survey Handbook, title 430-VI. Available: http://soils.usda.gov/technical/handbook/.

[57]    Bingham, G.E., Marchant, C.  C., Zavyalov, V. V., Ahlstrom, D. I, Moore, K. D., Jones,
       D. S., Wilkerson,  T.  D.,  Hipps, L.  E., Martin, R. S., Hatfield,  J. L.,  Prueger, J. H.,
       Pfeiffer,  R. L.  2009.  Lidar based  emissions measurement at the whole facility scale:
       Method  and  error analysis.  Journal of Applied  Remote Sensing  3(1):033510  [doi:
       10.1117/12.829411].

[58]    Cooper, D.C., Alley,  F.C. 2002. Air pollution control: A  design approach.  Waveland
       Press Inc. Prospect Heights, Illinois. 100.

[59]    Hinds, W.  C. 1999.  Aerosol Technology: Properties, Behavior,  and Measurement of
       Airborne Particles, 2nd Edition. John Wiley & Sons, New York. 402-408.
SDL/08-556                      California 2008 Tillage Campaign                              98

-------
                                10.   APPENDICES
10.1   APPENDIX A: DATA AND SETTINGS TABLES

Table 30. Settings for the ISCST3 and AERMOD dispersion models for the tillage study in the ISC-
AERMOD View software by Lakes Environmental, Inc. All settings were held constant across the sample
periods except the source area size and shape, which changed each day, and the downwind receptor locations,
which were specific to each field studies. (— = not applicable)
Setting
ISCST3
AERMOD
Control Pathway
Dispersion Options
Output type
Plume Deposition
Pollutant
Averaging Time
Dispersion Coefficient
Terrain Height Options
Terrain Calculation
Algorithms (ISC), Receptor
Elev./Hill Hghts (AERMOD)
Regulatory Default
Concentration
None
Other - PM
Period
Rural
Elevated
Simple terrain only
Regulatory Default
Concentration
—
Other - PM
Period
Rural
Elevated
Run using the AERMAP Receptor
Output file

Source Pathway
Source type
Base Elevation
Release height
Emission rate
Initial Vertical Dim.
of the plume
Building downwash
Area Poly
0.0m
0.0 m AGL
8.6 E-6 g/(s-m2)
Blank
None
Area Poly
0.0m
0.0 m AGL
8.6 E-6 g/(s-m2)
Blank
None

Receptor Pathway
Uniform Cartesian Grid
(# Receptors: 4200)
Discrete Cartesian Receptors
(# Receptors: 10)
138x130, lOxlOm spacing,
flagpole height z = 2.0 m AGL
Placed at sample locations, z = 2, 5,
and 9 m AGL
138x130, lOxlOm spacing,
flagpole height z = 2.0 m AGL
Placed at sample locations, z = 2, 5,
and 9 m AGL

AERMET View Settings
Hourly Surface Data
Adjustment to Local Time
Application Station
Elevation MSL
Upper Air Data
Mode
Sectors Parameters
Time Zone
Randomize NWS Wind
Directions
Anemometer Height
Wind direction sectors
Land Use Type
—
—
—
—
—
—
—
—
—
—
—
Source: on-site data, mixing height =
1000 m AGL
8 hr (Pacific)
76.2m
Source: calculated based on on-site
data, mixing height = 1000 m AGL
Upper Air Estimator

8 (Pacific)
Yes
6.2 m AGL
1 : Start = 0°, End =360°
Cultivated Land
SDL/08-556
California 2008 Tillage Campaign
99

-------
Setting
Surface parameters per
sector
ISCST3

AERMOD
Annually, using spring-time default
values for Midday Albedo = 0.14,
Bowen Ratio = 0.3, and Surface
Roughness = 0.03 m

Meteorology Pathway
Surface Met Data
Profile Met Data
Anemometer Height (agl)
Primary Met Tower Base
Elevation above MSL
Read Entire Met Data File
Specify Data Periods to
Process
Wind Speed Categories
Source: on-site data, mixing height =
1000 m AGL
—
6.2m
—
No
Set to sample period times, varied by
sample
Default
Source: calculated from on-site data
and default values
Source: estimated from on-site data
—
76.2m
No
Set to sample period times, varied by
sample
Default

Output Pathway
Tabular Outputs
All

Not available, no short term
averaging times selected

Not available, no short term
averaging times selected

Buildings
None
None

Terrain
Calculated values from AERMAP
using 7.5 Min DEM
Calculated values from AERMAP
using 7.5 Min DEM
SDL/08-556
California 2008 Tillage Campaign
100

-------
Table 31. Calculated PM concentrations (|ig/m3) measured during May 2008 at all sample locations.
Location
(height
agl)
Tl(9)
10.0 (2)
11.0(2)
T2(9)
T3(9)
AQT (5)
2.4/2.5 (2)
6.4/6.5 (2)
7.4/7.5 (2)
8.4/8.5 (2)
9.4/9.5 (2)
12.0 (2)
Size
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
PM10
PM25
TSP
PM10
PM25
PM10
PM25
PM10
PM25
TSP
PM10
PM25
Sample Period
May 17
Rl
138.2
64.6
59.7
135.4
75.7
43.5






559.3
154.4
48.8
380.0
167.0
53.9
610.7
161.3
49.4
157.7
65.2

140.5
62.3
191.4
57.4
140.6
65.9



May 17
R2
330.0
239.3
128.1
628.2
205.5
266.8



533.8
199.6
92.0



368.3
109.0
141.6
*
186.4
187.8
1458.4
3244.9

83.8
153.8
300.9
179.8
203.6
144.4



May 18
118.2
*
133.8
266.1
38.1
27.1



479.6
100.1
46.2



154.3
59.4
65.1
1233.4
135.3
63.3
140.2
90.1

75.7
60.9
141.2
36.9
149.1
*



May 19
Rl
242.0
83.2
35.2
178.6
*
29.2



1138.2
231.5
71.7



188.3
*
38.2
*
312.1
135.0
278.2
132.7

268.7
153.5
562.5
171.0
570.6
265.5



May 19
R2
153.5
65.1
23.2
267.9
*
24.8



965.0
195.9
41.4



224.5
76.9
34.5
1894.2
387.8
232.4
408.4
491.0

215.9
202.6
402.4
102.8
335.3
344.8



May 20
Rl
242.3
115.7
31.6
260.5
*
51.2



1511.7
253.5
77.5



286.1
98.1
52.3
*
347.1
459.7
287.3
465.9

250.9
398.0
*
284.9
347.3
392.7



May 20
R2
343.4
469.8
205.5
355.1
*
89.3



2276.9
258.4
324.4



709.1
172.9
107.3
*
399.2
215.7
730.6
171.1

364.8
136.5
437.0
139.3
348.9
186.7



        * Filter sample had a noted problem during or after sampling
        (blank) = no sample collected
SDL/08-556
California 2008 Tillage Campaign
101

-------
Table 32. Calculated PM concentrations (|ig/m3) measured during June 2008 at all sample locations.
Location
(height
agl)
Tl(9)
10.0 (2)
11.0(2)
T2(9)
T3(9)
AQT (5)
2.4/2.5 (2)
6.4/6.5 (2)
7.4/7.5 (2)
8.4/8.5 (2)
9.4/9.5 (2)
12.0 (2)
Size
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
PM10
PM25
TSP
PM10
PM25
PM10
PM25
PM10
PM25
TSP
PM10
PM25
Sam
June 5
Rl
102.9
47.4
26.7
227.9
86.2
101.5



*
58.0
32.2



104.3
58.6
38.2
*
80.4
57.6
102.5
62.6

53.5
42.8
80.8
86.9
77.9
43.9



June 5
R2
158.5
*
68.0
231.4
272.6
165.2



353.8
109.6
138.8



243.8
221.3
252.5
*
137.5
174.3
229.4
123.1

246.2
140.4
163.4
110.0
131.6
114.6



June 7
110.7
57.2
27.5
*
80.7
38.8






204.6
61.5
29.5
158.7
78.0
30.8
335.7
80.3
47.8
78.0
36.9

59.0
33.2
88.3
56.1
98.4
33.1



pie Period
June 11
73.6
188.8
117.9
183.1
137.6
106.3






262.9
96.7
76.1
250.0
75.1
122.7
246.9
80.2
213.6
172.9
102.6

176.3
149.8
182.6
73.2
111.9
101.4



June 18
195.4
56.5
30.1



233.6
68.6
28.2









1896.9
333.6
42.9
489.4
38.9
1495.1
195.3
60.2
207.1
35.1
61.6
30.0



June 24
180.1
101.8
71.2



195.6
89.6
64.4









554.2
142.2
74.0
116.7
70.7
717.0
142.4
64.3
126.9
76.4
196.9
65.0
700.7
203.4
74.9
        * Filter sample had a noted problem during or after sampling
        (blank) = no sample collected
SDL/08-556
California 2008 Tillage Campaign
102

-------
Table 33. PM concentrations (|ig/m3) used in emission rate calculations from sample periods in May 2008.
Location
(height
agl)
Tl(9)
10.0 (2)
11.0(2)
T2(9)
T3(9)
AQT (5)
2.4/2.5 (2)
6.4/6.5 (2)
7.4/7.5 (2)
8.4/8.5 (2)
9.4/9.5 (2)
12.0 (2)
Size
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
PM10
PM25
TSP
PM10
PM25
PM10
PM25
PM10
PM25
TSP
PM10
PM25
Sample Period
May 17
Rl
138.2
64.6
59.7
135.4
75.7
43.5






559.3
154.4
48.8
380.0
167.0
53.9
610.7
161.3
49.4
157.7
65.2

140.5
62.3
191.4
57.4
140.6
65.9



May 17
R2
330.0
—
—
+
+
+



533.8
§
§



368.3
§
§
*
§
§
§
§

§
§
§
§
§
§



May 18
§
*
§
§
§
§



§
§
§



§
§
§
§
§
§
§
§

§
§
§
§
§
*



May 19
Rl
§
§
§
§
*
§



§
§
§



§
*
§
*
§
—
§
§

§
§
§
§
§
§



May 19
R2
§
§
§
§
*
§



§
§
§



§
§
§
§
§
§
§
§

§
§
§
§
§
§



May 20
Rl
§
§
§
§
*
§



§
§
§



§
§
§
*
§
§
§
§

§
§
*
§
§
§



May 20
R2
§
§
§
§
*
§



§
§
§



§
§
§
*
§
§
§
§

§
§
§
§
§
§



                * = Filter sample had a noted problem during or after sampling
                — = Removed due to ring particle density classification of medium or higher
                + = Removed due to exceptionally high or low value and not supported by OPC time series
                0 = PM2 5 and PM10 levels indicated filters switched during sampling, concentrations switched
                § = Removed due to contaminated background or overloaded downwind filters
                (blank) = no sample  collected
SDL/08-556
California 2008 Tillage Campaign
103

-------
Table 34. PM concentrations (|ig/m3) used in emission rate calculations from sample periods in June 2008.
Location
(height
agl)
Tl(9)
10.0 (2)
11.0(2)
T2(9)
T3(9)
AQT (5)
2.4/2.5 (2)
6.4/6.5 (2)
7.4/7.5 (2)
8.4/8.5 (2)
9.4/9.5 (2)
12.0 (2)
Size
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
PM10
PM25
TSP
PM10
PM25
PM10
PM25
PM10
PM25
TSP
PM10
PM25
Sam
June 5
Rl
102.9
47.4
26.7
§
—
—



*
58.0
32.2



104.3
58.6
38.2
*
80.4
—
102.5
—

53.5
42.8
80.8
—
77.9
43.9



June 5
R2
158.5
*
68.0
§
—




353.8
§
—



243.8
—
+
*
§
—
§
123.1

§
140.4
§
—
§
114.6



June 7
110.7
57.2
27.5
*
80.7
38.8






204.6
61.5
29.5
158.7
—
30.8
335.7
80.3
—
—
—

59.0
33.2
88.3
56.1
98.4
33.1



pie Period
June 11
+
188.8
117.9
183.1
137.6
106.3






262.9
96.7
76.1
250.0
122.7 0
75.10
246.9
213.60
80.20
172.9
102.6

176.3
149.8
182.6
73.2
111.9
101.4



June 18
195.4
56.5
30.1



233.6
68.6
28.2









1896.9
333.6
42.9
489.4
38.9
1495.1
195.3
60.2
207.1
35.1
61.6
30.0



June 24
180.1
101.8
71.2



195.6
89.6
64.4









554.2
142.2
74.0
116.7
70.7
717.0
142.4
64.3
126.9
76.4
196.9
65.0
700.7
203.4
74.9
                * = Filter sample had a noted problem during or after sampling
                — = Removed due to ring particle density classification of medium or higher
                + = Removed due to exceptionally high or low and not supported by OPC time series
                0 = PM2 5 and PM10 levels indicated filters switched during sampling, concentrations switched
                § = Removed due to contaminated background or overloaded downwind filters
                (blank) = no sample  collected
SDL/08-556
California 2008 Tillage Campaign
104

-------
Table 35. PM concentrations (|ig/m3) as measured by OPCs from sample periods in May 2008. (OPC
V, x MCF,)
Location
(height
agl)
Tl(9)
10.0 (2)
11.0(2)
T2(9)
T3(9)
AQT (5)
2.4/2.5 (2)
7.4/7.5 (2)
8.4/8.5 (2)
12.0 (2)
Size
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
Sample Period
May 17
Rl
95.3
50.6
9.2









597.1
142.3
12.2
636.3
174.7
13.2
601.4
165.2
12.9
816.6
199.0
13.7






May 17
R2
181.6
68.6
4.3






1777.6
197.0
6.3



417.1
75.5
4.8
3251.9
318.8
7.5
2767.5
292.0
7.7






May 18
109.1
48.9
7.7






490.0
110.3
9.7



164.6
54.4
7.9
813.2
178.0
11.1
692.0
141.5
10.1






May 19
Rl
181.7
70.3
12.8
159.0
61.9
12.4



1585.2
283.4
18.9



227.8
74.0
12.9
2263.5
375.8
19.9
2523.2
406.8
20.5






May 19
R2
127.8
58.6
8.4
275.9
89.1
9.7



1296.9
241.6
12.5



207.1
71.3
8.9
3271.9
641.1
20.0
2553.9
451.5
15.9






May 20
Rl
224.7
74.5
12.6
296.3
87.9
12.2



1282.2
252.3
16.5



331.3
102.3
13.2
2550.4
462.7
20.7
1656.4
342.1
18.5






May 20
R2






























SDL/08-556
California 2008 Tillage Campaign
105

-------
Table 36. PM concentrations (|ig/m3) as measured by OPCs from sample periods in June 2008. (OPC
V, x MCF,)
Location
(height
agl)
Tl(9)
10.0 (2)
11.0(2)
T2(9)
T3(9)
AQT (5)
2.4/2.5 (2)
7.4/7.5 (2)
8.4/8.5 (2)
12.0 (2)
Size
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
Sam
June 5
Rl
95.3
41.2
11.7
172.4
69.5
13.6



205.9
63.8
13.3



113.3
47.2
11.6
386.1
93.2
14.7
180.4
64.0
12.6






June 5
R2
170.5
47.7
5.7
571.5
160.2
12.1



364.4
80.9
6.9



221.8
67.2
6.4
1162 A
213.0
12.9
769.2
193.9
11.6






June 7
104.8
43.1
16.2
194.6
94.4
19.3






190.9
59.2
16.9
194.2
62.3
16.8
318.3
83.4
18.3
245.8
72.2
17.1






pie Period
June 11
222.4
122.4
11.4
198.4
120.7
10.7






213.4
132.6
11.2
257.4
141.1
11.4
282.4
155.8
11.4
357.1
171.0
12.0






June 18
179.5
47.0
4.8
153.6
44.2
5.4



475.1
97.0
6.4






2816.4
467.4
22.0
2084.4
295.0
12.7
1703.5
214.5
9.6



June 24
127.0
85.0
56.9



172.4
81.7
55.8
334.9
109.4
57.8






720.8
165.8
60.2
822.3
177.1
58.5



868.3
203.6
58.3
SDL/08-556
California 2008 Tillage Campaign
106

-------
Table 37. PM concentrations (|ig/m3) as measured by OPCs used in emission rate calculations from sample
periods in May 2008. (OPC PMft = Vk x MCFft)
Location
(height
agl)
Tl(9)
10.0 (2)
11.0(2)
T2(9)
T3(9)
AQT (5)
2.4/2.5 (2)
7.4/7.5 (2)
8.4/8.5 (2)
12.0 (2)
Size
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
Sample Period
May 17
Rl
95.3
50.6
9.2









597.1
142.3
12.2
636.3
174.7
13.2
601.4
165.2
12.9
816.6
199.0
13.7






May 17
R2
181.6
68.6
4.3






1777.6
197.0
6.3



417.1
75.5
4.8
3251.9
318.8
7.5
2767.5
292.0
7.7






May 18
109.1
48.9
7.7






490.0
110.3
9.7



164.6
54.4
7.9
813.2
178.0
11.1
692.0
141.5
10.1






May 19
Rl
156.5*
63.3*
12.6*
159.0
61.9
12.4



1585.2
283.4
18.9



227.8
74.0
12.9
2263.5
375.8
19.9
2523.2
406.8
20.5






May 19
R2
127.8
58.6
8.4
137.1*
56.1*
8.4*



1296.9
241.6
12.5



207.1
71.3
8.9
3271.9
641.1
20.0
2553.9
451.5
15.9






May 20
Rl
224.7
74.5
12.6
169.0*
64.2*
11.7*



1282.2
252.3
16.5



331.3
102.3
13.2
2550.4
462.7
20.7
1656.4
342.1
18.5






May 20
R2






























           Significant plumes at upwind location removed
SDL/08-556
California 2008 Tillage Campaign
107

-------
Table 38. PM concentrations (|ig/m3) as measured by OPCs used in emission rate calculations from sample
periods in June 2008. (OPC PM* = Vk x MCFft)
Location
(height
agl)
Tl(9)
10.0 (2)
11.0(2)
T2(9)
T3(9)
AQT (5)
2.4/2.5 (2)
7.4/7.5 (2)
8.4/8.5 (2)
12.0 (2)
Size
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
TSP
PM10
PM25
Sam
June 5
Rl
95.3
41.2
11.7
93.9*
41.0*
11.8*



205.9
63.8
13.3



113.3
47.2
11.6
386.1
93.2
14.7
180.4
64.0
12.6






June 5
R2
170.5
47.7
5.7
127.9*
42.6*
6.2*



364.4
80.9
6.9



221.8
67.2
6.4
1162 A
213.0
12.9
769.2
193.9
11.6






June 7
104.8
43.1
16.2
98.6*
41.9*
16.3*






190.9
59.2
16.9
194.2
62.3
16.8
318.3
83.4
18.3
245.8
72.2
17.1






pie Period
June 11
222.4
122.4
11.4
198.4
120.7
10.7






213.4
132.6
11.2
257.4
141.1
11.4
282.4
155.8
11.4
357.1
171.0
12.0






June 18
179.5
47.0
4.8
153.6
44.2
5.4



475.1
97.0
6.4






2816.4
467.4
22.0
2084.4
295.0
12.7
1703.5
214.5
9.6



June 24
127.0
85.0
56.9



172.4
81.7
55.8
334.9
109.4
57.8






720.8
165.8
60.2
822.3
177.1
58.5



868.3
203.6
58.3
          Significant plumes at upwind location removed
10.2
       APPENDIX B: INVESTIGATIONS INTO AND CONCLUSIONS FROM FILTER-
       BASED DATA
During the post-weighing process it was noticed that filters used for PM2.s, PMio, and TSP had
particles large enough to be visible to the naked eye, not all of which were embedded in the filter
medium and thus could move around and off the filter. Particles small enough to pass through
the impactor assembly in the sample head are typically too small to be seen with the naked eye,
almost always being less than 10 jam and 2.5 jam in diameter for PMio and PM2.5, respectively.
Particles collected during sampling that were not stuck into or onto the filter medium may move
off of the filter, yielding a variable low bias in reported PM concentrations. In addition, several
of the calculated PM2.5 concentrations significantly exceeded the collocated PMio concentrations,
which is theoretically  impossible under the ideal sampling  scenario since PM2.5 is a subset  of
PMio and  should always be less than or equal to the PMio concentration. These inverted PM2.5
and PMio concentrations, as we will call them, were very prevalent in the data for several sample
periods, with up to 75% of the downwind PM2.s and PMio concentrations inverted for the lister
SDL/08-556
                               California 2008 Tillage Campaign
108

-------
pass on May 20, 2008. These size-inverted concentrations and the presence of visible particles on
the filters prompted an investigation to determine their causes as well as potential  solutions for
the current data set and  prevention methods for future  measurements. This section  describes
these investigations, as well as the conclusions drawn about the suitability of the dataset for use
in emission rate calculations, and options for preventing such problems in the future.
All aspects of the pre-, intra-,  and  post-sampling filter  handling were examined  in  detail for
potential problems. Filter identification numbers were  recorded throughout the  preparation,
sample, and  analysis  phases, which allowed  each  filter to be tracked throughout the entire
process. It was thought that the cause could be improper and/or inaccurate weighing  methods and
instruments. This possibility, however, was ruled  out as all filters were preconditioned according
to U.S. EPA standards  in dessicators before both pre-  and post-weights  were determined.
Measurements of weight  to the  1 ng (1x10~6 g) level were  performed by experienced personnel
using a vibration-isolated micro-balance at the UWRL that had  been calibrated on-site  by
manufacturer personnel earlier in 2008; a certified 1.000 mg calibration mass was used to
monitor balance accuracy and drift every ten readings. The correct calculation and use of average
pre- and post-weights for all filters were verified, and the filter numbers recorded on sample run
sheets were checked for potential errors. A few errors were found during this verification process
in the calculation and use of average pre- and post-weights, but they did not explain the majority
of the suspect filters with the problems stated above.  The errors found have been corrected in the
concentrations reported in Section 4.2.1 and found in Table 31 and Table 32.
Another potential cause of inverted collocated PM2.s-PMio concentrations was that the filters
were switched during sampling due to human error, causing the filter designated  for PM2.s to
actually sample PMio and vice versa. Data were examined on a location-by-location basis by
switching the suspected PM2.5 and  PMio  concentrations and  comparing  them with other
downwind PM concentrations of the  same size fraction measured during the  same sample period.
The results  suggested  that this likely occurred  on a total of four occasions at  two  sample
locations and across two sample periods. Switching these PM2.5 and PMio concentrations yielded
a more consistent set of concentrations in  all cases. In addition, examination of the OPC time
series yielded no irregularities that would suggest problems with either one or both samples. The
collocated PM2.s  and PMio concentrations were therefore switched for subsequent analysis.
Similar to the verification of  correct filter tracking  and  the average  weights used, this
investigation  found some inverted collocated PM2.s-PMio concentrations could be explained by
the filters being switched during sampling but  that a significant number of instances of the
phenomena were unexplained. It was determined  that this was likely not the cause in most cases
due to the pervasiveness of this phenomenon during only a few sample periods and because the
personnel that were operating the filter-based measurements were experienced and capable.
Further investigations were made through  visual inspection of the  filters through  microscopy.
Several representative filters were examined through a microscope to  size and count collected
particles. Examined filters included some from each of the following categories:  collocated
PM2.s and PMio samples where  the PM2.5-PMi0 concentrations were inverted; samples collected
in upwind/background locations on days with multiple suspect filters; and blank filters that were
taken to the field but not used. The microscope used  was a BX51 from Olympus (Center Valley,
PA) with a stage that  could be controlled either manually or through computer software. An
Olympus Model DP71  camera was mounted on the microscope to capture digital images of the
microscope field of view. The MicroSuite Five Imaging  software, also by Olympus, controlled
SDL/08-556                      California 2008 Tillage Campaign                            109

-------
the camera and stage automatically to photograph a user-specified area of the filter, combine
adjacent images, and identify and size particles in the images. The 5x, lOx, 50x, and lOOx lenses
were all tested, but the software analysis on images taken using 50x and lOOx lenses under the
given light, microscope, and software settings were  not used due to inaccurate  sizing and
counting. The  5x lens was mainly used to scan over the filter area for large particles (> -25 jam
in diameter), while most images used for counting and sizing particles were taken with the lOx
lens because smaller particles could be detected. The minimum particle diameters detected by the
5x and lOx, lenses under the light, microscope, and software settings were 5.22 jam and 2.61 jam,
respectively. Projected area particle diameter (dpA) is the diameter of a circle having the same
area as the 2D projected area of the 3D particle and is  a common representation of particle size
based on  microscopic sizing  [59]; the dpA were calculated based  on  individual particle area
values provided by the MicroSuite Five Imaging software. Smaller particles could be visually
identified  in  the images,  but  the software  could not  sufficiently resolve  them  from  the
background light intensity variation inherent in the Teflon filter medium. The entire microscope
and computer imaging  system were housed in a room maintained at Class  100 clean room
standards to prevent particle deposition on the filters during analysis. All of the microscope work
was performed by just one person in order to maintain consistent procedures and settings.
The 47 mm diameter Teflon filters  used in PM mass sampling have a collection area composed
of an exposed  mesh of Teflon fibers about 41 mm-diameter and a 3 mm-wide plastic ring which
holds the Teflon fibers in place; this plastic ring is used for filter handling and also for securing
the filter into the MiniVol sampling fixture. The Teflon area is fragile. The sample collection
area is about 1320 mm . The sampling fixture is a two-part, push-on assembly which sandwiches
the Teflon filter between two plastic fittings. Figure 43  shows an example of an image of a PMio
filter  sample  collection area taken using the microscope and the lOx  lens. This image is
composed of four adjacent pictures that were automatically combined by the software. Note the
varying shades of the grey background from the Teflon fibers. Particles are shown as both dark
and bright white spots, depending on chemical composition, size, and orientation with respect to
the light source and microscope lens. The average dpA ± la of detected particles in this image
was 4.47 ± 2.56 jam (n   2977) and the median mean diameter was 3.50 jam with 25 and 75
percentile values of 2.86 and 5.08 jam; 95.9 % of the detected particles  had a dpA less than 10.0
jam. The largest particle, marked by the blue ring, was measured at a dpA of 31.94 jam.
Due to the magnification and time limits of this analysis at each position, a total of about 1% of
the collection  area on most filters was sampled after integration  over three to four sample
locations.  Sample locations were chosen randomly within quadrants, with each sample from a
quadrant that was previously not sampled. While a total sample area  of at least 10% is preferred,
the variability of the cumulative distribution between sample locations on the same filter was less
than 10%. Therefore, multiple  sample areas per filter with a total area of  about  1%  of the
potential collection area were deemed adequate for the purposes of this analysis.
SDL/08-556                      California 2008 Tillage Campaign                            110

-------
Figure 43. Four adjacent microscopic images taken with the lOx lens that have been combined for
determining the particle size distribution and count at this location on a PM10 sample filter. The largest
particle with dPA = 31.94 |im is shown by the blue circle.

Results of the microscope particle sizing and counting on two  sets of PM2.5 and PMio samples
where  the concentrations were inverted  showed that the dpA size distributions on  the sample
collection areas were  similar in shape,  with  the PMio filters having  a  much higher particle
density than  the PM2.5 filter, as seen in Figure 44. Note that  the size distribution  is given in
number of particles per square mm of filter surface per jam of bin width; incorporating the bin
width into the  denominator removes the bin number concentration dependence on bin width.
Also, it must be remembered that the MDL was 2.61 jam and that dpA is related to but different
from aerodynamic diameter (daero). For a given particle, dpA and daero values should be similar but
will  vary according to shape,  density, and orientation at the time of detection.  These data,
coupled with visual scans of the entire collection area to find and size large particles, suggest that
the inverted  PM2.5-PMio concentrations  for these filters cannot likely be  explained by large
particle contamination on the PM2.5  collection  area. Additionally, the data show that there was
neither  a complete failure of  the size fractionating impactors nor were many of the filters
switched during sampling. Very similar particle count densities in each size fraction between the
two downwind sites, 7.4 and 2.4, suggest  that in-plume number size distribution was roughly
spatially homogenous during the May 20 Run 1 sample period, assuming spatially homogeneous
deposition on the filter area. The differences between the upwind  PM2.5 size distribution and the
downwind PM2.5 size distributions are due to the plumes that impacted the downwind samplers.
SDL/08-556
California 2008 Tillage Campaign
111

-------
A potentially significant number of larger-than-expected particles (dpA > ~2.5 jam for PM2.5 and
dpA > ~10 jam for PMio) were found on the downwind PMio and PM2.5 filters; there were even
particles greater than 100 jam in diameter. Particles with dpA greater than 25 jam were found on
the downwind PM2.s and PMio filters whose size distributions are presented in Figure 44.  Such
large particles are not unexpected on  TSP  samples collected downwind of agricultural tillage
operations, but they should  not be  present on  samples with a PMio or PM2.5 size-separating
device upstream of the filter in the sample head assembly. In addition, large particles were found
on the annular plastic ring of many of the examined filters by both the microscope and inspection
with the naked eye. This was  troubling since this annular area is  completely  covered by the
sampling fixture itself and  therefore cannot collect particles  during active sampling. Most
observed particles on the annular filter ring area were sitting on the surface, suggesting either
migration from the collection area or deposition after removal from the sampling device. Several
particles were even pressed into the plastic,  suggesting that they were either present on the filter
prior to being loaded for sampling or were on the bottom  of the filter holder piece when pressed
down onto the plastic ring medium and remained there upon removal from the sampling fixture.
            •Upwind PM2.5
•7.4 PMIO
•2.4 PMIO
7.4PM2.5
•2.4PM2.5
      300
    E
      250 -
      200 -
      150 -
      100 -
       50 -
                                      5/20 Run 1
                                                                                    100
Figure 44. Number distributions based on projected area diameter as measured via microscopy for two
downwind and one upwind PM2.S filters and two downwind PM10 filters used during the Lister pass on May
20, 2008.

The presence of such large particles on both the collection area and the annular area of multiple
filters may be explained by several potential factors, including but not limited to:
SDL/08-556
  California 2008 Tillage Campaign
                                           112

-------
   1)  The  collection  efficiency curve of  the  PMio  and PM2.5  size  fractionation  devices
       employed by both FRMs and MiniVols are S-curves with respect to particle aerodynamic
       diameter, designed to mimic the particle removal efficiency of the  human respiratory
       system. The removal efficiency curves have a 50% collection efficiency at the designed
       cut-off size, with some smaller particles being removed and some larger particles passing
       through the system. The potential exists for high numbers of large particles to be present
       in plumes emitted from sources such as agricultural tillage activities and, if the plume is
       being sampled by an instrument a short distance away as was the case  in this study, some
       particles larger than the cut size would be expected to pass through  the size-separating
       devices.  This potential factor would best  apply to particles close to  the cut size as the
       collection efficiency curve theoretically approaches  100% soon after  passing the design
       cut-off diameter, with the MiniVol  S-curve being less steep than  the FRM and
       approaching 100% at a larger size.

   2)  Improper assembly of the size-fractionating impactor is another potential explanation for
       passing  larger-than-expected particles through  the  impactor. While possible  on an
       individual basis, such an improper assembly problem across up to 75% of the downwind
       samplers is not  likely (75% of downwind PMio and  PM2.5 samples  had significant
       numbers of visible  particles). In addition,  two full sets of sample heads were used
       throughout  the  study period to minimize the time required  for preparation between
       sample periods, and  the problematic PM concentrations were present in  filters which
       were placed in both sets of sample heads.

   3)  A phenomenon commonly referred  to as "particle bounce" may have occurred.  Particle
       bounce refers to when a particle collides with the impactor assembly  but then returns  to
       the airstream  and is collected downstream at  the filter. To aid in removal of particles
       colliding with the impactor plate, grease may be applied on the impaction surface. If too
       many particles  accumulate in  the grease,  however, the impaction surface loses its
       "stickiness" and larger particles may bounce off, leading to higher reported PM2.5/PMi0
       concentrations than  actually  existed. Impactor  plate  stickiness  is influenced  by  a
       combination of total exposure time and PM concentrations during exposure. The smaller
       effective impact area and increased  particle velocity in the  MiniVol PM2.5 impactor
       assembly makes it more susceptible to particle bounce than the PMio assembly under the
       same conditions, which could result  in  higher reported levels  of  PM2.5 than PMio.
       Airmetrics,  Inc. suggests  cleaning and regreasing of the impactor plates every five  to
       seven samples to maintain the design removal efficiency, but states that the need to renew
       the plate may change based on exposure levels. In the case of this study, silicone-based,
       high-vacuum  grease was applied to all impaction surfaces on-site prior to sampling  in
       May. Personnel noted that impactor assemblies in downwind samplers had collected
       significant amounts of particles near the end of the seven sample periods in that month
       (May 17-20), which  suggests that particle bounce likely occurred during  some sample
       periods.  The sample heads and impactor assemblies were cleaned and  greased during the
       break in measurements,  with  evaluation  and  re-greasing  if necessary  after each
       measurement  in June. In addition, the enhanced  susceptibility  of the  PM2.5 assembly  to
       particle bounce  could explain the inverted PM2.s and PMio concentrations observed  at
       downwind locations during the May 19 and May 20 sample periods.
SDL/08-556                      California 2008 Tillage Campaign                            113

-------
   4)  Exposure to dust plumes while the sampler was deployed but not actively pulling air
       through its system is another potential cause. In  such a case, settling of particles may
       allow some to reach the filter. Dirt access roads surround each field at this location and
       the standoff distance for the samplers usually put them on the downwind side of these
       access roads. While travel was limited to farm activity, it is possible that a vehicle driving
       on these roads may have passed shortly before a programmed start time and resulted in
       large particles being deposited on the filter.

   5)  Contamination prior to or after sampling is a potential cause that cannot be ruled out,
       especially in a dry, dusty environment. Filters were held in individual Petri  dishes at all
       times except during sample collection. The filters  and Petri dishes were stored on-site in
       dessicators. All required filter handling for sampling was performed on-site in a trailer
       over white laboratory paper which was changed as needed.  Personnel handling the filters
       wore latex gloves and used tweezers to move filters; the tweezers were rinsed with DDI
       water as needed. Any filters dropped prior to sampling were not used, and those that were
       dropped after sampling were noted and not used in further calculations. While the trailer
       was kept as clean as possible, the environment outside of the trailer was dusty. In fact,
       blowing dust was a problem during a couple sample periods,  especially May 20 Rl and
       R2.

While any one of these potential causes can significantly impact the  mass collected on a  filter,
there is evidence from examination of the filters under the microscope that a combination  of
several of these phenomena potentially occurred. It was suspected that contamination prior to or
after sampling accounted  for the largest  portion  of  the problems  in  the  MiniVol  mass
concentration data, with additional contributions at locations heavily impacted by agricultural
activity, or traffic on nearby dirt roads  from particle bounce, and the passage of large particles
due to the S-shaped collection efficiency curve of the impactor assembly.
While the microscopic analysis of the filters provided insight into the number/size of particles on
the filters and causes of possible contamination, it did not provide a feasible solution to correct
this problem in the current data set. Microscopically examining and analyzing both the collection
area and the  ring of each of the 296 filters used was prohibitively time intensive.  A surrogate
method for determining the compromised/contamination level of each filter was devised in order
to feasibly remove those samples that were compromised. The ring on each filter was visually
inspected by just one  person with a Doublet  lOx single  lens magnifier by Selsi Company, Inc.
(Midland Park, NJ) against a white background and the relative amount of particles found on the
ring was classified into one of the following  ring particle density  categories: 1) None, 2) Very
light, 3) Light, 4) Medium, 5) Heavy, 6) Very heavy, and 7) Extremely heavy. The following two
categories were also used for the specified reasons: 8) Not available - filters were not available
for visual inspections because the 1C ion analysis had already been  performed; and 9) Filters
with notes - filters that had noted problems either during or after sampling that could affect the
mass collected, such as being dropped after sampling or a malfunction of the flow system. While
this method is  subjective it has the virtue of minimizing the impact of grossly contaminated
filters on the final PM emission values.
A significant number of particles were found on the rings of upwind/background and downwind
samples for all three PM mass fractions  measured. In fact, none of  the inspected filters  had a
completely clean ring that would  correspond to the "None" category.  Table 39 presents the
SDL/08-556                      California 2008 Tillage Campaign                             114

-------
results of the visual categorization according to the measured PM fraction. Filters with noted
problems (i.e., stable post-weights could not be obtained, filter was dropped during post-sample
handling, large particles/organic matter were present on the filter immediately after sampling)
numbered 17, 5.7% of the 296 samples collected during the study. Most samples (66.9%) were
classified as very light or light, having just a few small  detectable particles. As TSP sampling
does not use an impactor assembly, a TSP filter would be expected to have larger and much more
numerous particles  that may more easily move to the ring of the filter.  Therefore, more TSP
filters would be expected to have a significant number of particles on the ring than the collocated
PMio and PM2.5 filters collected at locations exposed to heavy PM plumes. This expected result
was supported in the ring particle density categorization of the filters: the only filter classified as
extremely heavy was a TSP filter. TSP  filters dominated the very heavy category, most of which
had visible particles in the Petri dish in addition to the filter ring.
Table 39. Size fractionated results of the visual inspection of annular filter rings.
Size
TSP
PM10
PM25
Total
%of
Total
Ring Particle Density Category
None
0
0
0
0
0
Very
Light
2
45
35
82
27.7
Light
22
51
43
116
39.2
Medium
19
9
19
47
15.9
Heavy
8
2
8
18
6.1
Very
Heavy
4
0
1
5
1.7
Extremely
Heavy
1
0
0
1
0.3
Not
available
0
1
9
10
3.4
Filter
w/ notes
8
8
1
17
5.7
Sum
64
116
116
296

A troubling finding was that a significant percentage (24%) of PM2.5 filters was categorized as
medium or greater. Most of these occurred on downwind samples collected May 18-20, while the
collocated  PMio  samples are generally  in  the very  light and light categories. This finding
suggests  that contamination during non-sample times is likely not a significant cause of the
problems in the data for these days because all the filters used on a given day were treated the
same during filter handling and storage and it would be expected that filters of all size fractions
would be impacted equally. This disparity in the number of PM2.s and PMio filters in the medium
or greater  ring particle density categories  suggests that the dominant  cause of larger-than-
expected particles and PM2.5 concentrations higher than PMio occurs at the impactor assembly in
the sample head,  which was the only difference between the PM2.5 and PMio MiniVol samplers
used.  As  the PM2.s impactor assembly removes  more  particles  from  the airstream than  a
collocated PMio impactor, it  is likely that particle bounce would occur sooner and at a higher rate
in the PM2.5 impactor than in the PMio impactor under the same plume conditions.
Table 40 presents the ring particle density  category results by sample period,  with a graphical
presentation in Figure 45. In general, the light  category  has the  greatest  number of filters,
followed by the very light category. The greatest  frequency of filters classified as medium and
greater occurs in  samples  collected May 18-20, which corresponds to the greatest occurrence of
PM2.5 samples classified  as medium  or  greater.  One  conclusion drawn from this is that the
downwind  filter-based PM samplers  were likely  overloaded during those  sample periods,
suggesting  that the reported downwind  concentrations should not be used for emission rate
calculations. This removes 104  samples, leaving 175 samples for emission rate calculations and
six others  with noted  problems.  Table 41  shows the operation(s) examined in each sample
SDL/08-556
California 2008 Tillage Campaign
115

-------
period, whether or not the filter-based PM dataset for each period was suitable for calculating
emission rates, and, if it was not used, the reasons for excluding the entire dataset.
Table 40. Results of the visual inspection of filter rings by sample date.
Sample
Period
May 17 Rl
May 17 R2
May 18
May 19 Rl
May 19 R2
May 20 Rl
May 20 R2
June 5 Rl
June 5 R2
June 7
June 11
June 18
June 25
Sum
%of
Total
Ring Particle Density Category
Very
Light
6
3
3
6
7
7
5
0
0
4
10
12
19
82
27.7
Light
13
13
7
5
8
3
11
12
12
12
13
6
1
116
39.2
Medium
3
4
8
6
3
2
3
3
7
5
0
1
2
47
15.9
Heavy
0
1
1
1
1
5
1
5
1
0
0
1
1
18
6.1
Very
Heavy
0
0
1
1
1
2
0
0
0
0
0
0
0
5
1.7
Extremely
Heavy
0
0
0
0
1
0
0
0
0
0
0
0
0
1
0.3
Not
available
1
1
1
1
1
1
1
1
1
1
0
0
0
10
3.4
Filter w/
notes
0
1
2
3
1
3
2
2
2
1
0
0
0
17
5.7
Sum
23
23
23
23
23
23
23
23
23
23
23
20
23
296
100.0
For the remaining samples, the ring particle density classifications were used to  separate out
problematic PM  concentrations. In the PM2.5  and PMio  size fractions, all  filters that were
classified as medium or greater and those with noted problems were removed, totaling 12 (6.9%)
and 8 (4.6%) PM2.5 and PMio samples, respectively, for remaining sample periods. In addition,
two  PM2.5, one  PMio, and  two  TSP samples  were  removed  from  consideration  due to
exceptionally high  or low  concentrations,  often  with significantly  greater PM2.5  or PMio
concentrations than PMio or TSP, respectively. Two of the five were collocated with an OPC and
the OPC time series (based on 20-second sample times) in each case was examined and provided
no evidence of contamination/concentrations significantly above background during sampling.
Four of the five removed for this reason were collected at background locations, and the fifth
was at the  AQT location while the tillage  operation was occurring in Field 4, several hundred
meters to the north. The other three samples removed were background TSP, PMio, and PM2.5
samples for May  17 Run 2, which were not collocated with an OPC at the time. Unfortunately,
this did not leave  a background PMio or PM2.5 concentration to use  in calculating the operation's
contribution to downwind concentrations for this run as the PMio  and PM2.5 filters at the other
background location had ring particle density categorizations of medium and were also removed.
Therefore,  a TSP  emission rate will be calculated using inverse modeling for May 17 Run 2, but
not PMio and PM2.5 emission rates.
SDL/08-556
California 2008 Tillage Campaign
116

-------
  c
  I
                                            • Very Light

                                            • Light

                                            • Medium

                                            • Heavy

                                            • Very Heavy

                                            • Extremely Heavy

                                              Not available

                                            • Filters w/notes
                                                                      f-Very Light
                                                                       Medium
                                                                      Heavy
                                                                     Very Heavy
                                                                    Extremely Heavy
                                                                   Notavailable
                                                                   Filters w/notes
                                                   Jll
                                                        J1S
                                                             J25
Figure 45. Line graphs showing the number of filters in each ring particle density category for each sample
period.
Traffic on access roads adjacent to upwind/background locations was recorded by field personnel
during most sample periods. Therefore, possible contamination due to sources nearby the upwind
sample locations was closely investigated during each sample period. This was accomplished by
examination of the available OPC  time  series for significant spikes in particle  counts that
indicate such an event. The OPC time series for upwind location 10.0 on June 5 Run 1 (Rl) and
June 5 Run 2  (R2) showed  that nearby road traffic significantly impacted the PM samplers:
between  50%  and 80% of the TSP volume concentrations were attributed to such  events.  In
addition, the ring particle density categories of the six filters collected at this location on June 5
were: PM2.5    heavy and medium;  PMio    heavy and medium; and TSP = medium and  light.
Based on these categorizations, the PM2.5 and PMio values  had previously been removed for
being medium  or greater. In light of this, the two TSP concentrations at location 10.0 were also
removed from  emission rate calculations. Samples collected at the other background location,
Tl, were not compromised by road traffic.
SDL/08-556
California 2008 Tillage Campaign
117

-------
Table 41. Sample period filter datasets used in calculating emissions, with reasons for why some datasets were
not used in further calculations.
Sample
Period
Operation(s)
Used
If not, reason excluded
Conservation Tillage
May 17
Rl
June 7
June 11
Strip-Till
Plant and Fertilize
Herbicide application
Yes
Yes
Yes
—
—
___
Conventional Tillage
May 17
R2
May 18
May 19
Rl
May 19
R2
May 20
Rl
May 20
R2
June 5 Rl
June 5 R2
June 18
June 25
Break down in-field borders
Chisel
Disc 1
Disc 2
Lister
Build ditch and field-edge
borders
Break down ditch and field
edge borders, Cultivator
passes 1 and 2, and Roller
Plant
Fertilize
Cultivator pass 3
TSP - Yes
PMio, PM2.5 - No
No
No
No
No
No
Yes
TSP, PM2.5 - Yes
PMio - No
Yes
Yes
Problems with most
background samples
Downwind samplers
overloaded by plumes
Downwind samplers
overloaded by plumes
Downwind samplers
overloaded by plumes
Downwind samplers
overloaded by plumes
Windblown dust
contamination

Problems with both PMio
background samples
___
—
Results of these investigations into the inverted PM2.5-PMi0 concentrations and visible particles
led to the removal of many collected filter samples from emission rate calculations using inverse
modeling, including all downwind samples collected during five consecutive sampling periods.
A graphical representation of the filters used and reasons for not using others is presented in
Figure 46. Note that the four samples for which it was determined that the PM2 5 and PMio filters
had been switched were corrected (i.e.,  the concentration values were reassigned) and used in
emission rate calculations. The final PM concentrations used in emission rate calculations are
tabulated in Table 33 and Table 34. Most of the samples removed during this investigation were
not used in MCF calculations, as explained in Section 4.3.2.
There  are several  practices and  procedures  which may prevent such problems in future filter-
based sampler deployments in such conditions. Filter handling steps include, but are not limited
to, the following: use EPA-recommended filter storage, handling, and mass determination
SDL/08-556
California 2008 Tillage Campaign
118

-------
                                    Total = 296 samples
                              5, 2%
                                                          I Pass Inspection
                                                          I PM2.5-PM10 Switched
                                                           Note on runsheet
                                                          I Contaminated/Overloaded
                                                          I Ring density > medium
                                                          I Abnormally high/low
Figure 46. Categorization of filter samples to determine suitability for use in emission rate calculations in
inverse modeling.
methodologies; use a calibrated mass balance, with continuing calibration verification; carefully
track each filter throughout the pre-weighing, sampling, and post-weighing steps; and ensure that
the filter storage and handling areas, both in the lab and the field, are clean and will minimally
contribute to the particles present  on the  filter.  During the sample preparation and  sampling
periods, the following practices will help to prevent problems with filter samples: inspect each
sample assembly before and after use for potential problems, specifically checking the condition
of the layer of high-vacuum grease on the impactor plate; place sample locations at a sufficient
distance downwind from the source(s) such that heavy plumes will not  overload the sampler; if
possible, restrict access to upwind areas to prevent contamination of upwind samplers, but if it is
not possible place upwind samplers in positions that will not be impacted by nearby activities.
Additionally, record  in detail all activities  under study (e.g.,  tractor operation time,  tractor
position via GPS) and any activities in upwind areas that may potentially affect upwind and/or
downwind  samplers. Proper training  of personnel is  also critical, both in prevention and
identification of potential problems. Placing a more time-resolved PM sampler, such as the OPCs
employed herein,  with  filter-based samplers provides the opportunity  to examine  changes in
aerosol loadings to identify impacts from  source  activities, as well as  potential  contamination
during sampling. It should be noted that most of these procedures were  already in use when the
problems in this campaign occurred; future mitigation of such problems requires more  careful
evaluation and application.
As discussed in the conclusions, the fact that we had to dedicate  so many resources to extracting
any useable data from our MiniVol samplers tells us that some field situations are in fact "too
dusty" for the sampling techniques as we  applied them.  Furthermore, even  when we tried to
counter some of these effects (by using a PMio impactor as a pre-filter for the PM2.5 impactor)
we continued to run into difficulty.  By employing as many of the best-practices listed above we
expect to avoid many of these difficulties in the future.
SDL/08-556
California 2008 Tillage Campaign
119

-------
10.3   APPENDIX C: RESPONSES TO COMMENTS RECEIVED RE: CALIFORNIA
       SPRING 2008 TILLAGE CAMPAIGN: DATA ANALYSIS REPORT
Responses to comments are in italics with all changes to the report text indented. Note that all
page numbers mentioned refer to the page number in the final version.

10.3.1    Draft Date:  16 March 2011 (Organization: U.S. EPA)

10.3.1.1     General Comments
    1)     Since the goal of this project was to calculate emissions for the purposes of assessing
          control efficiencies  and any emission calculations were ultimately used to compare
          emissions from a  conventional practice with one  that reduces emissions, we believe
          that it is appropriate to use a combination of point  samplers and inverse modeling as a
          part  of the methodology used to calculate emissions. EPA would want to further
          scrutinize this methodology if the main goal of the study was to inform  baseline
          emission factors.
          No response required
    2)     The  dependence on the use of historical emissions factors and the comparison to
          emissions  factors  estimated  in  this  study  is  understandable.  In  addition, the
          discussions describing  the variations of measured  emissions  and the uncertainties
          associated with historical emissions factors is refreshing. However, some of the cited
          uncertainties  appear quite low compared to  some recent  assessments of emissions
          factors uncertainties of several sources which would be thought to be less variable
          than tilling emissions (see  http://www.epa.gov/ttn/chief/efpac/uncertainty.html ). It is
          suggested that the uncertainties associated with the historical emissions factors and
          the estimated emissions factors associated with this study be re-evaluated based upon
          the conclusions in the  draft emissions factor uncertainty report (i.e. the number of
          independent test runs and the underlying variability of the measured emissions). Then
          some of the processes which are described as not being in close agreement  may be
          more in line with the overall uncertainties associated with the emissions factors and
          variabilities.
          The authors agree that a  better analysis of tillage  emissions factors uncertainty is
          warranted and have done so to the extent data are available, following the emissions
         factor dataset analysis procedure used in the emissions factor variability assessment
          referenced above. Results of two additional studies that  have been released since
          submittal of this document to EPA have been included in  the literature review. The
         following paragraph describing the uncertainty analysis methodology and results has
          been added to the end of the literature review section (page 10). Refer to page 11 in
          the document to view the tables and figures.
              "An  uncertainty analysis  was  conducted to determine  the  statistics of the
             preceding PM10 emissions factors  reported from measurements. This  analysis
             was performed following the  emissions factor dataset analysis technique used by
             RTI International in 'Emission Factor  Uncertainty Assessment, Review Draft'
             [29].  Data points were categorized according to the following tillage operations,
             with the number of values given in parenthesis: chisel (2), disc (67), land planing
             (1), listing (8), ripping (5),  root cutting (3), standard tillage planting (11), strip-
SDL/08-556                      California 2008 Tillage Campaign                            120

-------
              till planting (9), strip-tilling (6), and weeding (15). In cases where a report/paper
              only provided an average emissions factor, the average value was used only once.
              RTI International found that two parametric models, the lognormal and Weibull
              distributions,  best fit the analyzed datasets. These same two parametric models
              were fitted to the tillage operation datasets with eight or more data points. The
              estimated parameters for these models and the corresponding goodness-of-fit to
              the available data points, expressed as the root mean square  error (RMSE), are
              given in Table  5 with the better fit model values in bold. Note that a smaller
              RMSE represents a better fit to the data. The Weibull distribution proved to be a
              better fit for four out of the five examined datasets.  The fits to  the disc and
              weeding datasets were better than that for the remaining three, both visually and
              based on the RMSE values. Figure 1 presents the histogram for the disc operation
              dataset and cumulative density functions developed from the data and the Weibull
              distribution fit to the data. The  mean and median  values  of the fitted Weibull
              distribution are shown on the cumulative density function line, along with  the
              emissions factor value given  by ARE for a disc operation. The ARE emissions
             factor of 134.5  mg/m2 is very close to the Weibull distribution median of 136.3
              mg/m2. The emissions factor values corresponding to the 5%, 25%, median, 75%,
              and 95% levels along the cumulative  distribution curve, as well as the average
              emissions factor, were  calculated for the five operations  and are presented in
              Table 6. The 95% level emissions factors for the three operations with poorer fits
              (i.e.,  higher RMSE)  seem  very high; this  is likely an artifact of fitting  the
              distributions to a limited number of datapoints. In  this analysis,  the better fits
              were obtained for datasets with n > 15. "
          The discussion of the results on page 86 of the document has been updated to reflect
          the findings of the  uncertainty analysis,  which did indeed include a portion of the
          higher values as suspected by the reviewer. The paragraph now reads:
              "Some  of the values herein reported are in agreement with those reported  by
              Flocchini et al.  (2001) and Madden et al. (2008), as well as the PM10 emission
             factors used by CARB, such as the strip till and plant passes in the conservation
              tillage  method  and the  cultivate  and roll passes  in   conventional  tillage
              [12][16][13]. Other  emission rates are significantly different and larger than
             previous values reported in the  literature, especially the discing 1 and 2, chisel,
              and lister passes of the conventional tillage, though those derived through inverse
              modeling are  below the  estimated 95% level for  their respective distributions as
             presented  in  the uncertainty calculations  in  this report.  The  lidar-derived
              emission rates for the disc 1, disc  2, chisel, and lister passes are  high when
              compared to  values found by inverse modeling coupled with OPC PM data,
              values in the literature for the same  operations, and values reported in the 2007
             fall CMP tillage study which  used the same lidar methodology. These relatively
              high  emission rates provide indirect support for the conclusion that downwind
              PM samplers were likely overloaded during those sample periods by high aerosol
              concentrations.  While the values from listed published studies are generally not in
              close agreement, they are relatively well fit by lognormal or Weibull distributions,
              which have previously been shown to represent emissions factors datasets well
              [31]. In addition, they  are within  the range of the variability expected from
SDL/08-556                      California 2008 Tillage Campaign                             121

-------
             measurements made  under different meteorological and  soil conditions, as
             demonstrated by  the wide  range of values from Flocchini et al.  (2001)  [12]
             summarized in  Table 2.  The results from  this campaign are not in as good
             agreement as previous results have been [7]. In general our results are larger
             than their corresponding literature values. "
          A sentence has also been added to the executive  summary  on page  2  about the
          comparative results with the uncertainty analysis:
             "Emissions factor values estimated through inverse modeling for these operations
             are below the 95% level predicted by statistical distributions fitted to published
             data; emissions estimates from lidar for the same operations are above the  95%
             level"
   3)     The use of reverse modeling for particulate matter is of concern since the deposition
          and removal mechanisms are limited in EPA models. The EPA models only have a
          percentage of the parti culate reflect off a smooth surface when in reality, the ground
          surface is not smooth and if there is vegetation more removal may occur.  In addition,
          electrostatic and other forces may cause small particulate to aggregate to form larger
          particulate.  These effects are not included in EPA models and as a result a reverse
          model would tend to underestimate the actual source strength. The amount of
          underestimation may depend upon not only these items but also the distance at which
          the measurement is removed from the emissions source.
          The authors agree with the limitations of the ISC andAERMOD models stated by the
          reviewer concerning  deposition  and  removal  mechanisms,   resulting  in  an
          underestimation of the actual source strength. The effect of the distance  at which the
          measurement is removed from the source was minimized in this study by locating the
          sampling sites immediately adjacent  to  the field under  study;  additionally,  the
          constant motion of the tractor and implement lead to varying distances between the
          measurement and source. It should also be noted that any emission estimation
          methodology for a tillage source that does not include a correction for the deposition
          and removal of PM due  to surface roughness,  vegetation,  etc.  as a junction of
          distance from the source will underestimate actual emissions. The following text has
          been added to the inverse modeling methodology section on page  38:
             " In addition, the deposition and particle removal mechanisms are limited in the
             ISCST3 and AERMOD models herein employed. Insufficient correction within a
             given model for these and other processes that decrease downwind pollutant
             concentrations will lead to  an underestimation of emission rates  based on
             measured downwind impacts."

10.3.1.2     Comments about Appendix B
   1)     While the selection and use of low-cost, battery-powered samplers is understandable
          in a study of this type (particularly requiring a large number of concurrently operated
          samplers), it  should first be recognized that the MiniVol samplers are  not EPA-
          approved reference method or equivalent method samplers for making either PM2.5
          or PM10  compliance measurements.    Although  the MiniVol  samplers have
          sometimes  been  shown to provide similar PM concentrations  to collocated FRM
SDL/08-556                      California 2008 Tillage Campaign                            122

-------
          samplers under  some  sampling conditions, their ability to extract representative
          samples of large particles under high wind speeds has not been demonstrated.  In fact,
          a limited wind tunnel evaluation of the MiniVol samplers (conducted  by Research
          Triangle Institute in 1991) indicated that the sampler's ability to obtain representative
          aerosol  samples degrades  rather dramatically  as a function of wind speed.   In
          agricultural environments where elevated wind speeds are the not at all uncommon,
          representative sampling of large aerosols characteristic of tilling operations cannot be
          assumed.  This is particularly the case when interpreting data from MiniVol samplers
          operated  as  "TSP" samplers.   For  this reason, data obtained  with the MiniVol
          samplers in this type of study should  only be considered as approximate measures of
          PM2.5, PM10, and "TSP" concentrations.
          We thank the reviewer for  bringing to our attention the collection efficiency versus
          wind speed issued found in the 1991 MiniVol study. This is an issue of concern,
          especially when dealing with aerosols with relatively large size distributions as found
          downwind of most agricultural sources. It should be noted that the sampler model of
          those tested in 1991 were < 3.x, while the model number of the units employed during
          this study were 4.2. Attempts to contact Airmetrics to inquire about potential changes
          to the sampling inlet/size separator that might affect sampling efficiency between the
          models have not yet been successful.
          Text was added to the Section 3.1.4.1,  page 27, which describes the  MiniVols, to
          clarify that the MiniVol is neither a FRMnor a FEM.
              "The MiniVol is a  battery operated, ambient air sampler that gives results that
             closely approximate air quality data collected by a Federal Reference Method
             (FRM) PM sampler. The MiniVol is neither designated as an FRM nor a Federal
             Equivalency  Method (FEM)  by the  EPA, and results should be considered as
             approximate measures ofPM."
          A  short  discussion of general  meteorological conditions observed during  the
          sampling periods and  a table listing period-average values has been included in
          section 4.1.2.2, page 43. The period-average wind speeds observed during this study
          varied between 2 and 6 m/s, within the range of 0.5 m/s and 6.7 m/s at which the 1991
          study was conducted.
          Section 4.1.2.2, page 43, reads:
              "Meteorological characteristics  were monitored on-site throughout the field
             study.  Sample period average conditions were calculated and are  presented in
             Table 14  based on measurements taken at the WM and EM locations. As can be
             seen from these data, all measurements were made  during warm and  dry
             conditions.  Winds were consistently out of the northwest with average speeds
             between 2 and 6 m/s. "
   2)     Appendix B provides an account of discrepancies noted during visual observations of
          collected  field  filters  and  offers  several   possible  explanations  for  these
          observations. While the discussion on page 108 correctly points out that  there may be
          differences in the position and shape of the MiniVol and FRM's fractionation curves,
          it is  not correct to  state  "Ideally, the collection  efficiency of the size  fractionation
SDL/08-556                      California 2008 Tillage Campaign                            123

-------
          device would be a step function that goes from 0% to 100% at the desired cut-off
          size."  In fact, the slope of the FRM's PM10 fractionation curve was intentionally
          designed to match the performance of the human respiratory system - which does not
          provide step-function performance. The fractionation curve of the PM2.5 FRM was
          also not designed to provide step-function performance.
          This error has been corrected in the text with the first and second sentences of this
          paragraph, page 113, now stating
              "The collection efficiency curve of the PM10 and PM2.5 size fractionation
             devices employed by both FRMs and MiniVols are  S-curves  with  respect to
             particle aerodynamic diameter, designed to mimic the particle removal efficiency
             of the  human respiratory system.  The  removal efficiency curves have a 50%
             collection efficiency at the designed cut-off size, with some  smaller particles being
             removed and some larger particles passing through the system. "
   3)     As hypothesized on page 109 of the report, observations of very large particles on
          collected field filters and of uncharacteristic deposition patterns are  almost certainly
          due to "particle bounce" occurring in the MiniVol samplers.   Unlike EPA's PM2.5
          FRM which uses either the WINS well or the  very sharp cut cyclone (VSCC) to
          minimize substrate overloading and particle bounce, the MiniVol samplers were not
          designed to handle high concentrations of ambient aerosols.  The design of the
          impaction  substrate for both the PM2.5 and PM10 MiniVols is  such that particle
          bounce may  occur from the impaction stage once a few monolayers of particle
          deposits occurs during sampling.  Once particle bounce begins, the cutpoint of the
          stage can  shift dramatically upwards  and  result in inaccurate PM  concentration
          measurements. In the case of the MiniVol design, the reduction in a cutpoint from 10
          micrometers to 2.5 micrometers requires a reduction in impaction jet diameter from
          approximately 0.69 cm to  approximately 0.29  cm.  As  a result,  the jet velocity
          through the PM2.5 MiniVol is approximately 5.5  times that of the  PM10 MiniVol.
          This increase in jet velocity results in an increase in particle kinetic energy and makes
          the  PM2.5  MiniVol more susceptible to  particle  bounce  than the PM10 MiniVol.
          And because the PM2.5 MiniVol has a much smaller stage deposition area than that
          of the PM10  MiniVol,  it is possible for the mass  gain of PM2.5 MiniVol's filter to
          exceed that of the PM10 MiniVol filter. As  a result, measured PM2.5 concentrations
          may often exceed those of collocated PM10 concentration measurements.  Although
          occasional inadvertent switching  of PM2.5 and PM10 filters was noted in the report,
          substrate overloading was almost certainly the cause of the majority of the "inverted"
          PM2.5 and PM10 measurements.
          The authors thank the reviewer for providing a very good analytical description of
          the particle bounce phenomenon as it may occur in the MiniVol impactor assembly.
          The discussion under point 3 on page 113 in the report has been changed to include
          portions of this description. It now reads:
              "A phenomenon commonly referred to as "particle bounce " may have occurred.
             Particle bounce refers to when a particle collides with the  impactor assembly but
             then returns to the air stream  and is collected downstream at the filter.  To aid in
             removal of particles colliding with the impactor plate, grease may be applied on
SDL/08-556                      California 2008 Tillage Campaign                             124

-------
             the impaction surface. If too many particles accumulate in the grease, however,
             the impaction surface loses its  "stickiness" and larger particles may bound off,
             leading to higher reported PM2.s/PMjo concentrations than actually existed.
             Impactor plate stickiness is influenced by a combination of total exposure  time
             and PM concentrations during exposure. The smaller effective impact area and
             increased particle velocity in the MiniVol PM2.5 impactor assembly makes it more
             susceptible to particle bounce than the PMjo assembly under the same conditions,
             which could result in higher reported levels of PM2.5 than PMjo- Airmetrics,  Inc.
             suggests cleaning and  regreasing of the impactor plates  every five to seven
             samples to maintain the design removal efficiency, but states that  the need to
             renew the plate may change based on exposure levels. In the  case of this study,
             silicone-based, high-vacuum grease was applied to all impaction surfaces on-site
             prior to sampling in May. Personnel noted that impactor assemblies in downwind
             samplers had collected significant amounts of particles near the end  of the seven
             sample periods in that month (May 17-20), which suggests  that particle bounce
             likely occurred during  some sample periods. The sample heads and impactor
             assemblies were cleaned and greased during the break in  measurements,  with
             evaluation and re-greasing if necessary after  each measurement  in June. In
             addition, the enhanced susceptibility of the PM2.5 assembly to particle bounce
             could explain the inverted PM2.5 and PMw concentrations observed at downwind
             locations during the May 19 and May 20 sample periods. "
   4)     If funding were not a limitation, a much better alternative to the use of the MiniVols
          would be BGI PQ100 samplers.  These 16.7 Lpm, filter-based samplers  are EPA-
          designated for both PM2.5 and PM10 compliance testing, and operate with 120 VAC
          line power or internal  12 VDC battery (with solar panel  option). The PM2.5 version
          of the BGI PM2.5 PQ100  can be  equipped with a 2.5 micrometer cutpoint VSCC
          which has been shown to maintain its fractionation  performance under high loading
          conditions.  Use  of PQ100 samplers should  totally prevent the occurrence of PM2.5
          and PM10 data inversion that was reported in the study.  Even if a suite of these  BGI
          samplers could not be procured, it would be useful to collocate a pair of them with the
          currently operating MiniVol samplers in  order to  help  isolate and  resolve the
          performance issues currently noted with the MiniVol  samplers.  Funding required to
          purchase a pair  of the PQ100 samplers would more than be compensated for the
          resources currently being  expended  attempting  to  extract and interpret  useful
          information from field data of uncertain quality.
          The BGI PQ100 units have recently been investigated as potential future purchases.
          The  reviewer's  description and suggestion for  comparison  with  the  MiniVols
          currently deployed are compelling and a future purchase of some PQlOOs is planned.
          This comment highlights our ongoing effort to  utilize instruments  of higher fidelity,
          such as the BGI PQ100.
   5)     If continued use of the MiniVol samplers is envisioned, it is absolutely  critical that
          the MiniVol's impaction surface  be renewed  after every sampling event. In the
          MiniVol  design,  this is not  a difficult  or time  consuming process.  While  this
          procedure will not totally  prevent substrate overloading during  sampling of high
          concentration aerosols, it will at least minimize the effect and  should substantially
SDL/08-556                      California 2008 Tillage Campaign                             125

-------
          reduce the frequency of the PM2.5  and PM10  concentration "inversion" thats
          currently being observed.
          It is anticipated that the MiniVols will continue to  be used to measure ambient and
          near-source PM. The reviewer's suggestion to renew the impaction surface between
          each use will be implemented in future  uses.  This has also been included as a
          potential preventative action in the Lessons Learned section, page 90, under  the
          secondpoint:
                 "Another preventative action is to  clean and regrease the MiniVol impactor
                assemblies after each use if exposed to high aerosol concentrations. "

10.4   DRAFT DATE: 15 APRIL 2013 (ORGANIZATION: SAN JOAQUIN VALLEY
       AIR POLLUTION CONTROL DISTRICT)
   1)     Page 45  Section 4.2.1 discussion of 165 of 296 samples rejected by QA requires some
          immediate discussion  as well as reference  to Appendix B.  This amount of QA
          rejection is very  high and will raise an immediate red flag.  The importance of this
          issue merits added text in the paragraph.
          A summary of likely causes of the high failure rate has been included. The section on
          page 47 now reads as follows:
             "Of the 296 filter samples collected,  165 did not pass quality analysis/quality
             control (QA/QC) steps applied to the dataset, leaving 131 for use in calculating
             emission rates using inverse modeling. An  investigation into  this high rate of
             failure was conducted and a detailed description is provided in Appendix B. In
             summary, filters  that  did not pass  QA/QC  were  suspected to have  been
             contaminated either during sampling or during storage and handling. Evidence of
             "particle bounce " was found on many PM2.5 and PMio samples collected during
             May sample periods. Particle bounce occurs when particles that collide with the
             impactor plate, the mechanism used by the MiniVols to exclude particles larger
             than the design size, are re-entrained in the air stream and collected on the filter
             downstream and result in higher measured levels than actually existed. This issue
             is most likely due to exposing the MiniVol samplers to dust plumes exceeding the
             maximum recommended exposure level and improper instrument maintenance
             and  cleaning through  the May  sample periods. Corrective action was taken
             during the June sample periods and no issues associated with particle bounce
             were observed in the second portion of the study.
             "Additionally, some particles were observed on top of and imbedded into  the
             plastic annular ring around the Teflon filter material - the plastic ring is covered
             by the filter  holder  assembly  during deployment.  This  was likely  due to
             contamination during on-site filter storage  or  handling. Efforts were  made to
             minimize  this issue  throughout, especially during the  June sample  periods.
             However,  windblown dust did impact the handling and storage area during the
             last  sample periods in May. In addition,  contamination of upwind samplers
             prevented emission rate calculations for some size fractions for  two other
             operations. The size fraction distribution of approved filters was nearly identical
             to the total sample set: 51 (39%) were PM2.5, 50 (38%) were PM10, and 30 (23%)
SDL/08-556                     California 2008 Tillage Campaign                            126

-------
             were TSP. These finalized concentrations are given in Table 33 and Table 34 in
             Appendix A. See Appendix B for a detailed discussion of the QA/QC steps, filter
             inspection failure, possible causes of the failures, and preventative solutions for
             future sampling."
   2)     In sharp contrast to  section 4.2.1, sections 4.2.3 through 4.3.3 (starting at page 49)
          never mention QA.  Some discussion of data completeness and redactions should be
          included for these sections.
          Sentences and paragraphs addressing QA have been included in each of the sections
          referred to above. Some of the material were previously in the methodology portion of
          the report and have been pulled into the results section. While the lidar concentration
          measurements section had some discussion of QA screening, it has been augmented,
          clarified, and moved to the first paragraph of the section.
          Page 48, 4.2.2.1 Organic Carbon/Elemental Carbon Analyzer
              "As mentioned previously, the organic carbon/elemental carbon analyzer passed
             the manufacturer's  suggested in-field audits  and after completion of the field
             project the data were manually screened for completeness and potential outliers.
             During the two distinct periods of sampling, May 13-20 and June 3-21, 2008, the
             EC/OC  instrument  operated continually except for brief periods for QA/QC
             checks, servicing of the system generator,  or  significant breaks in the producer
             operations.  An unanticipated consequence  of the one hour sample times, coupled
             with the dual channel operation of the R P 5400 EC/OC Analyzer, was that the
             sampling/analysis/cleaning cycles extended beyond the planned two hours.  The
             net result was a sampling profile wherein  every third hour of data was missing
             (66% sample collection efficiency over the entire deployment). However, because
             the actual farm  practice periods varied from one  to  eight hours, the actual,
             observed data periods ranged from one to eight hours, the data coverage was
             from 50-100%, averaging 78.2% ± 10.3%. "
          Page 50, 4.2.2.2 Ion Chromatographic (1C) Analysis
              "As such, a  total of nine filters were analyzed for soluble ions. However, two of
             these filters,  from 5/20/08 and 6/5/08, were subsequently suspected as potentially
             having contamination  issues (see Section 4.2.1) and were discarded from further
             analysis."
          Page 54, 4.2.3 Aerosol Mass Spectrometer
              "During the tillage experiment,  the AMS  acquired chemical composition  data
             from May 14-May  19  with  some significant gaps in the data due to mass
             spectrometer malfunctioning.  This required manual screening of the AMS data.
             Approximately 35% of data over the time  period was found to be valid. Large
             gaps occurred throughout the  sampling  period with, for  example, no  data
             acquired on  May 17. "
          Page 56, 4.3.1 Met One Optical Particle Counter
              "The collected OPC data were  analyzed for  particle size distribution, particle
             volume  concentrations,  and converted to  particle mass concentration through
SDL/08-556                      California 2008 Tillage Campaign                            127

-------
             multiplication with the MCF, as described in Section 3.1.4.2.  Table 35 and Table
             36 present the PM2.5, PMw, and TSP concentrations as reported by the OPCs.
             Three to four OPCs were in positions immediately downwind of the field under
             study in each sample period, with between  one and four OPCs  in upwind
             locations.  Unlike the downwindMiniVol samplers, the downwind OPCs were not
             overwhelmed by the dust plumes from the tillage activities  - the manufacturer
             specified range of the OPC ofO to 318,000,000 particles/m3 was never exceeded-
             and thus provided usable data throughout all sample periods. Upwind OPC time
             series data were examined for contamination from activities upwind of the site,
             such as unpaved road traffic.  Contamination was found in six of the 12 sample
             periods, with five of those occurring at the 10.0 sample site that was immediately
             downwind of an unpaved road (see  Figure  7). Large spikes indicative  of
             contamination were removed from  the  upwind OPC time series data in  these
             instances  to  estimate the background aerosol  concentration; the  estimated
             background levels were in very good agreement with those measured by an OPC
             at a different, uncontaminated upwind location (see Table 37 and Table 38).
             Data completeness for the OPC datasets was calculated as a ratio of the number
             of valid samples per sample period over the possible number of valid samples and
             expressed as a percentage.  Data  completeness  was less  than 100%  due  to
             communication  errors between the  OPCs and the computer logging the  data,
             resulting in  lost packets  of 20 second  sample  data.  Communication  error
             frequency was variable between OPCs and across time. Data completeness per
             sample period ranged from 81.8% to 100.0%, and averaged (± la) 97.4 ± 3.7%. "
          Page 61, 4.3.2 Optical to PMmass concentration conversion
             "As  OPC data collected at each site  and during each sample period passed
             QA/QC, the calculation of MCF values was dependent solely  on the presence of a
             valid filter-based PM measurement. "
          Page 66, 4.3.3 Lidar Aerosol Concentration Measurements
             "Lidar  data was  collected  throughout all sample  periods,  except for the
             Cultivator 3 pass monitored on June  25 due to an equipment failure after the
             previous measurement. All lidar scans  collected during times when  no tillage
             activity was occurring, based on detailed field notes, were removed from further
             calculations. The remaining scans were visually checked to remove  scans  with
             data acquisition errors and to prevent the use of data that was contaminated by
             other sources;  sources of observed contamination were vehicular  traffic on
             unpaved roads,  agricultural activities immediately upwind, activities associated
             with the adjacent dairy, and windblown dust. Contaminated upwind scans, as well
             as the corresponding downwind scans, were removed from further emission rate
             calculations. In most cases two or more downwind scans use the same upwind
             scan as a reference because multiple downwind scans were made for each upwind
             scan. Downwind scans were not used in further calculations if the corresponding
             wind direction was outside of ± 70° from perpendicular to  the downwind lidar
             beam path; if the lidar scan contained apparent plumes from an outside source
             (such as from unpaved road traffic or the dairy); if no plumes were detected; or if
SDL/08-556                     California 2008 Tillage Campaign                            128

-------
             the tillage plume had a potentially significant portion crossing the lidar beam
             closer than 500 meters down beam. In addition, upwind scans were invalidated if
             none of the associated downwind scans were usable. Light wind speeds (< 1 m/s)
             were  recorded during portions  of some  measurement periods.  Light winds
             challenge the measurement system and resulted in additional invalidation of some
             lidar  scans.  The total number of lidar scans made  along  both  upwind  and
             downwind planes is presented in Table 19, along with the number of valid scans,
             i.e. those that passed all quality checks, and the percent of valid scans compared
             to the total number of scans.
             In several cases,  the percent of valid scans was  very low. Although wind and
             background conditions were good during the herbicide application on 6/11, none
             of the scans taken were deemed valid because the mass difference between the
             upwind and downwind scans was below the minimum detection level (MDL). This
             means that the operation didn't produce plumes significant enough to be detected
             by the lidar; this operation was performed by a very small tractor pulling a small
             spraying apparatus - the only disturbance of the ground was due to moving tires.
             Plumes  of insufficient concentration differences from background  levels led to
             downwind scan invalidation in many  other instances also. The dairy pen areas
             adjacent to  Field 4 proved to be sufficient PM sources such that lidar scans
             showing dust plumes passing over the pens prior to crossing the scanning plane
             were nearly  always invalidated. Additionally, windblown dust was entrained off of
             both Field 4 and upwind field surfaces during both the May 20 sample periods.
             All of these factors combined to significantly decrease the number of valid lidar
             scans available for emissions estimation from most operations. "
   3)     Strip till is mentioned beginning at page 1, but never clearly defined. Page  13 comes
          close but there is no adjacent description for comparison to conventional tilling.  The
          two terms, as observed for the  study, should be clearly defined.
          A sentence has  been added to Section 2.2, Operation Description, to describe  a
          conservation  tillage CMP and some text has also been added further down in  that
          paragraph to contrast the level of soil disturbance between the strip-till and discing
          and plow ing passes of the conventional method. It now reads as follows:
              "As described in the Conservation  Management Practices Program  Report
             (2006), the conservation tillage CMP "involves using a system in which the soil is
             being tilled  or cultivated to a lesser extent compared to a conventional system "
             and it is "intended to reduce primary soil disturbance operation such as plowing,
             discing, ripping, and chiseling". The Conservation  Tillage CMP under study is a
             strip-till method which combines multiple operations to reduce  the number of
             passes required and disturbs the soil  only in 8 "-wide strips centered every  30 "
             instead of disturbing the  entire surface like the  plowing,  discing, and listing
             operations of a conventional method. "
   4)     One question is why the total  |ig/(m2 x  s) PM emission rates were used to derive the
          percent emission  reductions when  comparing  the  conventional  tillage   and
          conservation tillage methods rather than comparing the total mg/m2  emission rates.
SDL/08-556                      California 2008 Tillage Campaign                             129

-------
          Please provide some clarification on why this metric was selected to compare PM
          emissions from the practices.
          PM emission factors for  tillage are generally  given in mass  per  area for  each
          operation (e.g. Ib-PMlO/acre-pass,  mg-PM10/m2-pass, etc.) and this was also the
          basis for comparing PM emission rates from this study with  emission rates found in
          literature and used by ARB (see Page 4, Table 1; Page 6, Tables 2 and  3; Page 7,
          Table 4  and following  paragraph;  and Page 87, Table 28).  It would  seem that
          comparing  the  total mg/m   PM  emission  rates  for  conventional  tillage and
          conservation tillage  would provide a more useful and accurate representation of the
          change in PM emissions.  At a minimum, the percent reductions based on the total
          mass of PM emitted per area tilled  should  also be  shown  and included in the
          conclusions. This would not really change the conclusions of the report  but would
          probably be a more accurate representation of the total PM emission reductions.
          The authors agree that  this comparison should be made using  an emission factor
          rather  than an emission  rate. The  values and corresponding control efficiency
          calculations in Table 29 have been changed to emissions in mg/m2 format.  Text in the
          conclusion, page 86, has been changed to the following and numbers in the executive
          summary have been updated:
              "Emission rates for PM2.5, PMjo, and TSP from  both  lidar data and  inverse
             modeling coupled with  OPC data by operation are presented in Table  29 in units
             of mass emitted per unit area tilled per operation pass.... The conservation tillage
             method produced 9.5% as much  PM2.5, 6.3% as much PMjo,  and 9.1% as much
             TSP as  the conventional method according to the lidar data and 14.7% as much
             PM2.5, 12.8% as much PMw, and 9.7% as much TSP as the conventional method
             according to the AERMOD-OPC combination. Therefore,  the control efficiency of
             the CMP for particulate emissions from these three data sets was as follows: lidar
             - 0.91,  0.94,  and 0.91 for  PM2.5, PMW, and TSP,  respectively; and AERMOD-
             OPC-0.85, 0.87, and 0.90 for PM2.5, PM10, and TSP, respectively. "

10.4.1.1      Comment for Further Discussion
   1)     Interagency discussion should review the difference between entrained and emitted.
          Measuring too close to a source will observe entrained material that will not remain in
          the air;  it  is entrained  but will not be observed as  an atmospheric  contaminant
          downwind.  We need to be  sure that we have reasonable definitions for quantification
          of emissions. This  impacts review of the best  modeling  approach to  determine
          emissions - do we mean  everything entrained or only that portion of material that will
          mix in the  atmosphere sufficiently to be observable downwind and impact public
          health.
          EPA and SJVAPCD agree  that  additional discussions need to occur about  definitions
          and methodologies.
SDL/08-556                      California 2008 Tillage Campaign                            130

-------
10.4.2    Presentation of Results and Discussion with the San Joaquin Valley Air
          Pollution Control District's Ag Technical Group: 22 April 2013

10.4.2.1      Notes
USDA and the Space Dynamics Lab (SDL) did a presentation on the study results, emphasizing
that the conservation tillage practice that was assessed reduced emissions by over 85% . USDA
also  noted that while different conservation management practices (CMPs)  have differed
reductions, the method with the greatest reduction in passes and soil disturbance of those studied
had the greatest reduction in emissions. EPA stated that it would use the study  results to help
create  incentives  to implement  conservation tillage, and that  the Agency understands that
variable conditions on farms restrict the applicability of this measure. Stakeholders and USDA
also noted that producers needed flexibility to address different crops and on-farm conditions,
and that the  measure that was assessed in this research is not appropriate for every  farm.
Stakeholders also pointed out that in addition to reducing emissions through reductions in passes,
California agriculture has reduced  emissions by  switching to equipment with cleaner burning
diesel engines. The discussion  included the questions  below from stakeholders, which were
answered by USDA and SDL.
       The following sentence was included in Section 2.1 Site Description, page  12, to state the
      point brought out by stakeholders and USDA  that tillage operations are not used in all
       agricultural crop production systems: "Tillage management practices are  often  crop
       specific and are not  appropriate for  use  in all  crop production activities.  The
       effectiveness of CMPs used in other crop systems at reducing PM emissions  should be
       investigated."

10.4.2.2      Questions and Answers
Additional comments and portions of the report related to questions are provided after the answer
in the indented sentences
Q: Do the results effectively capture the emission reductions associated with less fuel usage?
A: We were able to calculate this on the previous study because the producer quantified the
amount of fuel going into his tractor. We anticipate that it's tied to the amount of time using fuel,
and would guess that there's probably a close to 85% reduction in fuel emissions also.
       The previous study  referenced  in  this  answer  is Williams et  al.  (2012; available:
       http://cjpub.epa.gov/si/sijfublic record' report.cfm?dirEntryId=248752)  in which  the
       control effectiveness  of a Combined Operations CMP was  investigated  by USDA and
       SDL in Fall 2007.
       The following sentences have been added to the conclusion section of the report on page
       89 to address the additional reductions  through reduced tractor  operation time:" It
       should be noted that other reductions in emissions between the two tillage management
      practices are likely to have occurred, mainly due to decreased fuel usage  in tractor
       engines. Unlike  the previously conducted companion study, fuel usage was not quantified
       in  this study and prevented accounting for the  associated reductions in emissions.
       However,  these  reductions are  expected to be  similar to the reduction  in tractor
       operation time (-85%)."
Q: Is it possible that there are some extra reductions to be claimed?
SDL/08-556                      California 2008 Tillage Campaign                            131

-------
A:  Yes, if we looked at carbon emissions as well as potential PM coming from fuel usage we
could get additional reductions— we didn 't capture all of that with the samplers we had in the
field for this study.
       See the response to the above question.
Q: Did you analyze yield differences?
A:  We were primarily interested in emission aspects and didn't do any yield comparisons in this
study.
       The following sentence has been added on page  12 to clearly state this in the report:
        "The focus of this study was comparing emissions resulting from the tillage operations in
       each field. Therefore, no data were collected to make comparisons of other potentially
       varying characteristics,  such as crop yield, soil organic matter,  etc., between the two
       management practices."
Q: Were either of the fields already in conservation tillage?
A:  The field that was under conservation tillage had been under conservation tillage for several
years; the other one was still being operated with conventional tillage with all the passes.
       This is addressed on page 13 of the report
Q: Did you look at the difference between soil temperature or ambient air temperature?
A:  We did not make soil surface temperature measurements. We agree that this factor may have
a significant effect on emissions transport. We plan to make such measurements in future studies.
        An additional paragraph has been added to the Lessons Learned section,  pages 90-91, to
       address this comment in the report: "In addition  to the above  lessons learned, peer
       review of this report also  brought  to our  attention that  measurements  of additional
       parameters should be considered in future  studies. Specifically, it was suggested that soil
       surface temperature measurements be made. Differences in soil surface temperature may
       result in  different vertical and horizontal PM dispersion (with all other conditions being
       equal) as surface conditions are known to affect turbulence and dispersion. "
Q: Were different tractors used for conventional vs. conservation tillage?
A: It was usually the same tractor but for some of the passes a different tractor may have been
used.
       The equipment,  including the tractor, used for each tillage operation are provided in
       Table 8, page 17
Q: Did you do anything with N20?
A: No. We didn't have the equipment at the time of the study.
       All equipment deployed for this study is provided in Section 3, pages 20-40.
 SDL/08-556                      California 2008 Tillage Campaign                            132

-------

-------
&EPA
      United States
      Environmental Protection
      Agency

      Office of Research
      and Development (8101R)
      Washington, DC 20460

      Official Business
      Penalty for Private Use
      $300

      EPA/600/R-14/191
      August 2014
      www.epa.gov
Please make all necessary changes on the below label, detach
or copy and return to the address in the upper left hand corner.

If you do not wish to receive these reports CHECK HERE n!
detach, or copy this cover, and return to the address in the
upper left hand corner.
PRESORTED STANDARD
 POSTAGE & FEES PAID
 EPA PER MIT No, G-35
                                              Recy cl ed/Recy cl able
                                              Printed with vegetable-based ink on
                                              paper that contains a minimum of
                                              50% post-consumer fiber content
                                              processed chlorine free

-------