An Integrated Study of the Emissions of Ammonia, Odor and ... · Rohit Mathur, MCNC, Research Triangle Park Rich Gannon, NC Department of Environment and Natural Resources, Raleigh

An Integrated Study of the Emissions of Ammonia, Odor and Odorants, and Pathogens and Related Contaminants from Potential Environmentally Superior Technologies (ESTs) for Swine Facilities

(Program OPEN: Odor, Pathogens, and Emissions of Nitrogen)

Evaluation Findings for the ESTs at B.R. Harris (AgriClean), and Red Hill (Environmental Technologies);and Lake Wheeler Road

Laboratory (Black Soldier Fly): for Ammonia, Odor and Odorants, and Pathogens

North Carolina State University, Raleigh, NC Duke University, Durham, NC University of North Carolina, Chapel Hill, NC

December 6, 2005

i

1. Project Title: An Integrated Study of the Emissions of Ammonia, Odor and Odorants, and Pathogens and Related Contaminants from Potential Environmentally Superior Technologies for Swine Facilities (Program OPEN: Odor, Pathogens, and Emissions of Nitrogen)

2. Investigator:

Principal Investigator and Program Scientist: Viney P. Aneja Professor, Air Quality Professor, Environmental Technology Department of Marine, Earth and Atmospheric Sciences North Carolina State University Raleigh, NC 27695-8208 (919) 515-7808 (Voice) (919) 515-7802 (Fax) [email protected] Co-Principle Investigators: Susan Schiffman Mark D. Sobsey Professor Professor Department of psychiatry Department of Environmental Science

Duke University Medical School and Engineering Durham, NC 27710 University of North Carolina Chapel Hill, NC 27599-7400 Co-Investigators: S. Pal Arya, North Carolina State University, Raleigh Ian Rumsey, North Carolina State University, Raleigh Deug-Soo Kim, North Carolina State University, Raleigh Wayne Robarge, North Carolina State University, Raleigh David Dickey, North Carolina State University, Raleigh Len Stefanski, North Carolina State University, Raleigh Philip W. Westerman, North Carolina State University, Raleigh Mike Williams, North Carolina State University, Raleigh Otto D. Simmons, University of North Carolina, Chapel Hill Lori Todd, University of North Carolina, Chapel Hill Rohit Mathur, MCNC, Research Triangle Park Rich Gannon, NC Department of Environment and Natural Resources, Raleigh Hoke Kimball, NC Department of Environment and Natural Resources, Raleigh

ii

mailto:[email protected]

TABLE OF CONTENTS

Project Summary Acknowledgements Introduction

I. Evaluation of Potential Environmentally Superior Technology for Ammonia II. Appendix A (Addendum to the Black Soldier Fly Report)

iii

Project Summary

The need for developing sustainable solutions for managing the animal waste problem is vital for

shaping the future of North Carolina. As part of that process, the North Carolina Attorney

General has concluded that the public interest will be served by the development,

implementation, and evaluation of environmentally superior swine waste management

technologies appropriate to each category of hog farms in North Carolina. This is being done

through agreements (Agreements) between the Attorney General of North Carolina and

Smithfield Foods, Inc and Premium Standard Farm, Inc, providing funds to the Animal and

Poultry Waste Management Center (A&PWMC) at North Carolina State University, Raleigh,

North Carolina.

During the past four years, project OPEN (Odor, Pathogens, and Emissions of Nitrogen) funded

by A&PWMC, has demonstrated the effectiveness of a new paradigm for policy-relevant

environmental research in North Carolina’s animal waste management. This new paradigm is

based on a commitment to improve scientific understanding associated with all aspects of

environmental issues (air, water, soil, odor and odorants, and disease-transmitting vector and

airborne pathogens) and, as part of a comprehensive strategy, to facilitate in the development,

testing and evaluation of potential Environmentally Superior Technologies for the management

of swine waste.

The progress that the OPEN team has made is a result of the scientific and intellectual leadership

provided by the collaboration of scientists and engineers from three (3) universities (North

Carolina State University, University of North Carolina at Chapel Hill, and Duke University),

one (1) national laboratory (National Exposure Research Laboratory, U.S. Environmental

Protection Agency), one (1) State of North Carolina Department (Division of Air Quality, and

iv

Division of Water Quality, NC Department of Environment and Natural Resources), and one (1)

private research organization (MCNC- North Carolina Supercomputing Center). 11ESTs have

already been evaluated. Five were evaluated in the Phase 1 report, these are Brown’s of Carolina

(BOC) Farm #93 – Upflow biofilteration system (EKOKAN); Corbett #1, 3 & 4 – Solids

separation/gasification for energy and ash recovery centralized system (BEST); Goshen Ridge

Farm- Solids separation/nitrification-denitrification/soluble phosphorus removal/solids

processing system (Super Soils); Hickory Grove- Orbit High Solids Aerobic Digester

(Orbit/HSAD); Lake Wheeler-Belt system

In the Phase 2 report, six ESTs were evaluated, these are AHA Hunt (SBR), Carrolls (ISSUES-

ABS), Harrells (ISSUES-PCS), Hickory Grove (Super Soils Composting), Howard Farms

(Constructed Wetlands), and Vestal (ISSUES-RENEW).

For this current phase 3 report, two ESTs are evaluated, these are B.R. Harris (AgriClean), and

Red Hill (Environmental Technologies). B. R. Harris was evaluated for one cold season. Red

Hill was evaluated for two seasons (cold and warm). Both technologies results were compared

and contrasted with current lagoon and spray technologies at conventional swine farms (i.e.

Moore Farm and Stokes Farm). Additional data is also provided for Black Soldier Fly, which is

located at Lake Wheeler Road Field Laboratory (see Appendix A)

This report will show that targeted emissions were reduced under some of the environmental

conditions studied for the candidate technologies. However, based on the current research results

and analysis, and available information in the scientific literature, some of the evaluated

alternative technologies may require additional technical modifications to be qualified as

Environmentally Superior as defined by the NC Attorney General Agreements.

v

Acknowledgements

This research is funded by the Animal and Poultry Waste Management Center (A&PWMC), Raleigh, NC. We sincerely acknowledge the help and support provided by Dr. Mike Williams, Project Officer, and Ms. Lynn Worley-Davis. We thank the technology PIs, farm owners, Cavanaugh & Associates, and Mr. Bundy Lane, C. Stokes, and P. Moore for their cooperation. We acknowledge the discussions and gracious help provided by Dr. John Fountain, Dr. Richard Patty, Dr. Ray Fornes, and Dr. Johnny Wynne of North Carolina State University. We thank Wes Stephens, Mark Barnes, Guillermo Rameriz, and Rachael Huie. We also thank Mr. Hoke Kimball, Mr. Mark Yurka, and Mr. Wade Daniels all of NC Division of Air Quality for their support. Financial Support does not constitute an endorsement by the A&PWMC of the views expressed in the report, nor does mention of trade names of commercial or noncommercial products constitute endorsement or recommendation for use.

vi

Introduction This project is part of an overall research, development and demonstration effort to identify environmentally superior technologies for the treatment and management of swine waste. The project is being conducted for Smithfield Foods, Inc., Premium Standards Foods Inc. and the Attorney General of the State of North Carolina through agreements between these entities known as the “Smithfield Agreement” and the “Premium Standard Foods Agreement” (Agreements).

The agreements define “Environmentally Superior Technology or Technologies” as any technology, or combination of technologies that (1) is permittable by the appropriate governmental authority; (2) is determined to be technically, operationally, and economically feasible for an identified category or categories of farms [to be described in a technology determination]; and (3) meets the following performance standards:

1. Eliminates the discharge of animal waste to surface waters and groundwater through direct discharge, seepage, or runoff; 2. Substantially eliminates atmospheric emission of ammonia; 3. Substantially eliminates the emission of odor that is detectable beyond the boundaries of the parcel or tract of land on which the swine farm is located; 4. Substantially eliminates the release of disease-transmitting vectors and airborne pathogens; and 5. Substantially eliminates nutrient and heavy metal contamination of soil and groundwater.

Evaluation Summary The results of these findings are summarized in three evaluation tables for the AgriClean technology:

Table 1. Water holding structures emissions Table 2. Barn (Fan ventilated or naturally ventilated) emissions Table 3. Total emissions.

The results of these findings are summarized in four evaluation tables for the Environmental Technologies technology:

Table 1. Tank emissions Table 2. Water holding structures emissions Table 3. Barn (Fan ventilated or naturally ventilated) emissions Table 4. Total emissions.

vii

I. Evaluation of Environmentally Superior Technologies for Ammonia

Project OPEN Science Team for Ammonia:

- Project Director Viney P. Aneja1 - Science Team Members S. Pal Arya1; I. Rumsey1; Deug-Soo Kim1; Wayne Robarge2; David Dickey3; Len Stefanski3; Lori Todd4; K. Mottus4; * K. Bajwa1, H. Semunegus1, S.Goetz1, W. Stephens1, Chiping Nieh4

1. Dept. of Marine, Earth and Atmospheric Sciences, North Carolina State University 2. Dept. of Soil Sciences, North Carolina State University 3. Dept. of Statistics, North Carolina State University 4. Dept. of Environmental Science and Engineering, University of North Carolina-Chapel

Hill * Graduate Students

viii

1. Evaluation of Environmentally Superior Technologies for Ammonia Emissions: B.R. Harris Farm

AgriClean

Alternative Technology: AgriClean Location: B.R. Harris Farm (Greenville, NC) Period of Operation: The OPEN team monitored for evaluation during: Field experiment: 01/17 – 01/28/2005 Technology contact: Phil Lusk (605-224-4334) NCSU Representative PI: Leonard Bull / Lynn Worley-Davis Statement of Task:

- Measurement of ammonia (NH3) emissions from anaerobic lagoon and storage tank by using a dynamic flow-through chamber technology.

- Analysis of water samples from waste storage and treatment areas for Total Ammoniacal Nitrogen (TAN) and Total Kjeldahl Nitrogen (TKN) concentrations (one sample each day during the experimental period)

- On site monitoring of meteorological parameters at 10 meter height - FTIR technology used to determine ammonia emissions from barns - Parameters measured: NH3 flux , storage lagoon temperature and pH, wind speed and

direction, solar radiation, and air temperature Description of Alternative Technology: The AgriClean technology includes mesophilic digestion and solid separation. During the construction and installation of the AgriClean components, an AgriJet system was installed in 5 of the 12 finishing houses to provide flush – style solid/waste removal from the houses through water pressure. Each of the 5 AgriJet installed houses was manually flushed daily into an in-ground equalization (EQ) tank/ pump station. Waste was pumped from the EQ tank to the 255,000 gallon (total capacity) mesophilic digester and then stored post digestion in a 39,000 gallon settling/ EQ tank. The technology was designed to pump the undigested and settled solids through a Fan separator for additional solid removal with the liquid portion of the waste stream being delivered to the lagoon and the solid component being land applied or further processed. Biogas produced as result of mesophilic digestion was stored or flared.

1

• A conceptual flow-diagram of alternative technology;

1 234 5678 910 11 12

Anaerobic Lagoon Equalization tank

Methane storage tank

Solids

Solid Separator (Fan)

Heat exchanger / Recirculation

Storage Tank

Mesophilic Digester

N

= Houses w/ AgriJet system = Houses w/ existing pit recharge system

Flare

Figure 1.1 Conceptual flow diagram of AgriClean System (B R Harris Farm).

(Source; http //www.cals.ncsu.edu:8050/waste_mgt)

• Possible points of emissions of ammonia on conceptual flow-diagram and parameters that are important in controlling emissions:

o Water holding structures: Anaerobic lagoon and storage tank- water temperature and water chemistry (pH and TAN) are the major controlling factors.

o Animal houses: house operational technology flushing sequence and frequency are controlling variables as well as pH and TAN

o Solid Separator o Flare

2

An aerial photo of B. R. Harris farm with EST is given below:

Aerial photo of Agriclean (B.R. Harris farm).

3

• Table 1.1 Description of Animal Operation (value estimates provided by project investigators and/or animal contract company)

January 17-28, 2005 House number WEEK 1 1/17-1/23

1 2 3 4 5 6 7 8 9 10 11 12

# of pigs / house

875 953 992 897 914 877 931 1003 994 981 920 866

Wks in finishing

11 11 11 10 10 10 10 9 9 9 8 8

Ave. Wt of pigs (lbs.)

181 181 181 169 169 169 169 157 157 157 145 145

Feed consumed (lb/pig/wk)

34.4 34.4 34.4 33.3 33.3 33.3 33.3 32.1 32.1 32.1 31.2 31.2

House number

WEEK 2 1/24-1/28

1 2 3 4 5 6 7 8 9 10 11 12

# of pigs / house

873 950 991 895 914 877 927 1002 991 981 918 864

Wks in finishing

12 12 12 11 11 11 11 10 10 10 9 9


193 193 193 181 181 181 181 169 169 169 157 157


35.5 35.5 35.5 34.4 34.4 34.4 34.4 33.3 33.3 33.3 32.1 32.1

4

• Feed Nutrients

Table 1.2 Total elemental analysis of feed samples (5 samples in total, %N measurement is replicated 5 times, %P, Cu, Zn, measurements are replicated 3 times).

Date %N %P Cu(ppm) Zn(ppm)

January 17, 2005 2.46 ± 0.11 0.45±0.02 167.8±10.7 107±7

Nitrogen Excretion

Computation of Nitrogen Excretion Based on Animal Feed Data (B. R. Harris farm: AgriClean Technology-Evaluation period, January 17 – January 28, 2004) Note: Sampling was only conducted the week of January 24, therefore only that week’s production data was used to calculate nitrogen excretion.

• Animal population / Types: o Total number of pigs (finishing) in 12 finishing houses = 11183 o Weighted average weight of the pigs =177.00 lb/pig = 80.29 kg/pig

• Nitrogen Intake o Average feed consumed = 15.43 kg/pig/wk o Average nitrogen content of the feed = 2.46% (from Feed Analysis) o Average nitrogen intake per pig = 0.38 kg-N/pig/wk

• Nitrogen Excretion

o Average gain / feed or feed efficiency rate (ER) for feeder-finish operation, based on the 1999 Pig CHAMP data = 0.3

o Average N excretion = (1-0.3) x 0.38 = 0.27 kg-N/pig/wk o Average N excretion on animal weight (lw) basis = 3.31 kg-N/1000kg animal

live weight(lw)/wk

5

Meteorological Measurements Monthly/Annual Climate Data Results at the nearest weather station (Source: State Climatology Office) Approx 15 km from sampling site

Summary of monthly precipitation (cm) from 1995 to 2005

GREENVILLE, NC (UCAN: 14125,COOP: 313638) Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Ann 1995 11.63 14.83 10.08 5.46 10.64 29.57 13.61 8.92 7.24 11.20 9.32 4.57 137.08 1996 11.79 6.63 10.52 6.71 14.73 11.20 25.55 8.28 29.59 15.80 7.21 9.50 157.51 1997 7.98 7.77 9.09 9.93 8.81 8.10 10.97 15.24 12.60 5.00 10.31 12.60 118.41 1998 15.67 19.15 12.90 13.21 7.57 7.85 18.03 26.29 12.47 3.38 8.41 14.83 159.77 1999 13.41 3.89 6.76 5.44 6.73 7.77 9.58 12.14 67.84 10.41 2.92 3.35 150.24 2000 13.16 4.42 8.08 11.48 4.09 11.68 17.07 24.66 18.80 0.38 5.41 3.89 123.11 2001 3.00 7.16 13.46 5.13 9.55 15.21 14.33 13.61 2.87 2.21 2.18 2.21 90.93 2002 18.16 4.01 9.19 7.90 10.16 7.52 12.75 9.86 11.91 20.45 18.19 9.60 139.70 2003 3.89 13.21 13.31 11.94 19.63 2.84 18.90 35.43 25.98 11.28 3.76 11.89 172.06 2004 3.71 10.01 3.81 7.44 13.54 13.11 16.97 25.73 8.18 2.34 9.91 5.28 120.02 2005 5.92 4.85 11.23 6.10 12.62 8.74 11.79

AVG 9.85 8.72 9.86 8.25 10.73 11.24 15.41 18.02 19.75 8.24 7.76 7.77

B.R. Harris Precipitation Data Analysis GREENVILLE, NC (UCAN: 14125,COOP: 313638) Compared to the 10-year precipitation average of 9.9 cm for the month of January (1995-2004), B.R. Harris, conducted for January 17-28, 2005, showed a lower precipitation average of 5.9cm, a difference of 4.0 cm, however this is within the range for the last ten years.

Summary of monthly mean temperature (oC) from 1995 to 2005 GREENVILLE, NC (UCAN: 14125,COOP: 313638) Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Ann 1995 6.43 5.68 11.55 16.87 20.34 24.04 27.29 26.55 22.00 18.20 8.74 3.98 15.971996 4.50 6.37 8.55 16.00 20.82 24.96 26.40 24.69 22.66 16.97 8.49 8.96 15.781997 5.64 9.04 12.91 13.87 18.43 22.69 26.55 24.69 22.37 16.12 9.93 6.24 15.711998 8.29 9.39 11.34 16.04 21.31 25.68 26.94 26.19 23.87 16.58 11.78 9.29 17.231999 9.04 8.21 9.24 17.00 19.78 23.61 27.34 26.65 22.29 16.69 13.66 7.71 16.772000 4.72 8.70 12.88 15.25 22.29 25.43 25.10 24.89 22.21 16.02 9.64 3.14 15.862001 6.27 9.01 10.29 16.86 20.96 25.74 24.85 26.24 21.21 15.89 13.79 9.78 16.742002 6.87 8.11 12.72 18.07 20.18 25.53 27.48 26.68 24.18 18.06 10.82 6.06 17.062003 3.42 6.25 12.97 15.82 20.28 24.85 26.66 27.03 22.72 16.42 14.82 6.21 16.452004 4.06 6.08 11.74 16.86 23.42 25.00 27.06 24.72 23.12 17.22 12.33 6.77 16.532005 6.73 7.21 9.47 16.40 19.00 24.76 27.89

AVG 5.93 7.68 11.42 16.26 20.78 24.75 26.57 25.84 22.66 16.82 11.40 6.81

6

B.R. Harris Mean Temperature Data Analysis GREENVILLE, NC (UCAN: 14125,COOP: 313638) Compared to the 10-year temperature average of 5.9oC for the month of January (1995-2004), B.R. Harris, conducted for January 17th- 28th, 2005, showed a slightly higher temperature average of 6.7oC , a difference of 0.8oC, however this is within the range for the last ten years.

• Site Meteorological data measured during the measurement periods:

Hourly Averaged Wind Speeds

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

0:00 6:00 12:00 18:00 0:00

Time (EST)

Win

d Sp

eed

(m/s

ec)

Hourly Averaged Air Temperatures

-2.0

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

0:00 6:00 12:00 18:00 0:00

Time (EST)

Air T

empe

ratu

re (

o C)

Figure 1.2 Site meteorological data during the measurement period (January 17-28, 2005). Error bar indicates ±1 standard deviation of 15 minute averages.

7

Joint Frequency DistributionFor Raw Data File C:\DSKIM\IAN\AGRICLEAN\AgricleanWD.csv

N

S

W E

No observations were missing.Wind flow is FROM the directions shown.Rings drawn at 5% intervals.Calms excluded.

2.34 1.75

2.92

1.75

0.58

0.58

4.09

8.77

24.56

19.88

10.53

9.36

5.85

4.68

1.75

0.58

Wind Speed ( Meters Per Second)0.1 1.54 3.09 5.14 8.23 10.8

Figure 1.3 Wind rose depicting % wind direction during the measurement period (January 17-28 2005)

8

Measurement of Ammonia Fluxes and Emissions Emission Sources -

Major sources of NH3 are the hog houses, the anaerobic lagoon and the storage tank. Other possible sources include the solid separator and the flare. In all of the liquid waste environments, the NH3 fluxes are expected to depend on ambient air temperature, water temperature, pH, wind speed and N in waste effluent. The flux chamber was deployed on water-holding structures measuring NH3 fluxes directly from their surfaces. For the houses, NH3 emission was determined by using average NH3 concentration across plumes from exhaust fans and estimated air flow rate from fans. Dynamic-Chamber Technique for NH3 flux measurement

The measurement schedule followed for determining the flux of ammonia from the water-holding structures using the dynamic-chamber technique is described in Table 1.3. Measured flux (presented as hourly averages) as a function of time is presented in Figure 1.5. Tabulated hourly average flux values for each water-holding structure are presented in Table 1.4. Table 1.4 also contains the overall average flux values for each water-holding structure for each evaluation period. Table 1.5 contains TAN and TKN concentrations of the effluent from the water-holding structures. Table 1.6 presents total emissions of ammonia (kg-N) per week for each water-holding structure calculated for each evaluation period and normalized to 1000 kg live weight of animals present. - B.R. Harris Farm (Measurement Period: January 17-28, 2005)

Table 1.3 NH3 emission measurement schedule at B.R. Harris farm

Sample dates Parameters Instruments Sample plots Remarks January 24-26, 2005

NH3 flux, lagoon T, lagoon pH, WD, WS, SR, air T, RH

One NH3 analyzer, Meteorological instruments

anaerobic lagoon

Completed 2 diurnal measurements

January 26-27, 2005

NH3 flux, , lagoon T, lagoon pH, WD, WS, SR, air T,RH


Storage tank Completed 1 diurnal measurements

T = temperature; WD = wind direction; WS = wind speed; SR = solar radiation; RH = relative humidity Water samples at each plot were collected every day for analysis of TAN and TKN concentrations at the laboratory.

9

Site photo during experimental period

Overview of experiment at B.R. Harris Farm

1 234 5678 910 11 12

Equalization tankAnaerobic Lagoon

Mesophilic Digester

Storage Tank



= Houses w/ AgriJet system = Houses w/ existing pit recharge system

Flare

Sampling Locations

Meteorological Tower

Flow of Hog waste through pipes

Mobile Laboratory

Solids

Solid Separator (Fan)

N

Figure 1.4 Experimental site layout and measurement locations.

10

Measurement period (January 17-28, 2005)

Composite hourly average NH3 flux(Anaerobic lagoon)

0

200

400

600

800

0:00 6:00 12:00 18:00 0:00

Time (EST)

NH 3

flux

( µg-

Nm

-2m

in-1

)

Composite hourly averaged NH3 flux(Storage tank)

0

1000

2000

3000

4000

0:00 6:00 12:00 18:00 0:00

Time (EST)

NH

3 ( µ

g-N

m-2

min

-1)

Figure 1.5 Diurnal variation of NH3 flux from the anaerobic lagoon and the storage tank. Error bar indicates ±1 standard deviation of 15 minute averages.

11

Table 1.4 Summary of hourly and overall averaged NH3 flux from the water holding structures during the experimental period. AgriClean NH3 flux (µg-N m2 min-1) (Sampling period: 1/24-1/28/2005)

Anaerobic lagoon Storage tank

EST hrly avg stdev hrly avg stdev

0:00 264.3 10.7 1419.8 11.4 1:00 259.8 14.7 1295.6 55.8 2:00 258.1 15.6 1255.4 27.0 3:00 279.6 22.9 1205.3 29.1 4:00 280.1 36.0 1121.5 56.8 5:00 274.7 40.0 6:00 270.8 49.5 7:00 272.5 62.7 8:00 309.9 27.1 9:00 280.6 45.8

10:00 257.3 94.3 11:00 12:00 13:00 14:00 15:00 529.6 155.2 16:00 186.9 66.6 2736.9 264.7 17:00 197.9 53.3 1872.1 248.9 18:00 248.5 60.9 1597.7 25.8 19:00 243.4 57.2 1603.7 58.1 20:00 235.0 13.8 1583.6 42.5 21:00 229.8 29.6 1521.5 34.5 22:00 244.9 18.8 1434.7 25.1 23:00 258.1 14.9 1385.6 28.1

average† 269.1 1541.0 stdev 67.5 411.3

# of data 20.0 13

average‡ 260.7 1508.7 stdev 64.8 331.2

# of data 153.0 48

(15 min) Tlag=2.9±0.9(n=151)∆ † Statistics for hourly averages ‡ Statistics for 15 minute averages for the experimental period

∆ Does not include storage tank

12

Table 1.5 Total Ammoniacal Nitrogen (TAN) and Total Kjeldahl Nitrogen (TKN) averages and their standard deviation from water-holding structures at B.R. Harris farm. Anaerobic lagoon Storage tank TKN

(mg-N l-1) TAN (mg-N l-1)

TKN (mg-N l-1)

TAN (mg-N l-1)

1st Period (Jan 17- 28)

634.5±35.8 n=4

580.5±28.4 n=4

1356.5±27.6 n=2

1177.0±8.5 n=2

n represents the total number of effluent samples collected at each water-holding structure. Table 1.6 Summary of total emissions from water-holding structures at AgriClean during the experimental period. Water holding structure Anaerobic lagoon Storage tank Area (m2) 29809.4 44.2 Weekly NH3 emission (kg-N/wk)

78.3 0.7

Total emission from tanks and lagoon (kg-N/wk)

79.0

Total emission/pig (kg-N/pig/wk)

0.007

Total emission/1000 kg-lw (kg-N/1000kg-lw/wk)

0.09

13

Open-Path Fourier Transform Infrared (OP-FTIR) Spectrometers OP-FTIR spectrometer concentration measurements were obtained during January 25 and 26, 2005. Data was collected from two barns: a standard barn on January 25th and a barn with the AgriClean Jet system on January 26th, see Figure 1.6. The measurements were collected through the centerline of the fans on each barn; the operation of the fans was recorded during the measurement period. Figure 1.7 shows the 15-minute average concentrations and standard deviations of Nitrogen in mg—N/m3. The average concentrations were 0.806 and 1.12 mg—N/m3 for the barns with the AgriClean Jet system and the standard barn. Figure 1.6 Locations of Measurements taken with the OP-FTIR Spectrometers.

Digester

EQ Tank Solid Sep (fan)


EQ tank from Houses

Lagoon

Sep liquid back to lagoon


House 12

House1

FTIR Measurements

Flare

14

Figure 1.7 Fifteen minute Average Concentrations and Standard Deviations of Nitrogen Measured in January 2005.

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

930

1000

1030

1100

1130

1200

1230

1300

1330

1400

1430

1500

1530

1600

EST

Barn with JetsStandard Barn

Flux Calculations for the Barns The average nitrogen flux from the hog houses was calculated by multiplying concentrations of nitrogen (measured as ammonia mg/m3) across the midline of the fans by the factory calibrated fan rates (m3/min) for the fans that were on at each time point. The concentrations that were measured were adjusted for the length of the path across the operating fans at each time point. The concentration for each barn was then normalized by the total live weight of the hogs in the houses at the time of the sampling (1000 Kg LW). From this a weighted average of flux from Jet barns and Standard barns was calculated. The operation of the fans (monitoring when they are on or off) was performed during the entire sampling period.

Table 1.7 Flux Measurements KgN/Week/1000Kg Live Weight

Jet Barn Standard Barn Weighted Average of Jet barns and Standard barns

KgN/Week/1000 KG 0.572 0.975 0.81

15

Assessment of Ammonia Emissions from Alternative Technology:

At each alternative technology and conventional site, the estimated ammonia emissions are usually limited to two two-weeks long periods, representing warm and cold seasons. But, since measurements at different sites are made at different times of the year, environmental conditions are likely to be different at different sites, even during a representative "warm" or "cold" season. There is a need for accounting for these differences in our relative comparisons of the various alternative and conventional technologies.

The estimated emissions from water-holding structures at an alternative technology for each measurement period are compared with the average estimated emissions from baseline sites, after the later are adjusted to the average environmental parameters (lagoon temperature and air temperature) observed at the former (alternative technology) site. A rational basis for this adjustment for somewhat different environmental conditions is the multiple regression model developed for ammonia emissions and measured environmental parameters at the two baseline sites. The model is described in appendix 2 of the three-year progress report. Such a comparison would not require highly uncertain extrapolations of emissions at alternative technology sites beyond the two measurement periods.

Absolute numbers are not used in assessing ammonia emissions from the proposed alternative technology. A normalized measure of emissions (normalized to calculated N-excreted; %EEST) is compared to a similar normalized measure of emissions (%ECONV) from a baseline site using the conventional lagoon technology for handling swine waste in North Carolina. The %E values are an estimate of rate of loss of N compared to N excreted. Two baseline sites are used to account for differences in housing ventilation across the sites with the proposed EST’s. No method exists for adjusting baseline housing emissions to environmental conditions observed at an EST farm. Therefore, actual housing emissions measured at the baseline sites during comparable seasons of the year are used when generating the normalized measures of emissions from houses. It is acknowledged that the housing emissions for the baseline sites were not made under the exact meteorological conditions as the housing measurements for evaluation of an EST. The algorithm followed in deriving an index of performance (%reduction = [(%ECONV - %EEST)/%ECONV] * 100) by the EST in reducing ammonia emissions as compared to the conventional technology currently in use in North Carolina (baseline sites) is presented in Figure 1.8 for water holding structures.

16

Figure 1.8. Algorithm flow chart for evaluation of alternative technology ammonia emission from water holding structures.

Analysis of conventional farm (Moore & Stokes Farm) measurement data for NH3 lagoon emissions with lagoon and air temperature

Establish an observational model based on conventional farm measurement data of NH3 emission and lagoon and air temperature during different seasons

Log10 (NH3 emission) = a + b • Tlagoon+ c • D; a, b and c; experimental constants, Tlagoon; lagoon temperature, Tair; air temperature D; ∆T when ∆T>0, 0 when ∆T<0; ∆T= Tair- Tlagoon

Analysis of each technology farm (EST farms) measurement data for NH3 lagoon emissions with lagoon and air temperature

Estimate conventional NH3

emission for evaluation EST by using lagoon and air temperatures from EST, and multiple linear regression model

Plug temperatures into model

Measured EST average NH3 emission (Fmeas) during the experimental period

Estimated conventional average NH3 emission (Fproj) projected by the regression model

Subtract the measured EST emissions from the projected conventional model emissions, then divide by the projected conventional model emissions and multiply by 100 to find the % reduction

No reduction of NH3 emission

No

Yes Reduction of NH3 emission by alternative technology (The higher the % reduction, the more effective the technology.)

Multiple linear regression analysis

% reduction > 0

17

Evaluation of B.R. Harris farm (AgriClean) We compare the lagoon NH3-N emission from B.R. Harris farm with the projected average

emission from lagoon at the conventional farm, using the observational statistical (multiple linear regression) model.

Table 1.8 gives animal weight, feed consumed, and N-excretion at baseline farms and B.R.

Harris farm. Table 1.9 gives the NH3-N emissions (kg-N/1000 kg-live weight/wk) data summary for the B.R. Harris farm and baseline farms for evaluation of EST at the former. The emissions from different components of an EST or baseline farm should be viewed relative to the estimated nitrogen excretion from animal population, weight and feed data.

Table 1.8 Summary of animal weight, feed consumed, and N-excretion at conventional farms (Stokes and Moore) and the EST (B.R. Harris; AgriClean) farm.

Farm No. of pigs average pig

weight total pigs weight feed

consumed N-excretion, E

Information kg/pig kg kg/pig/wk kg-N/wk/

1000kg-lw Stokes (Jan.) 3,727 88.5 329,840 12.59 2.51 Moore (Feb.) 5,784 67.0 387,528 12.37 3.90

B.R. Harris (Jan.) 11,183 80.3 897,995 15.43 3.31

18

Table 1.9: Estimates of % reduction in NH3-N emissions from different components and their sum total at the EST (B.R. Harris: AgriClean) and conventional farms (kg-N/wk/1000kg-lw). (% reduction = [(%ECONV- %EEST)/%ECONV]*100

(1)Anaerobic lagoon and storage tank emissions Period

Average lagoon temperature (oC)

Average D (oC)

Conventional model emissions Fproj

% ECONV

AgriClean measured emission Fmeas

% EEST

% reduction

Jan 24-27 ,2005 2.9 0.7 0.09 2.8 0.09

2.7

3.6

(2) Barn Emissions Period

Moore Farm measured emission

% ECONV

AgriClean measured emission Fmeas

% EEST

% reduction

Jan 25-26,2005 0.89‡ 22.8 0.81

24.5

-7.5

Total Emissions (1)+(2) Period

conventional total emission

% ECONV

AgriClean measured emission

% EEST

% reduction

Jan 24-27,2005 0.98 25.6 0.90

27.2

-6.3

D is ∆T, the difference between the air temperature (Tair) and lagoon temperature (Tlag), when Tair > Tlag ; D = 0 when Tair < Tlag. Fproj is baseline lagoon area adjusted NH3 lagoon emission projected by the baseline multiple linear regression model corresponding to the average lagoon temperature and the average D during B.R. Harris (Agriclean) farm measurement period. % ECONV is the conventional model emissions relative to the N excreted. % EEST is the measured emission from the EST relative to the N excreted. Fmeas is the NH3 emissions from water holding structures and NH3 emissions from barn house measured at B.R. Harris (AgriClean) farm. Soil flux measurements were not taken because there was no lagoon spray and land application during the experimental period. ‡: Overall house emission measured at Moore farm during February 2003. % reduction is used to describe how effective a technology is, in reducing NH3 emissions. A number > 0 indicates a reduction in NH3.The larger the % reduction, the more effective the technology is in reducing NH3 emissions. Conversely a number < 0 indicates that there are has been no reduction in NH3 emissions.

19

2. Evaluation of Environmentally Superior Technologies for Ammonia Emissions: Red Hill Farm

Environmental Technologies

Alternative Technology: “Closed loop” Swine waste treatment system Location: Red Hill Farm (Ayden, NC) Period of Operation: The OPEN team monitored for evaluation during: 1st field experiment: 03/21 – 04/08/2005 2nd field experiment: 07/18 – 08/05/2005 Technology contact: Don Lloyd (919-922-5399) NCSU Representative PI: Kurt Creamer (919-515-4092) Statement of Task:

- Measurement of ammonia (NH3) emissions from anaerobic lagoon, settling tanks, water tank, by using a flow-through chamber technology during two different campaigns (warm and cold seasons)

- Analysis of water samples from waste storage and treatment areas for Total Ammoniacal Nitrogen (TAN) and Total Kjeldahl Nitrogen (TKN) concentrations (one sample each day during the experimental period)

- On site monitoring of meteorological parameters at 10 meter height - FTIR technology used to determine ammonia emissions from barns - Parameters measured: NH3 flux , storage lagoon temperature and pH, soil temperature,

wind speed and direction, solar radiation, and air temperature - Analysis of emissions from composter by the use of the denuder system

Description of Alternative Technology: The primary objective of this “closed-loop” system is to treat the liquid fraction of the waste in such a way that it can be used both for flushing the hog barns and for hog drinking water. This could eliminate the need for the traditional hog waste lagoon. The closed-loop system treats the waste from three hog barns, with a steady-state population of 3672 finishing hogs. These barns use a flush system for removing the manure from the barns, which, prior to installation of the treatment system, flushed the waste into a lagoon. The first step in the closed loop process is collection of the waste in an “equalization” or buffering tank. The waste in the tank is continuously pumped to an inclined separator where the solids are collected and will be composted prior to field application on-farm or sold. The liquid collected from the separator is injected with a polymer flocculant and sanitizer/disinfectant and pumped into a settling tank, where flocculated solids collect at the bottom over a period of approximately four hours.

Most of the liquid fraction from the settling tank is returned to the hog barns for re-use as flush water. But when the flush tanks are full, excess water is pumped to a tertiary treatment

1

system. This system provides filtration and aeration and is housed in a septic tank. The treated water is blended with well water to achieve a solids content consistent with human drinking water standards for use as hog drinking water. Solids from the settling tanks are composted along with the solids from the inclined separator.

• A conceptual flow-diagram of alternative technology;

Existing Anaerobic Lagoon N

Hog Houses

NORWECO system

Static mixers

Settling tank 1 H20 tank

Settling tank 2

TMC Polymer

Solid SeparatorTrailer (under shed)

Equalization tank

Pasture

Figure 2.1 Conceptual flow diagram of Environmental Technologies (Red Hill).

(Source; http //www.cals.ncsu.edu/waste_mgt)

• Possible points of emissions of ammonia on conceptual flow-diagram and parameters that are important in controlling emissions:

o Water holding structures: anaerobic lagoon, settling tanks and water tank - water temperature and water chemistry (pH and TAN) are the major controlling factors.

o Animal houses: house operational technology flushing sequence and frequency are controlling variables as well as pH and TAN

o Composter

2

An aerial photo of Red Hill farm with EST is given below:

Aerial photo of Environmental Technologies site (Red Hill farm).

3

Table 2.1 Description of Animal Operation (value estimates provided by project investigators and/or animal contract company) Sampling period (1st Evaluation) March 21- April 8, 2005 WEEK 1 3/21-3/27

House 1 Finishing

House 2 Finishing

House 3 Finishing

# of pigs / house

1046 671 673

Wks in finishing

13 13 13


145 145 145


35.0 35.0 35.0

WEEK 2 3/28-4/3

House 1 Finishing

House 2 Finishing

House 3 Finishing

# of pigs / house

1042 671 673

Wks in finishing

14 14 14


160 160 160


35.0 35.0 35.0

WEEK 3 4/4-4/8

House 1 Finishing

House 2 Finishing

House 3 Finishing

# of pigs / house

861 490 485

Wks in finishing

15 15 15


170 170 170


35.0 35.0 35.0

4

Table 2.2 Sampling period (2nd Evaluation): July 18 – August 5, 2005 WEEK 1 7/18-7/24

House 1 Finishing

House 2 Finishing

House 3 Finishing

# of pigs / house

883 756 1527

Wks in finishing

12 12 12


135 135 135


35 35 35

WEEK 2 7/25-7/31

House 1 Finishing

House 2 Finishing

House 3 Finishing

# of pigs / house

875 753 1523

Wks in finishing

13 13 13


145 145 145


35 35 35

WEEK 3 8/1-8/5

House 1 Finishing

House 2 Finishing

House 3 Finishing

# of pigs / house

811 727 1483

Wks in finishing

14 14 14


160 160 160


35 35 35

5

• Feed Nutrients

Table 2.3 Total elemental analysis of feed samples (5 samples in total, %N measurement is replicated 5 times, %P, Cu, Zn, measurements are replicated 3 times).

Date %N %P Cu(ppm) Zn(ppm)

March 28, 2005 3.17 ± 0.16 0.71±0.03 34.5±3.5 155±10

July 18, 2005 2.81 ± 0.11 0.68±0.03 34.6±6.2 155±7

Nitrogen Excretion

Computation of Nitrogen Excretion Based on Animal Feed Data (Red Hill farm: - Environmental Technologies Evaluation period, March 21 – April 8, 2005) Note: Sampling was only conducted the week’s of March 21 & March 28, therefore only those week’s production data was used to calculate nitrogen excretion.

• Animal population / Types: o Total number of pigs in 3 finishing houses = 2390 o Weighted average weight of the pigs =152.50 lb/pig = 69.2 kg/pig





live weight(lw)/wk

Computation of Nitrogen Excretion Based on Animal Feed (Red Hill Farm: Environmental Technologies-Evaluation period, July 18 – August 5, 2005)

• Animal population / Types:

o Total number of pigs in 3 finishing houses = 3113 o Weighted average weight of the pigs =146.7 lb/pig = 66.52 kg/pig

6





live weight(lw)/wk

7

Meteorological Measurements Monthly/Annual Climate Data Results at the nearest weather station (Source: State Climatology Office) Approx 20 km from sampling site

Summary of monthly precipitation (cm) from 1995 to 2005

KINSTON 5 SE, NC (UCAN: 14167,COOP: 314684) Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Ann 1995 9.45 12.40 11.20 1.65 6.71 35.28 11.23 5.84 7.32 12.90 7.57 4.27 125.811996 9.86 4.55 13.16 7.57 10.92 22.20 18.69 2.72 28.07 15.27 9.42 8.10 150.521997 7.29 7.57 9.30 6.68 6.38 4.60 22.38 12.85 8.46 3.68 11.84 13.67 114.681998 14.96 17.63 10.08 10.92 11.51 11.07 15.57 19.56 6.50 1.09 4.80 13.41 137.111999 15.06 4.39 9.35 6.78 5.44 9.37 19.94 9.93 58.50 20.09 4.42 2.13 165.402000 13.54 5.38 11.38 10.72 6.05 8.81 18.01 15.80 18.52 0.36 7.52 4.98 121.062001 2.90 7.39 10.92 3.45 6.63 23.09 18.16 12.09 5.74 2.90 2.34 2.18 97.792002 13.13 4.95 16.05 11.48 5.41 12.04 14.07 16.66 6.53 6.60 10.85 7.49 125.272003 4.04 14.07 12.27 13.89 19.46 10.87 23.83 11.89 17.02 18.49 6.50 12.19 164.522004 2.72 10.92 2.06 10.97 12.65 19.05 14.68 21.82 7.98 5.51 10.29 6.65 125.302005 6.55 5.21 9.07 9.47 13.39 10.44 15.77 4.09 0.00 73.99

AVG 9.29 8.93 10.58 8.41 9.11 15.64 17.66 12.92 16.46 8.69 7.55 7.51

Red Hill Precipitation Data Analysis KINSTON 5 SE, NC (UCAN: 14167,COOP: 314684) Compared to the 10-year precipitation average of 10.6 cm for the month of March (1995-2004), Red Hill, conducted for March 21 - April 8, 2005, showed a slightly lower precipitation average of 9.1 cm, a difference of 1.5 cm, however this is within the range for the last ten years. Compared to the ten year precipitation average of 17.7 cm for the month of July, Red Hill, conducted for July 18- August 5, 2005, showed a slightly lower precipitation average of 15.8 cm. Summary of monthly mean temperature (oC) from 1995 to 2005 KINSTON 5 SE, NC (UCAN: 14167,COOP: 314684) Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Ann 1995 6.09 4.93 11.31 16.16 19.78 23.28 26.39 25.36 21.41 17.63 8.38 3.78 15.371996 4.72 5.84 8.13 15.27 20.48 24.52 25.97 24.90 22.46 16.42 8.82 8.44 15.501997 5.61 8.51 13.60 13.69 18.22 22.52 26.32 24.05 22.12 16.08 9.85 7.02 15.631998 7.85 8.58 11.09 16.09 20.97 25.51 26.02 24.55 23.90 16.68 11.82 9.97 16.921999 8.57 7.86 9.40 16.83 19.59 23.68 26.77 26.14 21.47 16.14 13.74 7.26 16.452000 4.50 8.56 12.89 14.90 22.12 25.06 25.02 24.87 22.15 15.88 9.41 3.28 15.722001 5.41 8.55 9.87 16.25 20.38 25.11 24.64 25.27 21.21 15.51 13.23 10.19 16.302002 6.84 8.45 11.77 18.06 19.71 25.02 26.49 25.73 23.97 18.62 10.57 5.97 16.772003 3.78 6.08 12.65 15.08 20.26 24.32 26.17 26.17 22.06 15.49 14.93 5.69 16.062004 3.84 5.71 11.64 16.03 22.92 24.79 26.10 24.18 22.79 16.87 12.32 6.33 16.132005 6.87 7.38 9.36 15.22 18.39 24.27 27.37 26.50 25.83 AVG 5.72 7.31 11.24 15.84 20.44 24.38 25.99 25.12 22.35 16.53 11.31 6.79

8

Red Hill Mean Temperature Data Analysis KINSTON 5 SE, NC (UCAN: 14167,COOP: 314684) Compared to the 10-year temperature average of 11.2oC for the month of March (1995-2004), Red Hill, conducted for March 21st - April 8, 2005, showed a lower temperature average of 9.4oC, a difference of 1.8oC. Compared to the ten year temperature average of 27.4oC for the month of July, Red Hill, conducted for July 18- August 6th, 2005, showed a lower temperature average of 26.0oC, a difference of 1.4 oC.

• Site Meteorological data measured during the measurement periods:


0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

0:00 6:00 12:00 18:00 0:00

Time (EST)

Win

d Sp

eed

(m/s

ec)


0.02.04.0

6.08.0

10.012.014.0

16.018.020.0

0:00 6:00 12:00 18:00 0:00

Time (EST)

Air T

empe

ratu

re (

o C)

Figure 2.2 Site meteorological data during the 1st measurement period (March 21- April 8, 2005). Error bar indicates ±1 standard deviation of 15 minute averages.

9

Joint Frequency DistributionFor Raw Data File C:\DSKIM\IAN\SOS\SOSWD.csv

N

S

W E

0.67 0.00

2.00 23.33

18.67

5.33

4.67

8.67 5.33

1.33

2.00

2.67

0.67

0.67

7.33

16.67



Figure 2.3 Wind rose depicting % wind direction during the measurement period (March 21-April 8, 2005)

10


0.00.51.01.52.02.53.03.54.04.55.0

0:00 6:00 12:00 18:00 0:00

Time (EST)

Win

d Sp

eed

(m/s

ec)


0.0

5.0

10.0

15.0

20.0

25.0

30.0

0:00 6:00 12:00 18:00 0:00

Time (EST)

Air T

empe

ratu

re (

o C)

Figure 2.4 Site measurement data during 2nd measurement period (July 18-August 5, 2005). Error bar indicates ±1 standard deviation of 15 minute averages.

11

Joint Frequency DistributionFor Raw Data File C:\DSKIM\IAN\SOS2\SOSWD.csv

N

S

W E


0.00 0.00

0.00

0.00

0.00

3.45

0.00

0.00 0.00

13.79

65.52

17.24

0.00

0.00

0.00

0.00


Figure 2.5 Wind rose depicting % wind direction during the measurement period (July 18-August 5, 2005)

12

Measurement of Ammonia Fluxes and Emissions Emission Sources -

Major sources of NH3 are the hog houses, anaerobic lagoon, settling tanks, composter and biogenic emissions from soils during land applications. In all of the liquid waste environments, the NH3 fluxes are expected to depend on ambient air temperature, water temperature, pH, wind speed and N in waste effluent. The flux chamber was deployed on water-holding structures and soil in order to measure NH3 fluxes directly from their surfaces. For the houses, NH3 emission was determined by using average NH3 concentration across plumes from one side of hog house and estimated air flow rate from the side during the measurement period by open path FTIR. Emissions from the composter were determined by the denuder system. Dynamic-Chamber Technique for NH3 flux measurement

The measurement schedule followed for determining the flux of ammonia from the water-holding structures using the dynamic-chamber technique is described in Table 2.4. Measured flux (presented as hourly averages) as a function of time is presented in Figures 2.7 and 2.8. Tabulated hourly average flux values for each water-holding structure are presented in Table 2.5. Table 2.5 also contains the overall average flux values for each water-holding structure for each evaluation period. Table 2.6 contains TAN and TKN concentrations of the effluent from the water-holding structures. Table 2.7 presents total emissions of ammonia (kg-N) per week for each water-holding structure calculated for each evaluation period and normalized to 1000 kg live weight of animals present. Results are not shown for emissions from land application, this is due to the flux being negligible.

13

Table 2.4 NH3 emission measurement schedule at Red Hill Farm (1st and 2nd measurement periods, respectively: March 21- April 8, 2005; July 18- August 5, 2005)

Sample dates Parameters Instruments Sample plots Remarks March 21-29, 2005

NH3 flux, lagoon T, lagoon pH, WD, WS, SR, air T, RH


Anaerobic lagoon


March 30, 2005

NH3 flux, , lagoon T, lagoon pH


Settling Tank 1

Completed 2.5 hours of measurements

March 31, 2005

NH3 flux, lagoon T, lagoon pH


Settling Tank 2 Completed 1.25 hours of measurements

April 1, 2005



Water Tank Completed 2.5 hours of measurements

April 4, 2005 NH3 flux, Soil T One NH3 analyzer, Meteorological instruments

Soil Completed 3 hours of measurements


14

Table 2.4, continued. Sample dates Parameters Instruments Sample plots Remarks July 19-20 NH3 flux, lagoon T,

lagoon pH, WD, WS, SR, air T, RH


Anaerobic lagoon


July 26 & July 2, 2005



Settling tank 1 Completed 2 measurements of 2.25 hours and 0.75 hours, respectively

July 27 & August 1, 2005



Settling tank 2 Completed 2 measurements of 1.5 hours and 2.5 hours, respectively

July 21 & August 4, 2005



Water tank Completed 2 measurements of 1.25 hours and 4.75 hours, respectively


15

Site photos during experimental period

1st Evaluation: Overview of settling tanks and water tank

2nd Evaluation: View of anaerobic lagoon at Red Hill farm

16

Existing Anaerobic Lagoon

17

Figure 2.6 Experimental site layout and measurement locations.

Hog Houses

NORWECO system

Pasture Equalization tank

Solid SeparatorTrailer (under shed) Polymer TMC

Static mixers

Settling tank 1

H20 tank Settling tank 2

N

Sampling Locations

Meteorological Tower

Flow of Hog waste through pipes

Mobile Laboratory

1st Measurement period (March 21- April 8, 2005) Composite hourly averaged NH3 flux

(Anaerobic lagoon)

0

500

1000

1500

2000

2500

0:00 6:00 12:00 18:00 0:00Time (EST)

NH3 f

lux

( µg-

Nm-2

min

-1)

Composite hourly averaged NH3 flux (Settling Tank 1)

0500

100015002000250030003500

0:00 6:00 12:00 18:00 0:00Time (EST)

NH3 f

lux (

µg-N

m-2

min

-1)


0100020003000400050006000

0:00 6:00 12:00 18:00 0:00Time (EST)

NH3 f

lux

( µg-

Nm-2

min

-1)

18

Composite hourly averaged NH3 flux (Water Tank)

0255075

100125150175200

0:00 6:00 12:00 18:00

NH

3 flu

x ( µ

g-N

m-2

min

-1)

0:00Time (EST)

Figure 2.7 Diurnal variation of NH3 flux from anaerobic lagoon, settling tank 1, settling tank 2, and water tank during the 1st measurement period. Error bar indicates ±1 standard deviation of 15 minute averages. 2nd Measurement period (July 19-August 5, 2005)

Composite hourly averaged NH3 flux (anaerobic lagoon)

0500

10001500200025003000350040004500

0:00 6:00 12:00 18:00

NH3 f

lux

( µg-

Nm-2

min

-1)

0:00Time (EST)

19


0500

100015002000250030003500

0:00 6:00 12:00 18:00 0:00Time (EST)

NH3 f

lux ( µ

g-Nm-2 mi

n-1 )


0500

100015002000250030003500

0:00 6:00 12:00 18:00 0:00Time (EST)

NH

3 flu

x ( µ

g-N

m-2

min

-1)

Composite hourly averaged NH3 flux (Water Tank)

0

20

40

60

80

100

0:00 6:00 12:00 18:00 0:00Time (EST)

NH

3 flu

x ( µ

g-N

m-2

min

-1)

Figure 2.8 Diurnal variation of NH3 flux from anaerobic lagoon, settling tank 1, settling tank 2 and water tank during the 2nd measurement period. Error bar indicates ±1 standard deviation of 15 minute averages.

20

Table 2.5 Summary of hourly and overall averaged NH3 flux from the water-holding structures during the experimental periods. Red Hill 1st Period

NH3 flux (1st period: 3/21-4/8/2005)

Anaerobic

Lagoon Settling Tank 1

Settling Tank 2

Water Tank

EST hrly avg stdev hrly avg stdev hrly avg stdev hrly avg stdev

0:00 1230.2 436.8 1:00 1125.6 335.5 2:00 1270.1 450.3 3:00 1298.2 477.8 4:00 1199.6 417.8 5:00 1096.6 330.1 6:00 1092.9 377.3 7:00 1212.8 493.7 8:00 1334.1 508.1 9:00 1479.7 562.7

10:00 1425.8 517.0 11:00 1487.7 557.5 5469.4 26.4 12:00 1544.0 679.9 5504.5 0.0 120.5 57.3 13:00 1373.7 589.1 63.3 2.0 14:00 1166.3 322.3 2220.9 268.8 49.9 1.1 15:00 1087.8 174.5 1933.4 494.7 16:00 1047.0 143.6 2104.1 51.8 17:00 1173.7 138.3 18:00 1284.7 66.5 19:00 1314.2 131.0 20:00 1339.0 206.8 21:00 1388.4 370.4 22:00 1293.6 386.8 23:00 1282.2 425.9

average† 1272.8 2086.2 5486.9 77.9 stdev 135.6 144.6 24.8 37.5

# of data 24.0 3 2 3

average‡ 1279.0 2073.9 5492.8 80.4 stdev 399.3 321.6 26.2 45.0

# of data 280.0 11 6 11

(15 min) Tlag=14.9±1.5 (n=284)∆

† Statistics for hourly averages ‡ Statistics for 15 minute averages for the experimental period ∆ Does not include settling tank or water tank

21

Red Hill 2nd period NH3 flux (2nd period: 7/18-8/5/2005)

Anaerobic

Lagoon Settling Tank 1

Settling Tank 2

Water Tank

EST hrly avg stdev hrly avg stdev hrly avg stdev hrly avg stdev

0:00 2701.8 48.1 1:00 2603.5 32.2 2:00 2551.1 48.0 3:00 2487.4 38.6 4:00 2442.0 27.3 5:00 2402.6 30.7 6:00 2427.6 53.6 7:00 2296.6 150.5 8:00 9:00 58.1 1.0

10:00 679.2 42.3 594.8 16.2 49.9 3.6 11:00 612.0 38.6 930.9 390.0 46.9 2.8 12:00 925.6 576.6 1351.4 104.3 48.7 4.4 13:00 1111.5 528.2 1530.6 296.7 33.3 8.1 14:00 1526.7 12.4 1672.9 29.1 0.7 15:00 16:00 17:00 3713.2 130.5 18:00 3514.6 95.0 19:00 3338.3 77.2 20:00 3201.4 68.0 21:00 3033.3 40.1 22:00 2855.9 85.9 23:00 2726.6 74.9

average† 2819.7 971.0 1216.1 44.3 stdev 442.3 369.0 445.3 11.0

# of data 15.0 5 5 6

average‡ 2788.9 996.9 1223.3 43.3 stdev 408.8 488.4 491.6 10.4

# of data 58.0 24 22 22

(15 min) Tlag=31.6±0.8 (n=32)∆

† Statistics for hourly averages ‡ Statistics for 15 minute averages for the experimental period ∆ Does not include settling tanks or water tank

22

Table 2.6 Total Ammoniacal Nitrogen (TAN) and Total Kjeldahl Nitrogen (TKN) averages and their standard deviation from water-holding structures at Red Hill farm. n represents the total number of effluent samples collected at each water-holding structure. Anaerobic lagoon Settling Tank 1 Settling Tank 2 Water Tank TKN

(mg-N l-1) TAN (mg-N l-1)

TKN (mg-N l-1)

TAN (mg-N l-1)

TKN (mg-N l-1)

TAN (mg-N l-1)

TKN (mg-N l-1)

TAN (mg-N l-1)

1st Period (Mar 29- Apr 2)

925.0±22.4 n=4

722.3±37.5 n=4

1728 n=1

1480.0 n=1

3192.0 n=1

2415.0 n=1

0.64±0.18 n=2

0.0 n=2

2nd Period (Jun 28-Jul 2)

628.0 n=1

523.0 n=1

1171.0 ±44.1 n=1

1156.0 n=1

481.0±4.6 n=3

384.0±4.6 n=3

0.13±0.18 n=2

0.0 n=2

Table 2.7 Summary of total emissions from water-holding structures at Environmental technologies during the experimental periods. 1st Period Water holding structure Anaerobic lagoon Settling tank 1 Settling tank 2 Water tank Area (m2) 10091.3 5.8 5.8 5.8 Weekly NH3 emission (kg-N/wk)

130.1 0.12 0.32 0.00


130.5


0.05


0.79

2nd Period Water holding structure Anaerobic lagoon Settling tank 1 Settling tank 2 Water tank Area (m2) 10091.3 5.8 5.8 5.8 Weekly NH3 emission (kg-N/wk)

283.7 0.06 0.07 0.00


283.8


0.09


1.37

23

Open-Path Fourier Transform Infrared (OP-FTIR) Spectrometers OP-FTIR spectrometer concentration measurements were obtained March 23-24 and July 26-27, 2005. For both measurement periods, data was collected in the building with the tanks for the technology and at the long sides of one of the barns across the curtain opening see Figure 2.9 Figure 2.10 shows the 15-minute concentrations in mg--N/m3 for all locations in March, 2005. Figure 2.11 shows the 15-minute concentrations in mg--N/m3 for all locations in July, 2005. Table 2.8 lists the average daily concentrations of nitrogen in mgN/m3. Figure 2.9 Locations of Measurements taken with the OP-FTIR Spectrometers.

TMC Polymer

Static mixers

NORWECO system

Settling tanks This is enclosed in a building.

EQ tank for barn flush Solid separator to trailer (which under a shed)

Existing lagoon

N

FTIR Measure-

ments

24

Figure 2.10 Fifteen minute Average Concentrations and Standard Deviations of Nitrogen Measured in March, 2005.

0

0.5

1

1.5

2

2.594

5

1015

1045

1115

1145

1215

1245

1315

1345

1415

1445

1515

1545

EST

Inside BuildingUpwind Barn

Downwind Barn

Figure 2.11 Fifteen-minute Average Concentrations and Standard Deviations of Nitrogen Measured in July, 2005.

0.000

0.500

1.000

1.500

2.000

2.500

800

830

900

930

1000

1030

1100

1130

1200

1230

1300

1330

1400

1430

1500

1530

1600

EST

Inside BuildingUpwind Barn

Downwind Barn

25

Table 2.8 Average Daily FTIR Measurements Mg/M3 Nitrogen

Inside Building Upwind Barn Downwind BarnMarch 1.213 0.219 0.459July 0.620 0.278 1.492

Flux Calculations for the Barns To calculate the average nitrogen flux from the naturally ventilated houses, air flow measurements were made by sampling at one point in each of four equal sections along the upwind side of the building several times during the time period that the OP-FTIR spectrometer was measuring concentrations. Each point was sampled for 30-60 seconds and the maximum and minimum readings were recorded over a 5-7 minute period of time. The dimensions of the curtain opening were measured and the average velocity was calculated using the curtain opening and the sum of each of the four sampled sections. Calculations were performed in the following way. Average nitrogen concentrations (ppm) were obtained for the upwind and downwind sides of the house. The concentrations of ammonia were then converted to mg/m3 of nitrogen, and the upwind measurements were subtracted from the downwind measurements for each time period. A moving average was performed to reduce the effect of wind variations (times when the wind deviated from the predominate direction) and were combined with the calculated ventilation rates to calculate flux. The flux calculations for the house were then normalized by the total live weight of the hogs in the houses at the time of the sampling (1000 Kg LW), see Table 2.9. Table 2.9 Flux Measurements KgN/Week/1000Kg Live Weight March July KgN/Week/1000 KG 0.039 0.743

26

Assessment of Ammonia Emissions from Alternative Technology:

At each alternative technology and conventional site, the estimated ammonia emissions are limited to two two-weeks long periods, representing warm and cold seasons. But, since measurements at different sites are made at different times of the year, environmental conditions are likely to be different at different sites, even during a representative "warm" or "cold" season. There is a need for accounting for these differences in our relative comparisons of the various alternative and conventional technologies. The estimated emissions from water-holding structures at an alternative technology for each measurement period are compared with the average estimated emissions from baseline sites, after the later are adjusted to the average environmental parameters (lagoon temperature and air temperature) observed at the former (alternative technology) site. A rational basis for this adjustment for somewhat different environmental conditions is the multiple regression model developed for ammonia emissions and measured environmental parameters at the two baseline sites. The model is described in appendix 2 of the three-year progress report. Such a comparison would not require highly uncertain extrapolations of emissions at alternative technology sites beyond the two measurement periods.

Absolute numbers are not used in assessing ammonia emissions from the proposed alternative technology. A normalized measure of emissions (normalized to calculated N-excreted; %EEST) is compared to a similar normalized measure of emissions (%ECONV) from a baseline site using the conventional lagoon technology for handling swine waste in North Carolina. The %E values are an estimate of rate of loss of N compared to N excreted. Two baseline sites are used to account for differences in housing ventilation across the sites with the proposed EST’s. No method exists for adjusting baseline housing emissions to environmental conditions observed at an EST farm. Therefore, actual housing emissions measured at the baseline sites during comparable seasons of the year are used when generating the normalized measures of emissions from houses. It is acknowledged that the housing emissions for the baseline sites were not made under the exact meteorological conditions as the housing measurements for evaluation of an EST. The algorithm followed in deriving an index of performance (%reduction = [(%ECONV - %EEST)/%ECONV] * 100) by the EST in reducing ammonia emissions as compared to the conventional technology currently in use in North Carolina (baseline sites) is presented in Figure 2.12 for water holding structures.

27

Figure 2.12 Algorithm flow chart for evaluation of alternative technology ammonia emission from water holding structures.

Analysis of conventional farm (Moore & Stokes Farm) measurement data for NH3 lagoon emissions with lagoon and air temperature

Establish an observational model based on conventional farm measurement data of NH3 emission and lagoon and air temperature during different seasons

Log10 (NH3 emission) = a + b • Tlagoon+ c • D; a, b and c; experimental constants, Tlagoon; lagoon temperature, Tair; air temperature D; ∆T when ∆T>0, 0 when ∆T<0; ∆T= Tair- Tlagoon

Analysis of each technology farm (EST farms) measurement data for NH3 lagoon emissions with lagoon and air temperature

Estimate conventional NH3

emission for evaluation EST by using lagoon and air temperatures from EST, and multiple linear regression model

Plug temperatures into model

Measured EST average NH3 emission (Fmeas) during the experimental period

Estimated conventional average NH3 emission (Fproj) projected by the regression model

Subtract the measured EST emissions from the projected conventional model emissions, then divide by the projected conventional model emissions and multiply by 100 to find the % reduction

No reduction of NH3 emission

No

Yes Reduction of NH3 emission by alternative technology (The higher the % reduction, the more effective the technology.)

Multiple linear regression analysis

% reduction > 0

28

Evaluation of Red Hill farm (Environmental Technologies) We compare the lagoon NH3-N emission from Red Hill farm with the projected average

emission from lagoon at the conventional farm, using the observational statistical (multiple linear regression) model.

Table 2.10 gives animal weight, feed consumed, and N-excretion at baseline farms and Red

Hill farm. Table 2.11 gives the NH3-N emissions (kg-N/1000 kg-live weight/wk) data summary for the Red Hill farm and baseline farms for evaluation of EST at the former. The emissions from different components of an EST or baseline farm should be viewed relative to the estimated nitrogen excretion from animal population, weight and feed data.

Table 2.10 Summary of animal weight, feed consumed, and N-excretion at conventional farms (Stokes and Moore) and the EST (Red Hill; Environmental technologies) farm.

Farm No. of pigs average pig

weight total pigs

weight feed

consumed N-excretion, E

Information kg/pig kg kg/pig/wk kg-N/wk/

1000kg-lw Stokes (Sep.) 4,392 104.3 458,086 12.84 2.71

Jan. 3,727 88.5 329,840 12.59 2.51 Moore (Oct.) 7,611 52.3 398,055 10.99 4.39

Feb. 5,784 67.0 387,528 12.37 3.90 Red Hill(Mar-Apr.) 2390 69.0 164,910 15.88 5.09

July-Aug 3113 66.5 207,015 15.88 4.69

29

Table 2.11 Estimates of % reduction in NH3-N emissions from different components and their sum total at the EST (Red Hill: Environmental Technologies) and conventional farms (kg-N/wk/1000kg-lw). (% reduction = [(%ECONV- %EEST)/%ECONV]*100 (1) Settling tanks, and Water tank Period

Average lagoon temperature (oC) Average

D (oC)


% ECONV

Red Hill measured emission Fmeas

% EEST

% reduction

Mar 21-Apr 8 , 2005 14.9 0.5 0.32 10.0 0.003

0.06

99.4

July 18- August 5, 2005 31.6 0.0 1.95 54.9 0.001

0.02

99.96

(2) Anaerobic lagoon, Settling tanks, and Water tank Period

Average lagoon temperature (oC) Average

D (oC)


% ECONV


% EEST

% reduction

Mar 21-Apr 8 , 2005 14.9 0.5 0.32 10.0 0.79

15.5

-55.0

July 18- August 5, 2005 31.6 0.0 1.95 54.9 1.37

29.2

46.8

(3) Barn Emissions Period

Stokes Farm measured emission

% ECONV


% EEST

% reduction

Mar 21- Apr 8, 2005

0.25‡

10.0 0.04

0.8

92.0

July 18- August 5, 2005 0.25‡ 10.0 0.74

15.8

-58.0

30

Total Emissions (2)+(3) Period

conventional total emission

% ECONV

Red Hill measured emission

% EEST

% reduction

Mar 21- Apr 8, 2005 0.57 20.0 0.83

16.3

18.5

July 18- August 5, 2005 2.20 64.9 2.11

45.0

30.7

D is ∆T, the difference between the air temperature (Tair) and lagoon temperature (Tlag), when Tair > Tlag ; D = 0 when Tair < Tlag. Fproj is baseline lagoon area adjusted NH3 lagoon emission projected by the baseline multiple linear regression model corresponding to the average lagoon temperature and the average D during Red Hill (Environmental Technologies) farm measurement periods. % ECONV is the conventional model emissions relative to the N excreted. % EEST is the measured emission from the EST relative to the N excreted. Fmeas is the NH3 emissions from water holding structures and NH3 emissions from barn house measured at Red Hill (Environmental Technologies) farm. Soil flux measurements were made, but emissions were found to be negligible. Hog houses at Red Hill (Environmental Technologies) farm are naturally ventilated, like Stokes farm. ‡: Overall house emission measured at Stokes farm during January 2003. % reduction is used to describe how effective a technology is, in reducing NH3 emissions. A number > 0 indicates a reduction in NH3.The larger the % reduction, the more effective the technology is in reducing NH3 emissions. Conversely a number < 0 indicates that there are has been no reduction in NH3 emissions.

31

ADDENDUM Evaluation of Environmentally Superior Technologies for Ammonia Emissions:

Red Hill Farm

Environmental Technologies: Composting Unit Alternative Technology: Composting of Separated Solids Location: Red Hill Farm, (Ayden, NC) Evaluation Periods: August 18, 19, and 22, 2005 Technology contact: Environmental Technologies, LLC – Don Lloyd NCSU Representative PI: Kurt Creamer, NCSU APWMC (919-515-4092) Statement of Task:

- Measurement of ammonia (NH3) emissions from biofilter attached to heated, rotating composting unit using annular denuder technology

- Parameters measured: NH3 flux, temperature of air exiting biofilter Description of Alternative Technology: Separated swine solids from the Environmental Technologies, LLC, “Closed-Loop” swine waste treatment system is mixed with cotton offal and loaded daily into the composting unit via conveyor belt. The composting unit rotates clockwise with the material being rolled toward the exit end of the composting unit. Oxygen is injected into the interior of the composting unit periodically during processing. The estimated retention time is 4 days with temperatures inside the composting unit ranging from 135 to 160 degrees F. Exhaust from the composting unit is passed through a biofilter consisting of pine needles (Figure A3.1). Two by-products result from the composting process – composted solids and concentrated “T” which is considered fluid fertilizer.

(Source: Waste management Program, North Carolina State University, http://www.cals.ncsu.edu:8050/waste_mgt/)

50

http://www.cals.ncsu.edu:8050/waste_mgt/

• A conceptual flow-diagram of alternative technology

Figure A3.1 Conceptual flow diagram of composting unit.

• The only point of emission of ammonia sampled during this evaluation was the exhaust port on the biofilter.

51

Descriptive Information for Composting Unit:

Figure A3.2. Feed hopper, conveyor belt and rotating portion of composting unit.

Figure A3.3 Exit point for composted solids from rotating composting unit unit. Residence time = 4 days.

52

Figure A3.4 Biofilter packed with pine straw. Air flow is from bottom to top of biofilter driven by a fan mounted at outlet of biofilter.

A combination of 50% separated swine solids and 50% cotton offal was the feedstock for this evaluation period. The resulting mixture (~38% dry matter content) was loaded into the feed hopper and then fed into the composting unit via conveyor belt (Fig. A3.2). The loading rate was set at 3 tons (6000 pounds; 2,730 kilograms) per day with a retention time of 4 days. The solid final product exiting the composting unit (Fig. A3.3) had a dry matter content of approximately 35%. The liquid “T” exiting the composting unit had a dry matter content of approximately 0.24%. Once operational, the input rate equaled output rate at 3 tons per day. For this evaluation, therefore, it was assumed that the steady-state load in the composting unit was 3 tons of 50:50 separated swine solids and cotton offal. The N content of the input feedstock and the solid and liquid output from the composting unit were provided by Ms. Lynn Worley-Davis (APWMC, NCSU, Raleigh, NC). The 50:50 separated swine solids and cotton offal had a total N content of 1.99%, with an organic-N content of 1.85%, ammonium content of 0.12% and nitrate content of 0.02%. The solid final product had a total N content of 2.38%, with an organic-N content of 2.31%, ammonium content of 0.05% and nitrate content of 0.02%. The liquid “T” had a total N content of 0.056%, with an organic-N content of 0.007%, ammonium content of 0.049% and nitrate content of <0.001%.

53

• Ammonia Emission Measurements

Emissions of ammonia were measured at the exhaust point of the biofilter (Fig. A3.4). The exhaust pipe was 1.5 inches in diameter which confounded placement of the replicate inlets for the annular denuder technology used to measure the concentration of ammonia in the exiting gas stream. Inlets for the annular denuder technology were not placed directly into the exhaust pipe because of the temperature of the exhaust gas (50˚C) and the physical size of the inlets. Instead, the inlets for the annular denuder technology were clustered from 1.25 to 1.75 inches above the exhaust pipe. The temperature of the exhaust gas at this point was approximately 10˚C lower, but this distance away from the exhaust pipe may have added to the observed variability within the replicates. The individual and mean ammonia concentrations observed over three days are listed in Table A3.1. Table A3.1. Replicate and mean concentrations of ammonia at exhaust point on biofilter.

Time (EST) Replicate Date Start Stop 1 2 3 Mean % CV

- microgram NH3-N per cubic meter - 11:33 12:03 203 229 328 253 26.0 08/18/2005 12;23 13:08 391 447 425 421 6.7 10:32 11:02 323 267 404 331 21.9 08/19/2005 11:32 12:17 239 188 312 246 25.3 11:30 12:00 163 114 157 145 18.3 08/22/2005 12:21 12:51 192 103 161 152 29.6

Grand Mean = 258 41 The biofilter on the composting unit was equipped with a blower fan to draw exhaust gases through the filter. Actual airflow exiting the biofilter was not measured. The rated capacity of the blower fan was used to calculate an upper limit for airflow through the biofilter. This number was then used to calculate the estimated emissions of ammonia from the biofilter per unit time. The rated airflow rate for the fan mounted on the biofilter is 76 CFM (Induced Draft Blower, Grainger # 5C089; Source: Kurt Creamer PI; CFM = cubic feet per minute). This flow is equivalent to 129 cubic meters per hour. Using the grand mean for measured ammonia concentration (Table A3.1), the calculated mean ammonia emissions observed at the exhaust point of the biofilter was 33,300 +/- 13,600 µg NH3-N per hour, or 3,200,000 +/- 1,300,000 µg NH3-N per 4 days, which is equivalent to 3.2 +/- 1.3 grams of NH3-N per 4 days. The annular denuder technology is also capable of detecting oxidizing forms of gaseous nitrogen, specifically nitrous (HONO) and nitric acid (HNO3) in the exhaust gas stream. Neither gaseous nitrogen species was detected being emitted from the biofilter on the composting unit.

54

Evaluation of Ammonia Emissions from Alternative Technology: For this evaluation, the composting unit was considered a closed system with the only exit points for N loaded into the system being the composted solids, the liquid “T”, and the exhaust pipe on the biofilter. Only ammonia emissions from the exhaust point of the biofilter were included in this evaluation, thus an estimate of the efficiency of the composting unit to reduce ammonia emissions was based on a comparison of the amount of N lost as ammonia at the exhaust point of the biofilter to the steady-state mass of N loaded into the composting unit. As noted above, it was assumed that the steady-state load in the composting unit was 3 tons of 50:50 separated swine solids and cotton offal. The mixture of swine solids and cotton offal has a dry matter content of ~38%. This equals a total steady-state dry mass of 2,280 pounds (1,036 kilograms) of the separated swine solids and cotton offal in the composting unit. Assuming a total N content of 1.99% for the separated swine solids and cotton offal, the steady-state load of N into the composting unit is ~20,600 grams or ~ 21 kilograms of N. The measured emissions of ammonia-N were 3.2 +/- 1.3 grams of N per 4 days. Compared to the calculated steady-state load of N in the composting unit, the amount of loss as ammonia-N via the exhaust pipe on the biofilter was found to be 0.02%.

55

II. Appendix A Addendum

Prepared By Wayne P. Robarge To

Black Soldier Fly (BSF) Report Submitted by Kathleen Mottus and Lori Todd

No direct measurements of ammonia flux were made from the Larvae Bed, thus it is not possible to provide a direct measure of nitrogen loss as ammonia. An estimate of the potential loss of N from the Larvae Bed was made through the construction of a N-mass balance for the Larvae Bed (Table A4). Table A4. Nitrogen mass balance for Larvae Bed for 2003.

Category

Manure, Residue Larvae, Prepupae

Mass Nitrogen Content

Nitrogen Mass

INPUTS - grams - - % - - grams -

Manure - Wet 169,000 - - Manure - Dry 67,800 4.35 2,949

BSF Larvae (45,000 count) – Wet (Estimated from Prepupae Data) 31,000* - -

BSF Larvae – Dry (Assume 85% moisture content) 4,650 6.0* 279

BSF Larvae – Dry (Assume 75% moisture content) 7,750 6.0 465

Estimated Range in N Inputs - - 3,228 - 3,414 OUTPUTS

- grams - - % - - grams - BSF Residue - Dry 41,600 2.18˚ 907

BSF Prepupae (37,980 count) – Wet 26,200 - - BSF Prepupae – Dry

(Assume 12% Conversion Rate) 8,136 6.2@ 504 BSF Prepupae – Dry

(Assume 16% Conversion Rate) 10,848 6.2 672 Estimated Range in N Remaining in

Residue and Prepupae - - 1,411 – 1,579Estimated Range in Mass of N Loss Due to Volatilization as Ammonia - - 1,649 – 2,003

* - No data was available for mass of BSF larvae added to manure. As first approximation, wet mass of larvae added was derived from wet mass of prepupae harvested, corrected for population count. Nitrogen content derived from published crude protein content for dried BSF larvae meal; conversion factor 7 x %N = % crude protein. (see text for further details). ˚ - No analytical data available for residue produced in 2003. Used average of 50% nutrient reduction. @ - Based on reported crude protein content of 43.2% by BSF Principal Investigators. On a dry matter basis; conversion factor 7 x %N = % crude protein.

56

Construction of Table A4 was based on information provided by the Principal Investigators for the BSF project, by Ms. Lynn Worley-Davis (NCSU, Animal and Poultry Waste Management Project, Raleigh, NC), and information obtained via the Internet and peer reviewed literature. Inputs of N into the Larvae Bed include fresh hog manure and the original BSF larvae. The mass and N contribution of the BSF larvae was estimated by deriving an estimate of wet mass of larvae from the provided wet mass of harvested prepupae, corrected for assumed moisture contents. It was assumed that the larvae would have a higher moisture contents than the prepupae. The estimate of 6% N content for dried larvae was derived from the published crude protein content of 42% for ground BSF larvae meal, and assuming a conversion factor of 7 for multiplying %N to obtain % crude protein (CP). Bernard et al., (1997) posted a range of CP contents for various fly larvae and other insects. The minimum was 38.9 and the maximum was 60.4 for larvae (Bernard, J.B., M.E. Allen; D.E. Ullrey. 1997. Feeding captive insectivorous animals: Nutritional aspects of insects as food. Nutrition Advisory Group Handbook. Fact Sheet 003, August 1997). The assumption of 42% CP appears appropriate for these calculations. However, the assumed % moisture contents may be too high as Bernard et al., (1997) note the dry matter content of various insect larvae ranged from 20 to 34%. Only the wet mass of harvested BSF prepupae was available from the Principal Investigators. Estimates of the corresponding dry mass of harvested BSF prepupae were obtained by assuming a range of 12 to 16% conversion efficiency for the generation of prepupae mass by consumption of fresh manure. These estimates yield a range of ~0.2 to 0.3 grams per prepupae, which is slightly higher than the range of 0.1 to 0.24 grams per prepupae reported by Sheppard (See Table 1; Craig Sheppard, Black soldier fly and others for value-added manure management. http:// www.virtualcentre.org/ en/enl/vol1n2/article/ibs_conf.pdf). This same author cites a report of up to 24% dry matter conversion of food waste to soldier fly prepupal biomass, and has suggested that swine manure may be a superior medium for BSF production leading to higher observed conversion rates. Thus the range of conversion numbers used in Table A4. May be biased low. The estimate of 6.2% N content for the dried prepupae was derived from the provided crude protein content of 43.2% for dried ground prepupae, and assuming a conversion factor of 7 for multiplying %N to obtain % crude protein. Nitrogen reduction in the residue has been reported by one of the Principal Investigators (Craig Sheppard) as ranging from 25 to 60%. In constructing Table A4 a value of 50% nitrogen reduction in the residue was selected. Only the dry mass of BSF residue produced by the prepupae was available from the Principal Investigators. Nutrient data reported for 2002 trials used different feed stuffs and the determined N contents could not be used in constructing Table A4. Nitrogen reduction in the residue has been reported by one of the Principal Investigators (Craig Sheppard) as ranging from 25 to 70% (See Table 1 in above reference; and http://nespal.cpes.peachnet.edu/sustain/fly.asp). In constructing Table A4 a value of 50% nitrogen reduction in the residue was selected. With the assumptions outlined above, the estimated loss of N from the Larvae Bed via volatilization of ammonia ranges from 1,649 to 2,003 grams of the original N input. Expressed as a percentage, the estimated loss of N as ammonia from the Larvae Bed ranges between 48 to 62% of the original N input. Such relatively high losses of input N via ammonia volatilization are consistent with conditions measured in or above the Larvae Bed. Ammonia volatilization is a function of temperature, and

57

elevated temperatures in the Larvae Bed would favor the emission of ammonia. Temperature data loggers were incorporated into the first section of the Larvae Bed (the section which received fresh manure). Another data logger was suspended 0.6 m from the roof of the structure housing the Larvae Bed. Temperatures recorded for the duration of the First Evaluation period are shown in Fig. A4. Air temperatures within the structure housing the Larvae Bed varied by almost 20 degrees centigrade for each day of the evaluation period, reaching maximum temperatures of over 40 degrees. This extreme range in air temperatures did impact temperatures within the Larvae Bed, but the overall extremes in temperature within the Larvae Bed were much less (typically 5 degrees centigrade). The average temperature within the Larvae Bed was approximately 35 degrees centigrade, which is consistent with a substantial level of biological activity that would be associated both with the activity of the Black Soldier Fly larvae and well as microbial activity. The relatively constant and elevated temperature of the first portion of the Larvae Bed (the portion receiving fresh manure) would favor an enhanced and consistent volatilization of N as ammonia.

Figure A4. Recorded temperatures within the Larvae Bed and within the structure housing the Larvae Bed as a function of Day of the Year for the First Evaluation Period (July 5, 2003 to July 19, 2003). Note: temperatures recorded for the Larvae Bed exceeded maximum operating range of sealed temperature data loggers (> 38C). Temperature range for data logger recording air temperature was not exceeded.

58

Summary The calculated range of between 48 to 62% of N loss as ammonia from the Larvae Bed is consistent with the recorded consistent and elevated temperature within the Larvae Bed throughout the duration of the evaluation period. The relatively large calculated loss of N as ammonia is also consistent with the relatively small amount of N that was captured by the self-harvesting BSF Prepupae: 504 to 672 grams of N. The total mass of N in the self-harvesting BSF Prepupae accounts for only 15 to 21% of the calculated total N inputs.

59

Summary of Ammonia Emissions

The Ammonia science team of Project OPEN has successfully completed its assessment for potential reduction of ammonia emissions as part of the Phase 3 Technology Determinations. These assessments have been accomplished through a combination of field measurements conducted during approximately two-week intensives at each technology (both warm and cool season measurements), and the application of an algorithm for evaluation of alternative technologies whereby ammonia emissions from alternative technologies and baseline (conventional) sites are compared under the same environmental conditions.

Use of a dynamic flow-through chamber is the primary means by which the Ammonia science team directly measures flux from different components (aqueous/soil surfaces) of the alternative technologies. The Ammonia science team constantly strives to improve upon and validate use of the dynamic flow-through chamber system to measure flux. In regards to the overall goals of Project OPEN, the Ammonia science team has completed a comparison of data recorded by use of the dynamic flow-through chamber system to projected ammonia flux as predicted by the U.S. EPA WATER9 Model for the same environmental conditions. The chamber flux measurements of ammonia showed excellent agreement with the U.S. EPA WATER9 Model predictions. For more information on this model the reader is referred to http://www.epa.gov/ttn/chief/software/water/index.html Environmentally Superior Technology performance for ammonia reduction (Phase 3 Technology Determinations). Values shown are % reductions as compared to ammonia emissions from comparable conventional technology sites (positive values indicate reductions in emissions, negative values indicate enhancement of emissions).1

Technology

% Reduction in Emissions from Water

Holding Structures2

% Reduction in Barn Emissions

Total % Emission Reduction

at Technology site3,4

--- Season --- Warm Cool Warm Cool Warm CoolAgriClean (B.R. Harris)5 - 3.6 - -7.5 - -6.3 Environmental Technologies (Red Hill)

46.8 -55.0 -58.0 92.0 30.7 18.5

1 Conventional technology sites included a primary anaerobic lagoon and either tunnel (Moore Brothers farm) or naturally (Stokes farm) ventilated houses. 2 Percent reductions in water holding structures are based against average lagoon ammonia emissions measured at both conventional farm sites for the respective season. Percent reductions in barn emissions are based against the conventional technology using the corresponding housing ventilation technique. 3 Percent emission reduction figures are calculated using a precise algorithm that is documented in the respective reports for each technology. The summary numbers provided in this table should not be averaged or combined in any fashion across components of the technologies or across season. 4 Unless otherwise noted, % reduction in emissions from water holding structures means emissions from all measured structures at a technology were combined together for a single season to arrive at the single % reduction figure. 5 There was no warm season evaluation of the AgriClean Technology.

60

Documents

An Integrated Study of the Emissions of Ammonia, Odor and ... · Rohit Mathur, MCNC, Research Triangle Park Rich Gannon, NC Department of Environment and Natural Resources, Raleigh