Air quality issues on composting horizon

FLUX CHAMBER VS. OPEN PATH LASER TESTING

South Coast Air Quality Management District and the California Integrated Waste Management Board seek solution to controlling emissions from composting and mulching operations while maintaining waste diversion.

THE SOUTH Coast Basin is a 12,000 square mile area in southern California that includes Los Angeles and Orange counties and parts of San Bernardino and Riverside counties. The South Coast Basin is an extreme nonattainment area for ozone and a serious nonattainment area for PM10 (particulate matter less than 10 microns) as determined by U.S. Environmental Protection Agency (USEPA) air quality standards. Composting facilities and related operations handle materials that emit volatile organic compounds (VOC), ammonia (NH3) and dust, which are precursors to the formation of ozone and PM10.

The South Coast Air Quality Management District (SCAQMD), the local air pollution control authority, is responsible for implementing emission reduction requirements to achieve clean air standards and is targeting emissions from composting and cocomposting operations. Originally, in August 2001, the SCAQMD proposed a rule that would require all composting facilities to be enclosed with emissions vented to a biofilter.

Based on further technical assessment and on subsequent input from stakeholders including industry, local government, and the California Integrated Waste Management Board (CIWMB), SCAQMD is now proposing to develop a series of rules, rather than a single rule, to assess and control VOC, NH3, and PM10 emissions from composting facilities and related operations. This series of proposed rules (PR) will include PR 1133, defining registration and reporting requirements for all facilities; PR 1133.1, establishing requirements for chipping and grinding operations; PR 1133.2, establishing requirements for cocomposting facilities; and PR 1133.3, establishing requirements for green waste (yard trimmings) composting facilities. Currently, SCAQMD has issued draft language for three of the rules in the series, PR 1133, PR 1133.1, and PR 1133.2, and has scheduled a board hearing for adoption on December 6, 2002. PR 1133.3 for green waste composting facilities will be addressed in 2003.

This paper discusses emission testing by SCAQMD and CIWMB that has been conducted at composting facilities and related operations associated with the development of the proposed rules. SCAQMD and CIWMB are working together to develop a solution to the multimedia environmental concerns of air quality and waste diversion from landfills. The answer is an emissions control strategy for composting facilities that will result in improved air quality while maintaining responsible management and operation of composting facilities, a solution that protects AB 939 waste diversion mandates in California. (AB 939 mandated all jurisdictions in California to achieve 25 percent diversion of waste from landfills by 1995 and 50 percent diversion by 2000.)

BASELINE EMISSIONS FOR COCOMPOSTING

SCAQMD conducted assessment tests at the source to determine existing baseline emissions of VOC, NH^sub 3^, and PM10 from composting facilities and related operations. Tests were conducted in 1995 and 1996 at three cocomposting facilities - EKO Systems in Corona, Rancho Las Virgenes Municipal Water District in Calabasas, and San Joaquin Composting, Inc. in Lost Hills - that process biosolids along with green waste and other bulking agents. These three facilities are described below:

EKO Systems composts in windrows that are turned one to three times per week using a Scarab. Feedstocks include manure, biosolids, wood chips, agricultural residues, stable waste, and bulking agents such as green waste and other organics. Typical composition is 20 percent biosolids and 80 percent manure by weight. Active composting in windrow piles is typically 50 days. Tests (using flux chambers) were conducted by SCAQMD in November 1995 and January 1996.

Rancho Las Virgenes Municipal Water District's composting facility uses U.S. Filter agitated composting bins enclosed within a large building that is vented to a biofilter for odor and emission control. Typical feedstocks are biosolids and wood chips. There are a total of eight bins and two mechanical agitators, which turn the material approximately once per day. Newly mixed material is added at one end of the bin as material is removed from the building at the other. Temperature controlled forced aeration is provided by blowers underneath the material. The 46-day cycle yields a product that is in compliance with EPA 503 regulations for Class A. Tests were conducted by SCAQMD in December 1995.

San Joaquin Composting, Inc. composts in windrows that are turned using a Scarab one to three times per week. Typical composition of feedstock is 50 percent biosolids and 50 percent green waste by weight. The latter includes yard trimmings from curbside recycling programs and municipal tree and shrub trimmings from the Los Angeles area. Active composting is achieved in windrow piles for typically 57 days. Tests were conducted by SCAQMD in February 1996 and March 1996.

Emission results from the tests conducted at these three cocomposting facilities and reported in the SCAQMD Source Test Reports are summarized in Table 1.

BASELINE EMISSIONS FOR GREEN WASTE COMPOSTING

In its original form, PR 1133 was based on the source tests summarized in Table 1. Since SCAQMD had not measured emissions from green waste composting sites, additional testing was warranted at facilities that produce compost from green waste only and do not handle biosolids. In September 2001 (summer conditions) and November-December 2001 (winter conditions), SCAQMD conducted more source assessment tests to determine existing baseline emissions of VOC, NH^sub 3^, and PM10 from green waste composting facilities.

The tests were conducted at the Inland Composting and Organics Recycling (ICOR) facility in Colton, also known as Inland Empire (the site currently is closed). ICOR occupied approximately 40 acres of land in San Bernardino County and received feedstock from Riverside and San Bernardino counties. The site processed 800 to 900 tons/day - 80 percent curbside green waste and the remainder wood waste. End products included alternative daily cover (ADC) at landfills and various grades of compost. Fresh green waste was received, litter was sorted out, materials were screened and the fines used for ADC. The coarse fraction was ground and composted. Compost was stored in static piles, some measuring nine feet in height and 25 feet in width. Some of the piles were turned with a front loader. The compost was screened on a time schedule with the fines marketed as various grades of composts. Screening overs were typically set aside for further processing. Typical duration of composting was about 45 days, although some materials (e.g., the overs) were composted for about 90 days.

For the September 2001 tests, SCAQMD measured emissions using the USEPA surface emission isolation flux chamber. For the November-December 2001 tests, SCAQMD used an indirect, line-of-sight open path laser technology to measure the concentration of emission gases, NH^sub 3^ and methane (CH^sub 4^), and optical scintillation to measure the flow rate. For the most part, the laser tests were used to develop the emission factors reported by SCAQMD in the Source Assessment Report for ICOR, December 2001 and summarized in Table 2. The lasers can produce a continuous measurement of a single component, such as NH^sub 3^, but cannot accurately measure the combination of several components such as VOCs. Therefore, SCAQMD set one laser to measure NH^sub 3^ and the other laser to measure methane (CH^sub 4^).

The CH^sub 4^: VOC ratio, measured with an isolation flux chamber, was used to estimate VOC emissions based on laser readings of CH^sub 4^. The November-December 2001 testing occurred at ICOR with joint testing being conducted simultaneously by CIWMB using the USEPA surface emission isolation flux chamber.

Emission results from the ICOR testing were reported in SCAQMD Source Test Reports and are summarized in Table 2. CIWMB results are discussed later in this paper.

CIWMB TESTS

As mentioned above, CIWMB also conducted emission tests at mulching and green waste composting facilities in southern California in December 2001 and encouraged SCAQMD's simultaneous participation in the testing efforts. The three green waste processing sites selected were ICOR in Colton and two facilities owned and operated by the City of Los Angeles - the Anchorage composting facility near San Pedro Harbor and the Van Norman mulching facility in the San Fernando Valley. The following are descriptions of the Anchorage and Van

Norman sites:

Anchorage: The Anchorage facility is located in southern Los Angeles County near San Pedro Harbor. It receives materials from the nearby Harbor facility, which takes in approximately 60 to 80 tons/day of mostly residential green waste along with some material from the Department of Parks and Recreation and the Harbor. Feedstocks are preprocessed through a litter picking line, screening, and grinding, then transferred to the nearby Anchorage facility where they are composted in windrows for about 21 days. These materials are used to produce various grades of mulch and compost.

Van Norman: Green waste is processed but not stored at the Van Norman facility (i.e. it is chopped and shipped as mulch). The site is on approximately 20 acres in northem Los Angeles County near the Van Norman Lakes Reservoir. Van Norman processes roughly 300 tons/day of green materials comprised of an estimated 50 percent curbside-collected yard trimmings and 50 percent brush and logs. The raw feed is processed through various sized screens to produce different grades of mulch product.

TEST METHODOLOGY

Tests were performed by Dr. Chuck E. Schmidt, an independent CIWMB consultant, in partnership with the City of Los Angeles. CIWMB developed test protocols prior to conducting the tests, with input from Dr. Schmidt, and technical staff from both the City of Los Angeles and SCAQMD. A consensus was reached on the following protocols:

* Surface area emissions measured using USEPA surface emission isolation flux chamber and USEPA testing standards to evaluate NH^sub 3^, CH^sub 4^, and VOC compounds; Analysis of NH^sub 3^ by impinger sample collection from the flux chamber and NMAM 6015 analytical method;

* Analysis of CH^sub 4^ and VOC as total nonmethane hydrocarbons (TNMHC) by tedlar bag collection samples from the flux chamber and Method 25C;

* Hydrocarbon speciation analysis via grab samples and composite sample collections for analysis by USEPA Method TO-15 for VOC and SCAQMD Method 25.3 for condensable and volatile organic compounds (analyzed by SCAQMD lab);

* Side-by-side testing with the SCAQMD staff at the ICOR facility to ensure common sampling integrity and to develop a correlation between the open path laser technology employed by SCAQMD and the traditional flux chamber sampling techniques employed in the CIWMB tests;

* Testing of an entire 90-day life cycle composting process as well as a chipping and grinding and mulching process as simulated by the Anchorage and Van Norman facilities. Tests were taken on compost that ranged in age from 1-day old to 90-days old, e.g. 1-day, 3-day, 7-day, 14-day, 28-day, 60day, and 90-day. This was done to observe a 90-day emission profile; and

Advective flow measured by recovery of a tracer gas (carbon monoxide) added to the flux chamber sweep air (recovery used to adjust the emissions estimate for volumetric flow into the chamber).

CIWMB tests were done on December 3, 2001 at the Anchorage facility, December 6, 2001 at the ICOR facility, and December 7, 2001 at the Van Norman facility. At ICOR, a total of 14 surface flux measurements were taken on five material types - raw green waste materials, 17-day old static pile, 45-day old compost, 90-day old "overs," and screened product fines. The highest average NH^sub 3^ measurement was from the 17day-old static pile compost; the highest average VOC measurement was from the 90 day-old "overs" compost.

At the Anchorage facility, a total of 18 surface flux measurements were taken at multiple locations on green waste materials and compost/mulch. The highest average measurement from all materials tested on site occurred within the first seven days after grinding. The highest average NH^sub 3^ measurement was from Day 1 mulch; the highest average VOC measurement was from Day 3 mulch. At the Van Norman facility, a total of 24 surface flux measurements were taken at multiple locations on green waste materials and mulch products. The highest average NH^sub 3^ measurement was from Day 3 coarse mulch; the highest average VOC measurement was from Day 1 coarse mulch.

CIWMB TEST RESULTS

The test results, shown as flux measurements from samples taken at all three green waste processing facilities, included a wide range of values that varied by several orders of magnitude. The flux measurements represent 54 data points that depict a full spectrum of facility types, feed compositions, operating conditions, compost ages, product grades, and climatic conditions as related to wind velocities and moisture.

Table 3 provides an overview of the statistical analysis approach. Descriptive statistics are shown for the entire data set, all three facilities combined, and also for the individual facilities. The mean indicates the numerical average of the numbers in each data set; the median indicates the midpoint of each data set (the point at which one-half of the data lies above and one-half of the data lies below); quartiles divide the data into four parts with each sector containing one-quarter of the data; and standard deviation is a measure of data dispersion determined as a square root of the variance. As shown in most of the data sets, a large standard deviation compared to the mean indicates high variability in the data set. The median, not influenced by extreme values and outliers, may be more indicative of the data set than the mean.

The data set provides a representative grouping suitable for statistical analysis to determine universal NH^sub 3^ and VOC emissions from green material composting processes. Statistical analyses were conducted on the entire data set to show normal distribution of the data, data centralities, variability, and potential outlier data. Basic statistical analysis presents the following results: NH^sub 3^ emissions = 0.002 lb/hr1,000ft^sup 2^ mean with standard deviation of +/- 0.006; NH^sub 3^ emissions = 0.000 lb/hr1,000ft^sup 2^ median (Note: 34 out of 54, or 63 percent of the samples were below the detection limit); VOC emissions = 0.19 lb/hr1,000ft^sup 2^ mean with standard deviation of +/- 0.34; and VOC emissions = 0.025 lb/hr1,000ft^sup 2^ median.

SCAQMD AND CIWMB SIMULTANEOUS SAMPLING

The simultaneous tests conducted by SCAQMD and CIWMB at the ICOR facility in December 2001 presented an opportunity to compile two data sets into one larger data set. All of the CIWMB tests were conducted using the USEPA surface emission isolation flux chamber. The results of these tests are shown in Table 3.

During the simultaneous testing, SCAQMD primarily focused on testing using the open path laser technology; however SCAQMD also collected a limited number of samples using the USEPA surface emission isolation flux chamber, SCAQMD used the ratio of CH^sub 4^ to total VOCs, as determined by the SCAQMD Test Method 25.3 results, to predict total VOCs from the CH4 laser readings. Speciation samples were taken and analyzed by USEPA Method TO-15 and SCAQMD Test Method 25.3 to determine the various components of VOCs and their respective concentrations. It is important to note that the speciation samples indicate a high degree of variability for the CH^sub 4^ to TNMHC (VOCs) ratio. The ratio of CH^sub 4^ to total TNMHC (VOCs) varied from 0.19 to 54. Given this variation, it may be difficult to predict VOCs accurately using this method. Additional considerations when using laser technology are the difficulties in assessing if the lasers are capturing the target plume from the study source, or including additive background emissions from adjacent windrows as the wind blows across the pile or windrow that is being tested.

COMBINING SCAQMD AND CIWMB FLUX CHAMBER DATA

The December 2001 tests taken by SCAQMD are the first emission tests that used indirect open path laser technology instead of the more traditional USEPA isolation flux chamber method of testing. Because CIWMB had some concerns regarding the accuracy of the results using the open path laser technology, it completed a data evaluation exercise to combine the two data sets obtained with the USEPA surface emission isolation flux chamber.

For the ICOR facility, CIWMB tests included a total of 14 flux chamber results and SCAQMD tests included a total of four flux chamber results. CIWMB flux chamber samples were analyzed for VOC and NH^sub 3^; SCAQMD flux chamber samples were analyzed for VOC only. CIWMB analyzed for VOC using Test Method 25C. SCAQMD used a different test method to measure VOC emissions, SCAQMD Test Method 25.3, which they prefer because it measures VOC emissions including the condensable fraction that Method 25C does not account for. CIWMB test protocol prescribed Test Method 25C for VOC analyses because there were no laboratories available to run SCAQMD Test Method 25.3 at the time of the tests, except the SCAQMD lab. Typically, a simultaneous sampling event would result in a higher number for VOCs by Method 25.3 (due to the condensable fraction) than would be measured by Method 25C.

To produce combined emission results representing both CIWMB and SCAQMD measurements, a test method correlation was developed that presents all data on the same basis. Based on available simultaneous sampling data, a multiplier of 1.16 can be applied to Method 25C results to provide data results on the same basis as SCAQMD Method 25.3. After the VOC data were normalized to the same test method, SCAQMD Method 25.3, then CIWMB and SCAQMD data were combined and statistical analyses performed on the data sets.

Table 4 summarizes the results. Basic statistical analysis of the data sets indicate the following results: NH^sub 3^ emissions are in the range of 0.002 to 0.007 lb/hr-1,000ft^sup 2^ (mean values); and VOC emissions are in the range of 0.013 to 0.17 lb/hr-1,000ft^sup 2^ (mean values).

Most of the data sets in Table 4 show a standard deviation that is less than or equal to the mean which indicates reasonable grouping of the data within the data set. The exception to this observation is the VOC statistical result for the windrows grouping. The standard deviation is almost two times larger than the mean, which indicates high variability of the data within the data set or a possible outlier that is affecting the statistical calculations. Therefore, for the windrows VOC grouping, the median, which is not as influenced by extreme values and outliers, may be more indicative of the data set than the mean.

FLUX CHAMBER VS. OPEN PATH LASER RESULTS

Using the mean values from the statistical analysis, Table 5 presents a comparison of the combined data (SCAQMD and CIWMB emission results) with the data reported in the SCAQMD Source Assessment Report for the ICOR facility. As shown in Table 5, and Figures 1 and 2, the NH3 emissions are lower for the CIWMB flux chamber data than the NH^sub 3^ emissions reported in the SCAQMD Source Assessment Report. For the most part, the VOC emissions for the combined CIWMB and SCAQMD flux chamber data are also lower than the VOC emissions reported in the SCAQMD Source Assessment Report, with the exception of the windrow tests. However, the windrow tests have limited data points that include a likely outlier since the standard deviation is twice the mean for that data set. Additional baseline emission tests will provide a higher degree of confidence in the data sets and insight into the discrepancy between combined CIWMB and SCAQMD data and SCAQMD Source Assessment Report data.

While SCAQMD and CIWMB tests have looked at baseline emissions, SCAQMD has not set an acceptable emissions level to date. The approach currently being considered is to set a percentage reduction once a control technology has been put in place (e.g. 70 percent reduction in VOC and 65 percent reduction in NH^sub 3^ from the baseline) - versus stating an absolute threshold of "x" tons/day VOC and "x" tons/day NH^sub 3^.

EMISSION CONTROL STRATEGIES, BEST MANAGEMENT PRACTICES

Initially, SCAQMD proposed emission control strategies to implement a 75 percent reduction in NH^sub 3^ and a 90 percent reduction in VOC over existing baseline emissions. The prescribed control strategies defined in the original drafts of PR 1133 for cocomposting and green waste composting facilities specified enclosure of all or a portion of the composting facility with negatively aerated static piles (ASP) and emissions vented through a biofilter. Various scenarios of this emission control strategy were evaluated by SCAQMD along with emission reduction effectiveness, costs, and affordability impacts to the industry. The estimated financial impact of complying with enclosure, ASP, and biofilter control requirements could be devastating to the composting industry that has a minimal profit margin, especially in the case of green waste composting facilities competing with landfill tipping fees for feedstock. SCAQMD acknowledged the financial impacts of emission control strategies in its Technical Assessment Report to its board on April 5, 2002 and is currently considering alternative control strategies in its rulemaking process.

CIWMB and local jurisdictions are concerned about the subsequent impacts on waste diversion in California if the composting industry is negatively impacted by SCAQMD emission reduction requirements. CIWMB has proposed best management practices (BMPs) as a more cost-effective way to reduce emissions from composting facilities. BMPs are operational procedures that achieve optimal composting conditions. They can include simple practices - like controlling feedstocks, pH and moisture, and aeration by turning piles - to more sophisticated systems like negative-ASP with biofilters. CIWMB is planning to work with SCAQMD to conduct additional testing at composting facilities to quantify emission reductions attributable to various BMPs. Testing likely will be conducted this fall in support of SCAQMD's development of PR 1133.2 for cocomposting facilities and PR 1133.3 for green waste composting facilities.

[Sidebar]

The CIWMB testing included an entire 90-day life cycle composting process as well as a chipping and grinding and mulching process.

[Sidebar]

Considerations when using laser technology are the difficulties in assessing if the lasers are capturing the target plume from the study source, or including additive background emissions from adjacent windrows.

[Sidebar]

CIWMB has proposed best management practices as a more cost-effective way to reduce emissions from composting facilities.

[Reference]

REFERENCES

[Reference]

Schmidt, Dr. C.E., Technical Memorandum: Reporting of Ammonia and TNMHC Emission Factors from the Surface Flux Chamber Testing Conducted at the Anchorage, Van Norman, Inland Empire, and LACSD Scholl Canyon Landfill Green Waste Facilities Located in Southern California, May 2002.

Schmidt, Dr. C.E., Technical Memorandum: Results of the Surface Flux Chamber Testing at the Inland Composting and Organic Recycling Facility Located in Colton, California, February 2002.

[Reference]

Schmidt, Dr. C.E., Technical Memorandum: Results of the Surface Flux Chamber Testing at the Anchorage and Van Norman Green Waste Facilities in Southern California, February 2002.

Smyth, Brenda K., California Integrated Waste Management Board, Technical Summary Report: Air Emissions Tests Conducted at Green Material Processing Facilities by California Integrated Waste Management Board, February 2002.

Smyth, Brenda K., California Integrated Waste Management Board, Technical Summary Report: Test Method Correlation and Air Emissions Based on Combined Data for Tests Conducted by California Integrated Waste Management Board and South Coast Air Quality Management District at a Green Material Composting Facility, June 2002.

Stedwick, Wayne A., South Coast Air Quality Management District, Source Test Report

[Reference]

01-171 Conducted at Inland Empire Composting, Ammonia and Volatile Organic Compound (VOC) Emissions from Greenwaste Composting Operations, 2002.

Wang, Mei, South Coast Air Quality Management District, Source Test Report 01-176 Conducted at Inland Empire Composting, Remote Sensing Tests for Ammonia and Volatile Organic Compound (VOC) Emissions from Greenwaste Composting Operations, 2002.

Willoughby, Carey, South Coast Air Quality Management District, Source Test Report 95-0032/96-0003 Conducted at EKO Systems, Characterization of Ammonia, Total Amine, Organic Sulfur Compound, and Total Non-Methane Organic Compound (TGNMOC) Emissions from Composting Operations, May 17, 1996.

Willoughby, Carey, South Coast Air Quality Management District, Source Test Report 95-0034 Conducted at Rancho Las Virgenes Municipal Water District, Characterization of Ammonia, Total Amine, Organic Sulfur Compound, and Total Non-Methane Organic Compound (TGNMOC) Emissions from Composting Operations, May 15, 1996.

Willoughby, Carey, South Coast Air Quality Management District, Source Test Report 960007/96-0008/96-0009 Conducted at San Joaquin Composting, Inc., Characterization of Ammonia, Total Amine, Organic Sulfur Compound, and Total Non-Methane Organic Compound (TGNMOC) Emissions from Composting Operations, November 16, 1996.

[Author Affiliation]

Brenda Smyth is a senior integrated waste management specialist in the Organics and Business Resource Efficiency program at the California Integrated Waste Management Board. Chuck Schmidt is an independent consultant working under contract to CIWMB.