The CIWMB has suggested
that air emissions can be improved, without severely impacting a facilitys
costs, simply by changing some operational practices by adopting best
management practices (BMPs). However, regulatory agencies, like SCAQMD, need
to see the data. Thus, the CIWMB implemented a project to investigate the effect
of management practices on air emissions from green waste composting. Tests
were conducted at a composting facility to measure the ammonia and VOC emissions
for different feedstock mixes and aeration strategies. The results add much
needed insight into the relationship between operational factors and specific
air emissions and pose important consideration of alternative regulatory strategies.
The objectives of the CIWMB tests were to determine whether and how management practices impact air emissions. The tests were not intended to be a rigorous experiment time and funding were both limited. Instead, CIWMB sought a measured and realistic assessment of the potential for reducing air emissions by adjusting management practices.
A large number of factors can potentially affect air emissions, including feedstock composition, temperature, pH, moisture, aeration and pile geometry. Because all of these variables could not be investigated by this project, investigators chose to focus on two factors feedstock composition and aeration while keeping the other variables relatively constant. Two green waste feedstock blends were tested, one with a high percentage of grass clippings (curbside green waste) and the second with a high percentage of woody materials (mostly leaves, brush and wood with some grass clippings). These feedstock blends were characterized by their carbon to nitrogen (C:N) ratios. Similarly two aeration strategies were investigated static windrows and turned windrows.
To test the influence of these factors, full-scale experimental windrows were established at the Tierra Verde Industries green waste composting facility in Irvine, California. The windrows were constructed in late October 2002 and composting continued into February 2003. Four full-scale test windrows were established to evaluate the four variables (Figure 1). Two windrows were constructed with the high C:N ratio feedstock blend (woody) and two contained the low C:N blend (grass-heavy). As measured during the first week, the C:N ratios of the grass-prominent blends ranged from 16 to 28 and averaged 18.5 and 26 for the static and turned windrows, respectively. In comparison, the C:N ratio of the woody blends ranged between 51 and 100, with averages of 54 and 74 for the static and turned windrows. Two of the windrows were static while two were turned regularly. The static windrows were not turned at all throughout the 102-day composting cycle. The turned windrows were turned with a Scarab machine when windrow temperatures were high enough to maintain 131°F even after the cooling effects of turning. This turning regime resulted in about three turnings per week.
Air emissions were measured during the first and second weeks of composting (days 3 and 4, 11 and 12) by Charles Schmidt, an independent technical consultant for CIWMB. Schmidt captured samples for ammonia and VOC analysis using the USEPA isolation flux chamber method. Emissions samples were collected from six locations on each windrow, including spots along the ridge and the sides of the windrows. For the turned windrows, samples were taken on the day that the windrows were turned but at least 30 minutes after the turning was completed (to allow the VOC emissions from the freshly turned material to subside to a steady rate). During the first two weeks and at the end of the test, four grab samples of material were collected from each windrow at evenly spaced locations to characterize the feedstocks and compost quality. These samples were analyzed for C:N ratio, bulk density, moisture and volatile solids. Also, the compost samples, collected on days 101 and 102, were analyzed using the Solvita Maturity Index to determine the relative completion of the composting process.
Perhaps the most significant finding from the tests is that ammonia does not appear to be a concern with green waste composting. Of the 48 samples analyzed for ammonia emissions, only one was higher than the analytical detection limit of 0.1 ug/ml. Ammonia emission levels were low even from the windrows with a high proportion of grass, i.e. the low C:N feedstock blends. Given that ammonia is not an issue, the main consideration becomes how do BMPs impact VOC emissions.
The test results generally suggest that increasing the C:N ratio decreases VOC emissions. Figure 2 shows the average VOC emissions for both sampling periods plotted against the average C:N ratios from the high C:N (woody) windrows and the low C:N (grassy) windrows (63 and 20, respectively). The line slopes downward, indicating a prevailing trend of decreasing VOC with increasing C:N ratio. This trend held for all but one situation. Lower VOC emissions were detected for the high C:N windrow when comparing static windrows during the first and second weeks of composting and when comparing turned windrows during the first week. However, for the second weeks measurements, greater VOC emissions were measured from the high C:N windrow that was turned compared to the lower C:N windrow that was turned. There is no apparent reason why the turned windrow dodged the general trend in the second week. The result may simply be an experimental anomaly because all of the VOC data was highly variable.
Relative to feedstock composition, turning had a more pronounced effect on VOC emissions, at least during the first and second weeks of composting. In both weeks, the VOC measurements from the turned windrows were much greater than the VOC emissions measured from the static windrows (Figure 3). For both static and turned windrows, the VOC emissions were higher during the first week than the second. However, the reduction in VOC from the first to second week was more dramatic in the case of the turned windrows. Thus, the VOC emission levels of the static and turned windrows grew closer together with time.
It is easy to speculate why VOC levels were higher in the turned windrows. Turning carries raw feedstocks from the core of the windrow towards the outer layer where VOC compounds can more easily volatilize and escape. In contrast, static piles tend to develop a relatively stable outer layer that filters some of the VOCs that migrate from the center of the windrow. In addition, turning accelerates the rate of composting, which in turn leads to faster VOC formation and volatilization. The process-enhancing effect of turning was confirmed by the compost maturity evaluations conducted during the CIWMB tests. The average Solvita Maturity Index for the compost samples from static windrows, collected on day 101, was 5.9, which indicates a compost that is approaching the curing stage. The turned windrow compost samples, gathered on day 102, had an average Solvita Maturity Index of 6.6, which is indicative of a nearly finished compost. The composts from static windrows needed more time to reach the same level of maturity as the composts from the turned windrows.
In the CIWMB tests, VOC emissions were not measured beyond the first two weeks of composting. In that time, the turned windrows emitted more VOC. However, it is possible, even likely, that the rate of VOC emissions from the static windrows eventually exceeded those from the turned windrows. Because the decomposition rate in the turned windrows advanced faster, the VOC release fell off faster. The static windrows, on the other hand, released VOC at a slow pace, with a gradual decline. It is reasonable to conceptualize a hypothetical VOC emissions pattern for both aeration approaches, such as the curves shown in Figure 4. Initially, the rate of VOC release from turned windrows is much greater but it declines quickly relative to the slow and steady emissions from static windrows. What matters most is the total amount (e.g. kgs.) of VOC emitted for the entire duration of the composting cycle (represented by the area under the curves in Figure 4). Therefore, whether turned windrows or static windrows emit more VOC cannot be determined unless emissions are measured regularly over the life of the composting process. To make a fair comparison, the measurements should continue until the composts from both management systems reach a similar level of maturity.
In general, the results of these tests should be considered indicative rather than conclusive because of the variability of the data and the conditions of the tests. For example, the air emissions were collected from various locations of the windrows top, side and base. As shown in Figure 5, the flux, or rate of emissions, varied greatly among these sections of the pile, and also from one spot to another. The data generally confirmed that the flux was greatest at the top of the windrow and lowest near the base but again, this trend was not consistent. The air emission flow through a passively aerated windrow (as both static and turned windrows are) follows channels and vents within the mass of material. Finding the locations where the major emissions occur is a hit and miss proposition.
TAKE HOME MESSAGES
The CIWMB test results provide several interesting insights. First, ammonia appears to be a non-issue with composting of typical green waste feedstocks. Ammonia is generally undetectable, even during the first weeks of composting. This may be due to the biochemical nature of green waste feedstocks or the moderate to high C:N ratio. However, the CIWMB tests found little ammonia emitted even when the average C:N ratio was less than 20. In any case, ammonia should not be a regulatory concern for green waste composting.
Second, simple management practices can affect the rate, and probably the amount, of VOC emitted during green waste composting. The tests conducted here suggest that VOC emissions decrease with higher C:N ratios and increase with turning, at least in the early stages of composting. In the case of turned windrows, emissions dropped off dramatically from the first to the second week while the decrease was less severe with static windrows. With its slower pace of decomposition, static windrow composting is likely to surpass turned windrows in VOC emissions after several weeks. It is uncertain from the data collected by CIWMB which approach would be expected to yield a lesser total amount of VOC over the entire course of composting.
While the ammonia evidence is convincing, only general conclusions can be made about VOC emissions from the CIWMB test results because the scope of the tests and the data collected were limited. Due to the wide range of operating variables in the composting process and the cost of collecting emission data, the tests were intended to provide a glimpse, rather than a full picture, of the potential for BMPs to control air emissions. Having glimpsed the situation, CIWMB is moving in for a closer look. CIWMB has contracted with San Diego State University to conduct more comprehensive and controlled experiments in order to establish baseline data on ammonia and VOC emissions from green waste decomposition and to investigate the effect of multiple management factors through the full composting cycle (see sidebar). Hopefully, the information generated will add weight to the argument that management, rather than enclosures, can allow the composting industry to contribute to improved air quality in their home regions.
Brenda Smyth is Senior Integrated Waste Management Specialist with the Organics and Business Resource Efficiency program at the California Integrated Waste Management Board. The full report of the CIWMB air emissions BMP tests is available from firstname.lastname@example.org.
BMPS TO MINIMIZE AIR EMISSIONS
The California Integrated Waste Management Board (CIWMB) funded research that is underway at San Diego State University to investigate the baseline VOC and ammonia emissions resulting from natural decay of selected green waste composting feedstock materials. In addition, the research team is searching for the best management practices (BMP) to minimize the emissions. After surveying the composting facilities located in California for the type and the amount of feedstock they compost wood chips, ground prunings and leaves, and grass clippings were selected for the BMP investigation. The effects of several operational parameters including C:N ratio, moisture content, aeration intensity, pH level, and chemical additions on the VOCs and ammonia emissions are being studied at laboratory-scale. A number of flux chambers are being used to collect the VOCs and ammonia emissions as the feedstock undergoes natural decay. In this case, owing to the simplicity of the experimental setup, it is possible to include two additional feedstock materials food waste and leaves into the investigation. Preliminary findings suggest that a portion of the emissions can be attributed to the natural, "biogenic" emissions. For further information, contact Fatih Büyüksönmez (email@example.com). A presentation on the research will be given at the BioCycle West Coast Conference 2004 in Portland, Oregon.