Cedar Grove Composting Odor

7/29/2019 Cedar Grove Composting Odor

1/42

{00235104.DOC /3}

Odor Impact Evaluation forCedar Grove CompostingsEverett Facility

Prepared for

Cedar Grove Composting

3620 36th Place NEEverett, WA 98205

Submitted to

City of Everett

June 2010


2/42

{00235104.DOC /3}- 2 -

1. Introduction

Cedar Grove Composting, Inc. (Cedar Grove) requested that CH2M HILL evaluate the odorimpacts of its Everett facility on the surrounding community. This report considers CedarGroves current operations in Everett and also evaluates the potential impacts from changes thatCedar Grove is considering to its Everett operation, such as the addition of anaerobic digesters toits operations.

To put the technical analysis in this report in context, a historical overview of Cedar Grovesoperations in Everett since the facility opened in 2003, actions taken by the Puget Sound CleanAir Agency (PSCAA) in response to odor complaints regarding the facility and Cedar Grovesresponses to PSCAAs actions, including a recently-concluded Agreed Order on Consent underwhich Cedar Grove committed to spend more than $1 million on odor mitigation.

2. Executive Summary

Cedar Grove opened its composting facility in Everett in 2003 and operated that facility from2003 until 2008 without any odor complaints. In 2008, a combination of operational issues,natural events and heightened sensitivity based on the distribution of a post-card inviting odorcomplaints resulted in a large number of those complaints being filed.

Cedar Grove has made a substantial commitment to odor mitigation at its Everett facility,spending more than $1 million on odor mitigation, as well as adopting operational changes to

minimize odors. This report validates that those efforts have had a significant impact onminimizing odors from the facility. Modeling conducted by CH2M Hill, Envirometrics andPSCAA demonstrates that the vast majority of odors at the facility would be diluted to belowdetection levels before reaching the closest residence. In addition, reviewing the complaints thathave been received during part of 2008 and all 2009 indicates that a large percentage of thosecomplaints were either meteorologically improbable or were generated by an odor source otherthan Cedar Grove.

Cedar Grove is proposing the introduction of anaerobic digesters to its existing compostingoperations. Diverting a significant portion of incoming feedstock to those digesters will have abeneficial effect on odors. This is because the feedstock diverted to the digesters will be kept in

enclosed spaces and any gases generated by the digestion will be captured and used for either theproduction of electricity or as compressed natural gas.

Cedar Groves investment in odor mitigation and the introduction of anaerobic digestion has thepotential to reduce the odor impacts of the facility to well below the level of significance.


3/42

{00235104.DOC /3} - 3 -

3. Cedar Groves Existing Operations

3.1 Cedar Groves Operations in Everett

The Everett site occupies slightly less than 30 acres of a 128-acre parcel owned by Cedar Grove.1A layout of the site is provided on Figure 1.

Cedar Groves Everett plant consists of a receiving (tipping) building, grinder, conveyors, fourGore Cover Compost pads, finished product screening, material storage, leachate collectionand treatment, and stormwater treatment.

Cedar Groves existing operations in Everett were initially permitted in 2003. Two GoreCover Compost pads were installed in 2004. In 2005 two additional Gore Compost pads werecompleted and were placed into operation in the summer of 2005. Operations remained

essentially the same until winter 2008, when the initial stage of the composting process waschanged to reduce Phase I from 28 days to 21 days.

Figure 1 shows both the current configuration of site, as well as the location of the plannedanaerobic digester system.

Figure 1

1 A 20-acre parcel in the southeast corner of the property was sold to Concrete Norwest in 2005 and is partiallydeveloped as a ready-mix batch plant.


4/42

{00235104.DOC /3} - 4 -

3.2 Cedar Groves Current Composting Process

In order to understand the odor issues associated with Cedar Groves Everett facility, anoverview of the composting process used by Cedar Grove is helpful. The following pictures alsoprovide a point of reference as to where various parts of the process take place.

TippingTipping

buildingbuilding

ScalesScales

GrinderGrinder

ConveyorsConveyors

GoreCoveredAerobicGoreCoveredAerobic

ProcessProcess

Phase1Phase1

Phase2Phase2

Phase3Phase3

StormwaterStormwater

treatmenttreatmentLeachateLeachate

tanktank

Screening,finishedScreening,finishedproductproduct

storage/distributionstorage/distribution

CurrentCedarGroveProcessbasicelements

Phase1Phase1Phase2Phase2

Phase3Phase3

FinishedproductFinishedproduct

IncomingmaterialIncomingmaterial

TippingTipping

buildingbuilding

GrinderGrinder

CurrentCedarGroveProcessFlow


5/42

{00235104.DOC /3} - 5 -

3.2.1 Incoming Feedstock

Incoming feedstock is supplied by commercial waste collection companies and municipalities inSnohomish and King Counties Monday through Friday. Wastes collected in nearby areas aredelivered directly by the route trucks. In areas further away, e.g., south Snohomish County,

Seattle and King County, route trucks take the collected organic wastes to a transfer station or re-load facility. Wastes from several trucks are loaded into long-haul vans and delivered to theCedar Grove facility. Long haul trucks from transfer stations mostly arrive from about 7 am toabout 4 pm. Route trucks from local areas arrive from about 9 am to 11 am and 2 pm to about 4pm.

3.2.2 Mixing and Grinding

Waste is mixed in the tipping building to achieve the proper mix of carbon (woody) and nitrogen(food, leafy). A proper balance between carbon and nitrogen is a vital part of the compostingprocess. Waste from the tipping building is removed from the building by a front-end loader and

placed directly onto a conveyor that leads to a grinder. The conveyor contains a screen thatallows the smaller particles to bypass the grinder while the oversize material continues across thescreen and into a grinder. Grinding decreases the size of individual pieces of feedstock so thatbacteria can act on it efficiently, as well as providing texture and porosity to allow air movement.

Most wastes are processed and ground within a couple of hours of being received, meaning thatmost materials are processed the day they are received.

An enclosed conveyor moves material from the grinder to the middle of the Phase I cells of theGore Cover system. As needed, moisture in the form of leachate (collected from under theGore Cover pads and treated in a tank), stormwater or freshwater is sprayed onto the materialas it is moved on the conveyor.

3.2.3 Constructing Compost Heaps and

Moving Heaps in Phases I and II

A front-end loader collects compostable waste at a drop at the end of a conveyor and moves it toa previously cleaned cell location. The Gore Cover has been rolled up on a winder rig thatlocated at the head of the new row. The heap is built one bucket load at a time. Phase I heapsare built while the grinder is operating, from about 6 am to about 1 pm. Phase I heaps are takenapart and the Phase II heaps are built from about 3 pm to about midnight.

After a heap is constructed, the Gore Cover is pulled over the pile. This cover is a speciallydeveloped Gore-Tex type membrane laminated between two polyester layers. The cover protectsthe pile from weather conditions, while still allowing release of CO2 and moisture. Thesecontrolled conditions allow consistent product to be produced without the risk of damp pockets,which could create anaerobic conditions.

The Gore Cover controls odor emissions in three ways: (1) prevention of anaerobic pockets,(2) creating a physical barrier to the emission of gas from the decomposing pile, and (3) bycontrolling condensation on the interior of the Cover. During the composting process, a fine film


6/42

{00235104.DOC /3} - 6 -

of condensation collects on the inside of the cover and then drips back onto the pile. Themoisture helps to keep odorous compounds in solution and prevents them from volatilizing intothe air phase..

The cover is rolled over the constructed pile as soon as the heap is complete. Once the cover isin place, temperature and oxygen probes are installed and the edges of the cover are sealed. The

air supply is then controlled by the oxygen and temperature sensors. Airflow to the heap iscycled on and off to maintain the oxygen and temperature levels within the desired parameters.

Once the heap has remained in place for its required period of composting, the Gore Covercover is rolled up and the front-end loader digs out the row and transfers it to a conveyor thatwill carry the partially composted waste to the next section of the facility.

3.2.4 Construction of Phase III Heaps

Construction and disassembly of the Phase III heaps proceeds in the same way as for the Phase Iand Phase II, except that a Gore Cover is not utilized. Because the material has already gone

through Phase I and II, most of the odorous compounds in the material have been broken downby the anaerobic composting process. An active air supply is provided to each heap during itscomposting period.

3.2.5 Screening and Storage of Finished

Product

In the final stage, finished compost is screened by size for product uses (compost, mulch, for mixwith soil blends, etc.) Oversized material is returned to the front end of the compost process andblended with unprocessed material at the grinder. This helps inoculate the incoming material

and accelerate the onset of the composting process.


7/42

{00235104.DOC /3} - 7 -

4. Cedar Groves Response to Past Odor

Issues in Everett

4.1 Odor Complaints Generated in 2008 Based on a

Combination of Natural Factors, Process Problems

and Solicitation

By June 2008, Cedar Groves Everett facility had been operating for over four years withoutreceiving a single odor complaint. During that time period the facility had composted over432,000 tons of material.

In June and July of 2008, after a late spring and rainfall of over 45 inches, the facility received

large volumes of wet grass. At the same time, the slough adjacent to Cedar Groves facilityexperienced extremely low, negative tides. The aeration system for the leachate tank becameplugged, causing the leachate system to become anaerobic. The addition of anaerobic leachatewater to the grinder contributed to the odors at the facility.

Incoming feedstock had been held in one of the transfer stations that supplies Cedar Grove for aperiod of between three to five weeks. When this material was initially received, odors from thismaterial were released. Combined with the anaerobic leachate, this added to the potential forodors.

A final factor contributing to odor complaints during the summer of 2008 was the distribution of

a post-card identifying Cedar Grove as a source of odors and suggesting that odors be reported tothe Puget Sound Clean Air Agency. Cedar Grove is not aware of the source of the card;inquiries to PSCAA indicated that they were not involved in the generation of the card. A copyof the post-card has been included as Appendix A to this Report for reference.

The combination of all of these factors resulted in a large number of odor complaints being filedwith PSCAA. After complaints began to be made, theEverett Heraldran an article implyingCedar Grove was solely responsible for odors in the area and local television stations had similarcoverage. In 2009 the City of Marysville printed an announcement in its Parks and Recreationguide suggesting that if people experienced odors like those from Cedar Grove they should callPSCAA. TheHeraldhas had subsequent articles with headlines referring to Cedar Groveodor. These communications have increased the surrounding communitys perception thatCedar Grove is the primary source of odors in the area adjacent to the composting facility. Asexplained below, the evidence shows that this perception was unfounded. There are a number ofother strong odor sources present in the same area, whose odors have been mistakenly attributedto Cedar Grove.


8/42

{00235104.DOC /3} - 8 -

4.2 Cedar Groves Response to Odor Complaints and

Investment in Odor Mitigation

In response to odor complaints in the summer of 2008, Cedar Grove began investigating multipleoptions for odor mitigation. This investigation included evaluation of different components ofthe composting process and an assessment of how odors could be generated by different parts ofthe facilities. In addition, Cedar Grove looked at a variety of operational changes and equipmentto prevent odors from being transmitted off-site.

Cedar Grove developed an Odor Mitigation Plan which was presented to and approved byPSCAA. The approved Odor Mitigation Plan included both operational changes and capitalimprovements agreed to by both Cedar Grove and PSCAA and integrated into an Agreed Orderof Consent between Cedar Grove and PSCAA. The elements of the approved Odor MitigationPlan are outlined below:

Voluntary Odor Mitigation Measures Implemented by Cedar Grove

Cedar Grove voluntarily instituted a number of odor mitigation measures. The cost of thesemitigation measures, as estimated by Cedar Grove, is included for reference purposes.

x Odor Studies and Development of Mitigation OptionsEstimated Cost: $100,000

Cedar Grove engaged both CH2M Hill and Envirometrics, Inc. to conduct odor evaluations atCedar Groves Smith Island Facility, and these consultants developed and evaluated themitigation options included in the Odor Mitigation Plan.

x Diversion of Excess FeedstockEstimated Cost: $500,000 (includes transportation costs, processing fees and lost

revenue from diverted feedstock)

Arrangements were made with several other permitted composting facilities to enable CedarGrove to handle peak volumes of incoming feedstock. Those facilities are:

Green Earth Technology Whatcom County

Bailey Farms Snohomish County

Lenz Enterprises Snohomish County

Cedar Grove Composting, Maple Valley King County

Diversion of excess material is accomplished by sending feedstock directly to these alternativefacilities. If materials have already been tipped at the tipping area, materials may be reloadedand transported out of the facility.


9/42

{00235104.DOC /3} - 9 -

x Hired Additional Personnel Devoted to Odor MitigationTotal Estimated Cost: $110,000

A new staff member was hired and given responsibility for monitoring incoming feedstockspecifically to identify excessive odors. Estimated Cost: $55,000

Cedar Grove also hired an Odor Specialist whose full-time responsibility is monitoring odorcomplaints in the surrounding community and investigating and documenting odors from theSmith Island Facility. Estimated Cost: $55,000

x Voluntary Operational Changes at the Smith Island Facility

Cedar Grove made the following modifications to its operations at the Smith Island Facility in aneffort to mitigate potential odors from that Facility.

Tipping Building

When incoming feedstock has already begun to generate odors, it is mixed in with other wastealready stored in the tipping building, allowing any odors to be addressed by the tippingbuildings aeration system and biofilter. In addition, cleanliness in the tipping building andtipping building apron is maintained by water spray and sweeping.

Leachate System

Total Estimated Cost: $2,000

Added a daily inspection of the leachate tank to confirm that the air diffusers inside the tank areworking properly. In addition, the leachate collection and handling systems are cleaned daily.

Grinding Operations

Estimated Cost: $ 15,000

Addition of misting system with odor control agent to reduce dust and release of odors. Fugitivedust control is accomplished by frequent watering and sweeping of the site.

Aeration of the stormwater pond

Estimated Cost: $29,500

The aeration system for the stormwater detention pond is checked daily to make sure the pondstays in an aerobic state.

Transfer operations and uncovered Phase 1 and 2 heaps


In the event it is necessary to stop work on any pile, the pile will be covered as soon aspracticable and not later than the end of the shift.


10/42

{00235104.DOC /3} - 10 -

x Voluntary Capital Improvements Made for Purposes of Odor Mitigation

Tipping Building


An apron was constructed for the tipping building and rollup doors were installed in the front ofthe Tipping Building to minimize the escape of air from the face of the building.

Conveyors and Trommels


Shrouds were installed on the trommel.

Installation of a tree barrier/wind break


Cedar Grove designed and installed a windbreak around the Smith Island Facility. Installation ofthe windbreak was done on a berm using both conifers and deciduous trees. The plantedwindbreak included mature trees between 15-20 feet tall, along with bare root shrubs.

Installation of an on-site weather station and use of data frommultiple weather stations


To allow the Odor Specialists to correlate odor complaints with the meteorological conditions atthe time of the complaint, Cedar Grove installed a weather station on-site and entered intoarrangements allowing it to use weather data from other stations so that it can obtain the mostaccurate weather data (temperature, wind speed and wind direction) at its Everett facility andfrom the surrounding area.

The remainder of this report focuses on subjective odor perception; specific odor issuescommonly associated with composting generally, an analysis of the particular complaintsreceived in 2009, an examination of the proposed modifications to Cedar Groves compostingprocess under consideration and the potential those changes have to reduce odors at the facility.

5. Perception and Measurement of Odors

5.1 How Odors are Perceived

An odor is a sensation our brains generate in response to chemicals in the air which arebreathed in through the nose (Dalton, 2003b). Odors are generated by airborne environmentalchemicals odorants, which react with receptors in the nose and produce the odor sensation inthe brain. Humans can perceive most odors even at relatively low concentrations of thechemicals, in parts per billion in many cases (Dalton, 2003b).


11/42

{00235104.DOC /3} - 11 -

Odors are subjective and humans differ in their sensitivity to odorous compounds in the air.Where some people detect an odor to be strong, others sense it as weak or even non-existent(Schiffman, 1998). In addition, people differ in their reactions, preferences and aversions toodors. An objectionable odor to one person can be tolerable to another. The perception of odordepends on both the characteristics and concentration of the odorant and the person smelling it.

5.2 How Odors are Measured

Because odors are perceptions of numerous chemical compounds, detecting, measuring,monitoring and even describing odors are not straightforward processes. The human elementcannot be completely removed from the process because odors are perceived subjectively at verylow odorant concentrations. Odor science has developed special terms and techniques todescribe and measure the nature and strength of specific odors.

2Both qualitative and

quantitative descriptors are necessary an odor may be measured according its thresholdconcentrations, pervasiveness, descriptive quality, degree of pleasant or unpleasantness and theconcentration of the odor-causing chemicals present.

5.2.1 Odor Thresholds

The most common measure of odors is the odor threshold value (OTV). Odor strength isquantified by determining the amount of dilution needed to bring the odorous air sample to itsthreshold of perception. The higher the threshold value, the more dilution necessary to bring theodor to threshold, and thus, the stronger the odor.

The odor threshold is determined by trained human assessors observing presentations of the

odorous air sample dynamically diluted with an olfactometer.3 Because individuals perceiveodors differently, odor panels contain several members, preferably 5 to 10 (Haug, 1993). Their

collective response is expressed statistically. For example, odor thresholds are usually definedby the point at which 50 percent of the panel no longer detects the odor (e.g. D/T50). Results arecomputed for each assessor based on the dilution levels where correct detection orrecognition responses are recorded. The responses of all assessors are averaged to determinethe samples detection and recognition thresholds.

The dilution ratio is an estimate of the number of dilutions needed to make the actual odoremission just detectable. This is known as the Detection Threshold (DT). The RecognitionThreshold (RT) is the dilution ratio at which the assessor first detects the odors character(smells like).

2More detailed discussions of odor detection and measurement are provided by St. Croix Sensory, Inc., 2005 andBrant and Elliot, 2004).

3 The testing procedures follow ASTM International E679-04, Standard Practice for Determination of Odor andTaste Thresholds by a Forced-Choice Ascending Concentration Series Method of Limits, and EN13725:2003, AirQuality Determination of Odour Concentration by Dynamic Olfactometry. EN13725, the official standard of allEuropean Union countries, exceeds the requirements of ASTM E679-04. The standardization organizations ofAustralia and New Zealand have also adopted an identical standard (AS/NZ 4323.3-2001).


12/42

{00235104.DOC /3} - 12 -

The odor threshold is reported as a dimensionless dilution ratio; however, often the pseudo-dimensions of Odor Units (O.U.) are used. Units of Odor Units per cubic meter (O.U./m

3)

are also commonly applied in order to calculate odor emission rates.

5.2.2 Odor Intensity

Odor intensity is the relative strength of the odor above the Recognition Threshold.4 The odorreferencing is accomplished by comparing odor intensity of the odor sample to the odor intensityof a series of concentrations of the reference odorant nbutanol. The Odor Intensity ReferencingScale serves as a standard method to quantify the intensity of odors for documentation andcomparison purposes.

5.2.3 Hedonic Tone

The hedonic tone is a rating of the pleasantness or unpleasantness of an odor. It is a subjectivemeasure, but it can be averaged for a group of individuals evaluating an odor. Hedonic tone can

be rated on a scale from 10 for very unpleasant to +10 for pleasant (Chiumenti et al., 2005).Hedonic tone depends on the odor intensity, concentration, duration and frequency of exposure,as well as the perceptions and associations of the individual.

5.3 Odors and Health Effects

The following was taken from the San Diego University Report; With the possible exception ofenclosed spaces and vessels, the chemical compounds responsible for odors at a compostingfacility are not present in large enough concentrations to cause direct physiological harm(Dalton, 2003b; Pelossi, 2003). That is, the concentrations are below the threshold levels thatcause irritation via the trigeminal system. Yet, people still claim to experience symptoms of ill

health due to odors. The symptoms most often reported include nausea, eye, nose and throatirritation, headache, shortness of breath and drowsiness (Chiumenti et al., 2005; Schiffman,1998). Without a cause-and-effect relationship, the question that follows is: Are the symptomsreal (i.e. physiological) or perceived (i.e. psychological)?

The answers offered by researchers appear to suggest that symptoms arise from peoples reactionto odors, rather than a direct physiological effect. In 1998, a conference of experts at DukeUniversity examined the potential health effects of odors from livestock farms, wastewatertreatment and biosolids recycling (Schiffman et al., 1998). The participants identified thefollowing three ways in which odors might cause health symptoms:

1. A person is exposed to an odorous chemical compound at concentrations highenough to produce sensory irritations. The person associates the symptoms with

4 The intensity of an odor is referenced on the ASTM Odor Referencing Scale described in ASTM E544-99,Standard Practice for Referencing Suprathreshold Odor Intensity. The IITRI Dynamic Dilution BinaryOlfactometer (Butanol Wheel) is the dynamic presentation method St. Croix Sensory utilizes for the procedure ofodor intensity referencing.


13/42

{00235104.DOC /3} - 13 -

the odor, although they are actually caused by the chemical odorant. Asmentioned above, odorous compounds are rarely present in high enoughconcentrations to cause direct irritation. Therefore, this situation is not prevalent.

2. A second agent may accompany the odor, such as pathogenic bacteria. Theperson becomes ill from exposure to this second agent but associates the

symptoms with the odor. Although this scenario cannot be easily ruled out, thereis little or no evidence that suggests it occurs.

3. Odors at concentrations high enough to be detected but below the irritantthreshold affect people in other ways. They change their behavior or thoughts inways that lead them to experience the symptoms. (For example, when exposed toan odor, a person may alter their breathing by holding their breath, taking shallowbreaths or breathing only through their mouths. Hence, they may become light-headed or develop a sore throat (Dalton, 2003b)). Persistent or frequent odorsmay increase stress, depress a persons mood or cause them to negatively changetheir behavior in ways that bring on physical symptoms. The conference

participants concluded that more research is needed to assess and betterunderstand the associations between these biological andbehavioral/psychosocial factors (Schiffman, 2000).

Assuming that the third scenario is a reasonable explanation of the situation, the odorsthemselves are not the direct cause of the reported health effects.

6. Odors Generally Associated with

Composting

There are four potential odor sources at an active green waste/food waste composting facility.First, odors may be present in the material received for processing. Second, odors may developfrom the compostable materials while they are on-site. Third, compostable materials may bedeposited on the ground around the site and develop odors when it becomes wet and decays.Fourth, leachate and storm water on the site may accumulate in puddles, ponds or tanks anddevelop odors as the nutrients in the liquid decay.

The enzymes and bacteria that occur naturally in plant material begin a rotting process almost assoon as the plant is cut or uprooted. These bacteria and enzymes reduce the naturally occurringamines to odorous amines and ammonia. Other bacteria, especially those that prefer to grow in

the absence of oxygen, produce hydrogen sulfide, dimethyldisulfide and methyl mercaptan fromthe naturally occurring amino acids, producing a sewer odor. Wet grass matted in a closedcontainer for several days is an example of such anaerobic decay odors. If there are any meats,fats or oils in the waste, oxidation of the fats and fatty acids and produces a rancid odor. Theodors most commonly associated with plant material decay are hydrogen sulfide, reduced sulfurcompounds, and ammonia. When food waste is present there can be various odorous fatty acids,ketones and aldehydes released. Because food wastes are generally only a very small portion ofresidential wastes, food waste has not been a significant source of odors at composting facilities.


14/42

{00235104.DOC /3} - 14 -

According to San Diego State Universitys Comprehensive Compost Odor Response Project,dated March 2007, the odor associated with composting is not the result of a single smell.Organic materials inherently generate a large number and variety of volatile chemicalcompounds that humans can sense as odors both good and bad. These compounds can coexistand interact with one another to produce a diverse menu of aromas.

As an organic material decomposes, the mix of volatile compounds changes, and so does thecharacteristic odor. A smell may appear where one was not apparent before, or the reversemay be truea smell may disappear. The quality also changes with the variation in conditionsresponsible for the formation and movement of specific volatile compounds. Before consideringthe specific compounds associated with composting odors, it is important to recognize thefollowing truisms:

x Usually, the smell of a given material is a result of a mix of several volatile compoundsbased on the chemical composition of that material.

x Most odorous compounds are transient. After forming and being liberated, they

decompose, immobilize, change phase and/or disperse relatively quickly, depending onthe environmental conditions.

x The concentration of a particular compound determines whether or not its odor is detected,recognized and considered objectionable. Odorous substances that are generallyconsidered pleasing can become offensive at high concentrations (e.g. perfume, pine oil).Whether people find an odor good or bad, pleasant or unpleasant, are a quality known ashedonic tone. The hedonic tone says nothing about the character of the odor (e.g.pungent, septic, fishy), just how good or bad it is perceived to be.

x A compounds detection threshold, namely, the concentration at which a compound isdetected by people, varies greatly among volatile compounds. Some compounds can be

detected at extremely low concentrations while others require high concentrations.x The character and strength of odors are highly subjective.

x An individuals sensitivity to an odor and his /her reaction to it are greatly influenced bypersonal experience, gender, psychology and societal factors (Dalton, 2000a, 2000b).

Appendix B lists compounds and groups that are commonly associated with odors at compostingfacilities. A column has been added to the right of the original table which indicates otherpotential sources for the chemical listed. A review of this table indicates how it is possible forsimilar odors to come from a wide range of sources.

7. Dispersion of Odors

The only way an odor will be perceived off-site is if there is a person who: (1) detects the odor;(2) finds it objectionable and (3) becomes sufficiently annoyed to react negatively. As a result,potential odor impacts from a facility are influenced by conditions that affect the transport ofodorous compounds (e.g. weather, topography, and distance to neighbors).


15/42

{00235104.DOC /3} - 15 -

7.1 Transport Patterns Plumes and Puffs

Odorous compounds travel within air columns and gradually diffuse or wash out. Erosion of anair parcel depends on forces that diffuse, dilute, disperse and chemically alter compounds withinthe air. In air pollution terms, an intact but eroding air stream is called a plume. It is well

represented by the vapor plume from the exhaust stack of a power plant. The plume from thestack trails on for a distance, spreads out and eventually disappears. The wind, temperature andthe surrounding landscape determine the plumes direction, shape and longevity. (Lui, 1997).

Under calm conditions, plumes tend to gradually disperse, principally by diffusion, which causesthe plume to spread out both horizontally and vertically (de Nevers, 2000). The way in which aplume spreadsits spreading shapeis commonly described by a mathematical relationshipknow as a Gaussian distribution which is presented in Figure 2. To predict how a plume willbehave, air pollution professionals use mathematical equations that define the plumes Gaussianspreading pattern. The equations are used in computer models (e.g. the Gaussian model) topredict pollutant concentrations, like odor, at various directions and distances from the source.

FIGURE 2: Gaussian Distribution of a Plume

A plume represents a continuous stream of emissions at a constant strength and rate. However,odors are rarely emitted as a continuous stream at composting sites. For example, turning acompost heap could cause a release of odorous compounds. In those situations, the plume modeldoes not work well. In addition, under very stable atmospheric conditions (e.g. temperatureinversions, see below) the odorous air may hardly disperse at all (Haug, p 607). In these cases,odor emissions are sometimes modeled as a series of discrete puffs a brief burst of odorrelease.


16/42

{00235104.DOC /3} - 16 -

7.2 Proximity

The proximity of neighbors to an odor source has a huge influence in odor impacts. There isnothing more effective at preventing odor problems than an isolated site located a long distancefrom the nearest neighbors (Miner, 1995).

While a land buffer between a facility and its neighbors can be effective to limit odor impacts,the management of a facility, the surrounding topography and the attitude and expectation ofneighbors frequently have just as much impact on the frequency of odor complaints. This is whythe distribution of the postcard (Appendix A) resulted in an increase in odor complaints. Thepostcard impacted the attitude and expectations of the neighbors by implying a cause and effect.If you smell odors, they are coming for this particular source, so call PSCAA. Thus complaintswere received from distant locations or when the receptor was upwind of the compost facility.

This example demonstrates that meeting recommended buffer distances will not in itselfeliminate odor impacts, nor guarantee that odors wont impact neighbors beyond the buffer

distance.

7.3 Atmospheric stability

Winds generally mix in the atmosphere near the earths surface in the horizontal planeeast,west, north and south. Vertical mixing also occurs, from rising and falling air. The atmosphere isconsidered unstable when conditions are such that vertical mixing readily takes place. Whenconditions hinder vertical mixing, the atmosphere is stable. Vertical mixing is good forcomposting facilities because the plume of odors is carried upwards, where it is unlikely toencounter an objecting human nose. When it rises, it is also likely to be diluted and dispersed bythe more vigorous winds aloft.

Atmospheric turbulence is often categorized into six stability classes named A, B, C, D, E and F,with class A being the most unstable or most turbulent class, and class F being the most stable orleast turbulent class. Some of the characteristics of these source categories are shown in Table 1below.

Table 1: Atmospheric stability categories

Day Night

Surface WindSpeed Incoming Solar Radiation Night-time Cloud Cover1

(m/s) Strong Moderate Slight

>50% low

cloud


17/42

{00235104.DOC /3} - 17 -

A = extremely unstable D = neutral;B = moderately unstable; E = slightly stableC = slightly unstable; F = moderately stableAssume neutral condition D for all overcast conditions, day and night

Source: Pasquill, 1961In the lower atmosphere, up to about 6,000 ft. above the ground, vertical air movement occurs

because of temperature differences at different altitudes (de Nevers, 2000). The verticaltemperature profile the change in temperature with altitude is referred to as the lapse rate.When it refers to the actual temperature gradient of the ambient atmosphere, it is also called theambient lapse rate (Lui, 1997).

When air near the earths surface is heated by the suns energy, it becomes less dense, andtherefore buoyant, compared to the cooler air above. The warm air rises and the cool air falls,creating vertical mixing. However, if the air near the ground is cooler than the air aloft, it doesnot rise and no vertical mixing occurs. This situation is known as a temperature inversion.

7.3 Air Drainage

Air drainage occurs when the air near the earths surface cools at night. As the air at higherelevations cools, it flows down the slope. Only a slight slope is necessary. Cooling occurs fasterin wide-open expanses with a clear view of a cloudless night sky. In developed areas, both theground and adjacent air remain warmer as a result of heat retained by the thermal mass of thedeveloped landscape (e.g. asphalt parking lots, concrete buildings). On clear nights cooler airfrom uphill and less developed areas drains into the developed areas, displacing the warmer airdown slope.

7.4 Topography

Topography is determined by both natural features, such as hills, mountains, valleys, waterbodies, trees and vegetation, and manmade features such as buildings, roads, bridges, fences,power lines and planted landscapes. Topography may provide barriers to air flow and slowingwinds (e.g. wind breaks), establish airflow channels (e.g. valleys), increasing turbulence (e.g.trees) and generally creating localscale air circulation patterns (Lui et al., 1997).

Cedar Grove has attempted to use topography to its advantage by planting a row of trees aroundthe Everett site to minimize air drainage and create turbulent eddies when the wind is blowing.A soil berm (also employed by Cedar Grove around its facility) has a similar effect. In addition,

trees, fences or soils berms along the perimeter provide a visible barrier that shields the facilityfrom the public, and vice-versa.

8. Modeling Off-Site Odor Impacts

The remainder of this report focuses on Cedar Groves Everett facility, the odors generated bycomponents of the composting process, and the impact that Cedar Groves mitigation efforts


18/42

{00235104.DOC /3} - 18 -

have had on those odors. In addition, the report will discuss the effect other odor sources withinthe surrounding community on the cumulative impacts of odors. Finally, the report will assessspecific odor complaints that have been received.

8.1 Evaluation of the Process Components at the

Everett Facility

In January 2009, Cedar Grove, the consulting firm Envirometrics and PSCAA evaluated thenumber of dilutions that are necessary for specific processes not to create odor impacts at theclosest residence to the Everett facility. That home is located about 1,700 meters or 1 mile fromthe facility in the Everett Area. The EPA SCREEN3 model (version 95250), a numericaldispersion model, was used to estimate odor concentrations at downwind locations. The modelcalculates potential emissions under a wide range of meteorological conditions, including bothstagnant air conditions and turbulent air conditions. The model was run using rural dispersioncoefficients to simulate land use in the vicinity of the proposed project and assumed simpleterrain.

The dispersion modeling was conducted for four sources: (1) the Tipping Building Biofilter; (2)the proposed new Digester Building Biofilter; (3) Phase III of the composting process, and (4)the Curing Pile area. The detailed modeling analysis and results for these four sources areincluded in Appendix C of this document. The results were also used to evaluate the potentialimpact from the entire facility, discussed below.

As is elaborated below, under anything but upset conditions, even aggregating the emissionsfrom all of these sources and assuming the worst-case meteorological conditions and assumingevery source concentrated emissions on the same reception point, dilutions of the odor sourceswould dissipate sufficiently that off-site impacts would be no more than 1.6 OU under stable air

conditions, and 0.7 OU under normal conditions. Since background odor levels are typically 5-7OU, this modeling analysis demonstrates that even the most sensitive neighbor could not havedetected any odor from the Everett facility under normal weather conditions.

8.2 Aggregating Facility-Wide Impacts

The results from the Screen3 modeling can be used to model the combined impacts from the 5major areas investigated by PSCAA, namely, the Tipping Building Biofilter, Phases 1, 2 and 3,and the curing pile area. No odor data exists for Phase I and II, since the only sample collected

for Phase I was collected on September 15, 2008 and has been determined to be invalid.5 Noodor data exists for Phase I and II, since the only sample collected for Phase I was collected on

September 15, 2008 and has been determined to be invalid.6 However assumptions could be

5 All of the results provided by the lab for September 15, 2008 appear to be in error. This is discussed in greaterdetail in Appendix C All of the results were well outside of the expected ranges based on other samples collect atthis site and data collected at other sites.6 All of the results provided by the lab for September 15, 2008 appear to be in error. This is discussed in greaterdetail in Appendix C All of the results were well outside of the expected ranges based on other samples collect atthis site and data collected at other sites.


19/42

{00235104.DOC /3} - 19 -

made based on Phase III results. Two samples were collected on December 1, 2009 from aPhase III heap covered with a Gore

Cover. The results of the two tests indicated a DT of 150

OU and 160 OU. Based on this data, we have assumed Phase I has a DT of 250 OU and Phase IIhas a DT of 200 OU. This is slightly higher than Phase III, since the Phases occur earlier in theprocess. This assumption is also similar to odor data provided by Gore

for Phase I.

Phase I has two rows of 16 heaps and Phase II has one row of 16 heaps. In Screen 3, each row of16 heaps would be modeled in the same manner as Phase III was modeled. Table 2 below wasdeveloped based on all of these assumptions.

TABLE 2: Estimate of Facility-wide Impacts All Meteorological Conditions

Nearest Home Everett (1700m)

Nearest Home Marysville (2300m)

Marysville Homes(3000m)

SourceIdentification

DetectionThreshold

(OU)Reduction

(%)Impact(OU)

Reduction(%)

Impact(OU)

Reduction(%)

Impact(OU)

Biofilter 700 99.86% 0.98 99.89% 0.77 99.91% 0.63

Phase I(Row 1) 250 99.95% 0.125 99.96% 0.1 99.97% 0.075

Phase I (Row 2) 250 99.95% 0.125 99.96% 0.1 99.97% 0.075

Phase II 200 99.95% 0.1 99.96% 0.08 99.97% 0.06

Phase III 410 99.95% 0.205 99.96% 0.164 99.97% 0.123

Curing Pile 35 99.86% 0.049 99.91% 0.0315 99.94% 0.021

Total 1.6 1.2 1.0

Class F is stable and is characterized as occurring at night with wind speeds less than 3 miles per

hour and less than 38 percent cloud cover. This classification does not usually occur duringdaylight hours. Class D is considered to be neutral atmospheric conditions. Class D applies toheavily overcast days, at any wind speed, day or night.

TABLE 3: Estimate of Facility-wide Impacts Class D Meteorological Conditions

Nearest Home Everett (1700m)



SourceIdentification

DetectionThreshold

(OU)Reduction

(%)Impact(OU)

Reduction(%)

Impact(OU)

Reduction(%)

Impact(OU)

Biofilter 700 99.93% 0.49 99.95% 0.35 99.97% 0.21

Phase I(Row 1) 250 99.98% 0.05 99.99% 0.03 99.99% 0.03

Phase I (Row 2) 250 99.98% 0.05 99.99% 0.03 99.99% 0.03

Phase II 200 99.98% 0.04 99.99% 0.02 99.99% 0.02

Phase III 410 99.98% 0.08 99.99% 0.04 99.99% 0.04

Curing Pile 35 99.95% 0.18 99.97% 0.01 99.98% 0.01

Total 0.9 0.5 0.3


20/42

{00235104.DOC /3} - 20 -

When evaluating the dilution necessary to make an odor undetectable it is important to note thatambient air has a background odor level. When a DT is determined in the laboratory it isdetermined by comparison to clean, carbon-filtered air. By contrast, ambient air tends to have abackground or base level of odor that is around 5 to 7 OU. For this reason, the average person

will not detect an odor until the odor is present at greater than 5 OU. However, there are alwaysindividuals that are sensitive to a specific odor and may be able to detect that odor when it ispresent at concentrations as low as 1 OU.

8.3 Facility-Wide Modeled Emissions Indicate that

Off-Site Odor Impacts are Below Detection Levels

The use of the highest impact for modeling purposes renders the model conservative, because itis unlikely that the odor plumes from the 5 different sources would have their highestconcentration, or that the centerline of each plume would all impact in exactly the same location.

More likely the edges of the different plumes or areas of lower concentration would overlap,such that the total impact would be less than predicted by adding the highest impact for eachplume.7

Table 4 Summary of Odor Modeling Results

Modeled Odor Impact at Indicated Locations (OU)

Atmospheric StabilityNearest Home Everett (1700m)



Class F Stable 1.6 1.2 1.0

Class D Neutral 0.7 0.4 0.3

Figure 3, Impact Zones, depicts Cedar Grove and the surrounding area. Three circles have beenoverlaid on the picture indicating the areas with 1700, 2300, and 3000 meters. The red circleindicates the areas located within 1700 meters of the Cedar Grove facility. The yellow areas arebetween 1700 and 2300 meters from the facility and could potentially experience 1.6 to 1.2 OUduring stable meteorological conditions. The purple areas are between 2300 and 3000 meters

from the facility and could potentially experience 1.2 to 1.0 OU during stable conditions.

7 On the other hand, this modeling exercise does not include a number of smaller odor sources at the facility whichmay contribute to the predicted odor impact, because the odor concentrations are low and there is no air flowthrough media minimizing the potential for these sources to contribute significantly to off-site odor impacts. In

particular, the finished compost area was not included because it has no airflow, so an appropriate modeling scenariocould not be implemented for this analysis.


21/42

{00235104.DOC /3} - 21 -

Figure3

ImpactZones


22/42

{00235104.DOC /3}- 22 -

As was noted above, ambient air tends to have a background or base level of odor that is around 5 to 7OU, meaning that, for the average population to detect an odor, it will need to be present at greater than5 OU. While an impact of 1.6 OU would most likely not be detected by the average population, thereare certain individuals that are sensitive to a specific odor and may be able to detect that odor when it ispresent at concentrations as low as 0.5 OU. The analysis indicates that during very stable conditions,

sensitive receptors at the residence closest to the site may experience the odor. However, very stableconditions typically only occur a few times a year. When they do, it would be unlikely that the odorcould be detected at residences located greater than 3000 meters from the site.

The likely conclusion of the analysis is that an odor source other than Cedar Grove is located near thecomplainants. The other possibility is that stable air conditions could lead to cumulative impacts caused,not just by Cedar Grove, but by the other sources in the area. These two possibilities are discussed ingreater detail later in this document under Cumulative impacts and the History of Odor Complaints.

For the tipping building biofilter, when weather conditions are more neutral, or a stability class D, theimpact at the nearest home in Everett drops significantly to an impact of 0.7 OU. Even the mostsensitive individual should be unaware of an odor at this level.

The results of this modeling analysis indicate that, normally, odors could only be detected by theaverage population during periods of upsets within the process, or by sensitive individuals during stablemeteorological conditions. Upsets and stable meteorological conditions would explain a small numberof odor complaints per year. However, they will not explain the 331 odor complaints logged in 2009, orthe fact that 155 (47 percent) of these complaints were determined to be meteorologically improbable.Two alternative explanations are that a number of these odor complaints could be related to cumulativeimpacts from multiple sources or might be caused by other odor sources in the area. This is discussed ingreater detail later in this document under Cumulative impacts and the History of Odor Complaints.

9. Proposed Modifications to ExistingOperations

Cedar Grove is evaluating changes to its existing operations both as a means of addressing theincreasing demand for processing yard and food waste, as well in an effort to reduce odors from itscomposting operations. The primary modification currently under review is adding anaerobic digestersto the process. Cedar Grove is evaluating additional odor controls for its existing processes.

9.1 Potential New Operations Dry Fermentation

Cedar Grove is investigating a dry fermentation technology to replace some Phase 1 of its the currentprocess. Dry fermentation uses micro-organisms to break down biodegradable material in the absenceof oxygen. The dry fermentation process utilizes renewable resources as a feedstock to produce amethane and carbon dioxide rich biogas. The nutrient-rich solids left after digestion can be used as afertilizer and compost. The biogas can be combusted to produce renewable electricity, cleaned topipeline natural gas standards, or further processed into compressed natural gas (CNG). One of the


23/42

{00235104.DOC /3} - 23 -

benefits of dry fermentation is the reduction of greenhouse gas emissions (through replacement of fossilfuels) and landfill methane emissions, and displacement of chemical fertilizers.

In addition to the production of biogas, Cedar Grove expects the new technology to result in fewer odorsper ton of material processed. This is partly due to the fact that tipping and screening of material that isdestined for the digester will take place indoors. The buildings will be equipped with a new, largerbiofilter. During the anerobic digester mixing process, a majority of the gases produced will becollected and sent to the new, larger biofilter. There is also the potential for a reduction of odors duringPhase II and III of the process, since the anaerobic digester is expected to break down more of thevolatile fatty acids than the current Phase I process. Cedar Grove is initially planning to divert 50,000tons of material from the Phase I composting process to the dry fermentation system in order to evaluatethe effectiveness of the system and its effect on odor emissions.

The technology being evaluated uses a batch process under which organic waste is loaded intoindividual chambers on a 21-28 day cycle. Percolate (bacterial agent) is applied to the biomass whilespecialized gates prevent oxygen from entering the system. The organic biomass within thefermentation chamber is kept at a constant temperature through an in-floor heating system to maximizethe decomposition rate. No movement of biomass is needed during the entire 21-28 day cycle.

The handling, mixing and loading of organic waste is conducted inside a building, with building airvented to a biofilter. Biogas collected from the digesters is sent to a gas cleaning system and then toeither a generator or turbine for electrical generation or stored as compressed natural gas. The gascleaning system is designed to clean the biogas to whatever level is required for the end use of the gas,which often requires removing hydrogen sulfide and other sulfur compounds from the gas.

After the organic matter completes its 21-28 day fermentation cycle it is moved into Phase II of theGore Cover composting process.

9.1.1 Gas Use

Gas use will be through one of three methods. In the first option, gas will be fed to a reciprocatingengine and electricity will be created. The power not consumed at the site will be transmitted throughwire to the Snohomish PUD. This system could also capture waste heat for re-use. A second alternativewould be to pipe the gas to a nearby user (to be determined). Finally, the gas could be cleaned andcompressed for either pipeline sale or in a mobile distribution system that would make the gas availablefor fleet use (such as a transit system.)

9.1.2 Moving Digestate

After 21 to 28 days in a digester tunnel most of the gas potential has been realized. At this point thetunnel is opened and the material is unloaded. A variety of approaches are currently being evaluated,and the most efficient and odor sensitive method of moving the material from the digester to thecomposting process will be selected. From this point, the material is handled in the same way asdescribed in the current and future processes from Phase II onward.


24/42

{00235104.DOC /3} - 24 -

9.2 Modifications to Existing Operations

Cedar Grove is evaluating additional odor controls for its existing composting process. Future optionsbeing reviewed include the sprinkler system for the tipping building biofilter and options for coveringfinished product.

9.2.1 Tipping Building Biofilter

Cedar Grove is evaluating the effect of a new sprinkler system on the tipping building biofilter. Theexisting procedure for maintaining moisture in the biofilter included watering the surface of the biofilterusing a watering truck. The facility added a sprinkler system to the biofilter which operates 6 hours perday. This allows the water to be added at a slower rate over a longer period of time. Eventually a timermay be added to the system which would allow for intermittent watering of the biofilter throughout theday. During dry weather conditions, the watering truck may still be used to provide additional moistureto the biofilter if needed. Testing has indicated the change has resulted in decrease in odor emission.

10. Cumulative Impacts from AdditionalOdor Sources in the Area

According to a San Diego State University Report, Competing odor sources are probably underrated asa contributing factor to odor problems. In some cases a facility was sited intentionally near other odorsources in an attempt to be with like uses. However, competing odor sources can have negative effectson a facility, especially if the facility has other issues as well. If a compost facility is sited near acompeting odor source it may aggravate issues the neighbors may already have with that odor sourceand they may turn their displeasure on the composting facility. In addition, odors from two or more

facilities may commingle and change, especially over distance.

10.1 Additional Odor Sources Located Near Cedar Grove

Smith Island is home to many sources that produce compost or wood products. There are also additionalwood product industries located in the immediate area. The potential wood products sources and theiraddresses are provided in Table 5.

TABLE 5: Potential Wood Products Odor Sources

Source Type of Operation Address

Pacific Topsoils

Stock pilling topsoil, wood

products 3000 Frontage Road

Sringbrook Nursery and TruckingNursery 9022 84th Street Northeast,

Arlington, WA

Kimberly Clark Chip Storage Wood Chip Storage Weyerhaeuser Bridge Road

Kimberly Clark Pulp and Paper Mill 2600 Federal Avenue

Miller Shingle Wood Products Smith Island


25/42

{00235104.DOC /3} - 25 -

Miller Shingle Wood Products 2326 28th St NE

Buse Timber and Sales Wood Products 3812 28th Place Northeast

Everett Bark Bark Distribution and Storage 4116 34th Avenue Northeast

Sno-River Chip Reload Wood Chip Storage 1910 Ross Avenue, Everett, WA98205-3211

All of these sources are capable of producing odors similar to those generated by composting facilities.In addition, on many days, odors can be detected at the property line of these facilities. Alone orcumulatively, these sources likely generate effects downwind of Smith Island. Unless these other odorsources are required to reduce their odor impact levels, adding additional odor controls and reducingemissions from Cedar Grove are unlikely to improve the odor impacts on the community.

Examples of odor events that may have been caused by cumulative odors are presented in the analysis ofodor complaints for the Everett Area. In addition, an analysis was made of the history of complaintsfrom one frequent complainant, who lives in Everett and reported a total of 11 complaints in 2009.8

This complainant was invited to tour Cedar Groves Facility to help identify the source of the odor. Heindicated that the only odor source on site that was close to the odor he could smell at his residence wasthe odor from the overs storage pile. Overs are the portion of the finished compost that is too big togo through the final material screening. Normally this would be material larger than 7/16 of an inch.The volume of overs stored onsite is less than 10% of the total material onsite. It is also the product thatis most similar to the products located at wood handling facilities in the area.

The complainant lives more than 3500 meters from Cedar Grove. Based on the results of the odorimpact modeling outlined in Section 8 of this report, it is unlikely that the volume of overs located atCedar Grove would be able to generate enough odor to create an impact at a location that far from thesite. However, if calm weather conditions lead to a fairly intact parcel of air forming over Smith and

Spencer Island, than that parcel of air could contain the odors emitted from all of the sources on the twoislands. The total volume of overs, wood chips and other similar materials from multiple sources maybe enough to generate odors at that distance, during stable air conditions. Odors from Cedar Grovealone, even during stable air conditions, would not be.

In addition to these wood products facilities, other potential odor sources exist in the surrounding area.Appendix B identifies common odorous compounds released during the handling and decomposition oforganic materials. It also identifies other processes that occur in the Everett-Marysville area that likelycreate the same compounds or use materials that have a potential to release those compounds. Later inthis report, a number of the odor complaints recorded by Cedar Grove are analyzed. That analysisindicates that some of the complaints may be due to boat repair facilities in the area. Cedar Grove

employees and consultants have detected sweet woody odors apparently coming from marine repairyards both near and distant from the Everett composting facility. They have not, however, been able to

8 All of the complaints occurred during or immediately after a long period, generally more than 8 hours, of wind speeds lessthan 1 mph. This would indicate possible stable conditions. However, as discussed in Section 7.3 of this document,atmospheric stability is also dependent of solar radiation. Seven of those complaints, were determined to be meteorologically

possible since the winds were calm and from the North. The other four complaints occurred when winds were calm and fromthe WSW.


26/42

{00235104.DOC /3} - 26 -

identify the activity generating these odors. At these facilities epoxies and paints are frequently used onthe boats. Some of these facilities have undergone recent expansions. Others may have made a shift inthe last few years in the types of solvents and strippers used. The paints and hardeners usually containamines as well as other solvent. According to someone in the industry, some of them also containodorants to try to mask or improve the smell of the material. It is common for hardeners and paints for

the bottom of boat hulls to have a sweet-woody odor.

The list of boat repair facilities in the area is presented in Table 6.

TABLE 6: List of Shipyards and Marinas in Everett

Shipyards and Marinas Address

Everett Shipyards 2730 Federal Avenue

TC Systems, Inc 1032 West Marine View Drive

Everett Engineering 25 West Marine View Drive

Navel Station Everett 2000 W Marine View Dr. Everett

Hansen Boat 4124 34th Avenue Northeast

Geddes Marine 1326 1st Street, Marysville, WA

Harbor Marine 1402 W Marine View Dr,

Venture Pacific Marine 4000 Railway Ave Everett

There are many other potential odor sources in the Everett and Marysville area besides wood productand paper mills, other composting facilities, and the marina. The list includes wastewater treatmentplants, tanneries, and agriculture. The locations of some of the sources are shown in Figure 4.


27/42

{00235104.DOC /3} - 27 -

FIGURE 4: Location of Shipyards and Marinas in Everett

10.2 Impacts from Other Sources May Generate a

Cumulative Odor Impact on the Community

When odorous compounds are released into the air, they are assimilated by the surrounding air. PacificTopsoil, Miller Shingle, Buse Timber and Sales, Everett Bark, Seacrest Marina, Hansen Boat and theMarysville WWTP lagoons are all located within 1 mile of Cedar Groves Everett facility. During eitherstagnant air condition or during an inversion, odor emissions from all of those sources have the potentialto be released into the air parcel above Smith and Spencer Islands. As a result, odors from two or moreof these sources can commingle and change, especially over distance.

When the wind is blowing from the Southwest, the air parcel that moves over Cedar Grove will also bemoving over Everett Bark, Seacrest Marina, Hansen Boat, and the Marysville WWTP lagoons and liftstation before it reaches residences in Marysville. Consequently, odors from all sources located south ofCedar Grove will be traveling in the same general direction, towards Marysville. The cumulative impactof all of these sources may provide an explanation as to why odor complaints may be made byMarysville residents, despite the fact that the modeling outlined above indicates that any potential odorsfrom Cedar Grove should have been dispersed to a level below the DT.


28/42

{00235104.DOC /3} - 28 -

11. Analysis of Odor Complaints

As outlined above in Section 4, in response to the odor complaints received in 2008, Cedar Grove spentmore than $1 million on odor mitigation and also made a number of operational changes to reduce the

impacts of odor from its operations.

Section 8 outlines the results of the modeling conducted by PSCAA, CH2M Hill and Envirometrics.That modeling indicates that, barring upset conditions, off-site impacts from the largest potential odorsources should be diluted to below the detection threshold by the nearest neighbor in Everett. Inaddition, as outlined in Section 9, there a large number of potential odor sources, many of which emitcompounds that could be mistaken for odors from Cedar Grove, located in the immediate surroundingarea.

In order to test the effectiveness of its investment in odor mitigation, and based on the modeling and theexistence of a large number of alternative odor sources in the general location of Cedar Groves Everett

Facility, Cedar Grove decided to evaluate the odor complaints received by PSCAA or by Cedar Groveduring 2008 and 2009. Cedar Grove has recorded all odor complaints made during July and August in2008 and for the entire year of 2009. After assembling all of the complaints, a careful meteorologicalreview of most of the complaints was made in order to determine if the complaints weremeteorologically possible or improbable. This review included plotting both the complainants locationand the wind direction at the time of the complaint. A complaint was determined to be meteorologicallypossible if the wind direction indicated that Cedar Grove or a source near to, upwind or downwind ofCedar Grove could have been the source of the odor. The complaint was determined to bemeteorologically improbable if the wind direction indicated that it was very unlikely that the odor couldcome from Cedar Grove. Wind data was obtained from up to four meteorological stations that arereporting in the Everett and Marysville area. These include the station at Cedar Groves facility, a

Mesowest station and several amateur stations.

In a number of instances, the meteorological data suggests that it would be impossible for odors to movefrom the Cedar Grove facility to the complainants location. This indicates that a large number of theodor complaints directed at Cedar Grove may have been made based on the complainants perception ofodor at an earlier time than when the complaint was reported, or, that the complainant may be smellingodors from a source other than Cedar Grove. The remainder of this section examines specific odorcomplaints that appear to validate this hypothesis.

11.1 Complaints Received in July and August 2008

During July and August 2008, 126 odor complaints were received by either PSCAA or Cedar Grove.These complaints were reviewed and segregated by time of day. Complaints were further segregated

into those received between 6 am and 2 pm and complaints received from 2 pm to midnight.9 One ofthe reasons for analyzing based on time-of-day is that Cedar Grove has discretely different operationsoccurring in the earlier hours than the later hours. Table 16 summarizes the complaints, the number of

9 No complaints were received between midnight and 6 am in the time period analyzed..


29/42

{00235104.DOC /3} - 29 -

individuals making complaints, and complaints by time period. Based on the analysis of wind data,Table 16 also identifies those complaints that were considered meteorologically improbable.

Table 16: July and August 2008 Complaints

Total Individuals MeteorologicallyImprobable

July 104 64 18

August 23 17 5

time 6 am 2 pm 2 pm-midnight No time recorded

July 69 32 3

August 14 9 0

The number of complaints received each day is shown in Figure 5. The maximum number was 19complaints, which were received on July 17, 2008. Four of those complaints were received from one

person over a period of about 10 minutes. On the previous day, nine complaints were received. As canbe seen in the figure, the complaints on these two days were significantly greater than the numberreceived during the rest of July and August. On the six days during the period from July 9 through July18, complaints were generally elevated when compared to the rest of the days depicted. More than halfof all the July complaints were made on those six days. Complaints declined significantly in August.

FIGURE 5: Number of Complaints made each Day and Daily Maximum Temperature

Marysville Odor Complaints

0

5

10

15

2025

30

35

7/1/20

08

7/8/20

08

7/15/200

8

7/22

/200

8

7/29

/200

8

8/5/20

08

8/12/200

8

8/19/200

8

8/26

/200

8

Complaints

-C

Figure 5 also graphs the daily maximum temperature, shown in Celsius, allowing a comparison

between the number of complaints and the daily temperature. There is no apparent relationship thecorrelation coefficient is only 0.12, suggesting little to no relationship between the variables.

In addition, negative tides occurred in the Snohomish River estuary adjacent to Cedar Grove severaltimes during July and August. During the period when the greatest number of complaints was received,there were negative tides from July 14 through July 21.10

10 The low tide was at 9:10 am on the 14th and advanced about 40 minutes each day to 1:30 pm on the 21st.


30/42

{00235104.DOC /3} - 30 -

The diurnal pattern of winds in the Snohomish River estuary was relatively consistent throughout July.The night and early morning hours were almost always calm. When the night winds were high enoughto record a wind direction, it was generally blowing upriver toward the south southeast to southeast..With calm winds overnight it is possible for odors to accumulate in a quiescent air mass and then moveto the receptors, causing a rash of complaints immediately after the calm is broken in the morning.

However, review of the complaints during the six days in July with the highest numbers of complaintsshows that this could not be the case on five of the six days. On those days, the complaints began morethan one hour after the calm broke and on three of the days the complaints were predominately in theafternoon.

This analysis suggests that while odors from Cedar Grove could, from time to time, generate sporadicodor complaints from the neighborhoods northeast of the facility, the cause of many of the complaintson the six days in July may well have been an unusual set of circumstances. In the alternative, itsuggests that other sources of odors besides Cedar Grove might have been responsible for at least someof the odors on those days where complaints were the highest.

11.2 Complaints Made in 2009Cedar Grove has analyzed odor complaints for all of 2009. A total of 331 complaints were made in2009. Based on a correlation of wind data and the time and location of the complaint, 163 (49%) of thecomplaints were meteorologically improbable, and 21 (6%) were unable to be determined because noaddress was provided. Twenty-nine (29) individual households were responsible for 230 (69%) of the331 complaints.

Figure 5 shows the number of complaints by time of day for each month in 2009. The time period iscategorized the same way as it was in 2008, 6 am to 2 pm and 2 pm to midnight. For eight months, therewere more complaints from 2pm to midnight than from 6am to 2 pm. There were more complaints from6am to 2pm for four of the months, in particular for the month of June. In June, 135 complaints wererecorded, accounting for 41% of all of the complaints for 2009.


31/42

{00235104.DOC /3} - 31 -

FIGURE 6: 2009 Complaints by Time of Day

Figure 6 shows a sharp rise in complaints in May to June. A total of 29 complaints were recorded inMay, with eight of those complaints made on May 31. The number of complaints per day for June 2009is presented in Figure 7.

FIGURE 7: June 2009 Complaints

June 2009 Complaints

0

2

4

6

8

10

12

14

16

6/1

6/2

6/3

6/4

6/5

6/6

6/7

6/8

6/9

6/10

6/11

6/12

6/13

6/14

6/15

6/16

6/17

6/18

6/19

6/20

6/21

6/22

6/23

6/24

6/25

6/26

6/27

6/28

6/29

6/30

Day

NumberofComplaints

0

10

20

30

40

50

60

70

80

JanuaryFebruary March April May June July AugustSeptemberOctoberNovemberDecember

Month

Number of Complaints

6 am - 2 pm

2 pm - 12 am


32/42

{00235104.DOC /3} - 32 -

11.3 Example Complaints Analysis

Cedar Grove conducts an analysis of every complaint received by the company as well as those receivedby the PSCAA. An example of how the analysis is conducted is provided below. The example analyseswill indicate the criteria used to determine if a complaint is possible or improbable and also how theinclusion of additional data, like distance, the description of the odors, and proximity to other odorsources, often indicates that even when the location of a complaint is determined to be meteorologicallypossible, the odor may be attributable to another source or combination of sources.

The highest number of complaints was recorded on June 4, 2009, a total of 14 complaints were recorded.Ten (10) of the complaints were determined to be meteorologically possible (71 percent). One of thecomplaints did not provide an address so could not be included in the analysis. The complaints werelodged during two separate blocks of time; in the morning from 8:30 am to 11:15 am and from 20:58 to21:55 in the evening. Since the odor event during the morning hours included odor complaints fromboth Everett and Marysville, as well as one complaint from the south end of Marysville near LakeStevens, it was selected to be used as an example of the complaint analyses performed by Cedar Grove.It is worth noting, that on June 4 a Snohomish Health Department Inspector was onsite conducting aninspection. The inspector noted on his report:

6/04/09: During this site evaluation, I did not detect any negative odor associated with the operation. Inaddition, I noted that one of the doors to the tipping building was closed and that the building wasoperating at about 25 percent of capacity.

The Inspection reports are included in Appendix D of this report.

Six complaints were recorded on June 4 from 08:30 to 11:15. According to the Cedar Grovemeteorological station, on-site winds were calm, less than 1 mph, with no wind gust from the SWdirection until about 07:56. From 07:56 to noon, the wind continued to come from the generally SWdirection with light wind speeds from 1 mph to 9 mph. The MJHS meteorological station recorded thewinds from the WNW at approximately 2 mph, until the winds shifted at around 9:40 am to a lightWSW wind. The Sunnyside meteorological station recorded winds from the WNW from 08:19 to 0929at approximately 2 mph, until the winds shifted at around 09:44 to a light WSW wind.

Figure 8 shows the locations of the complaints and the wind directions recorded by the Cedar Grove andMJHS meteorological stations during the morning hours. The houses indicated in red are a residencewhere a complaint was made that was deemed meteorologically possible, based on the wind data. A

green house indicates that a complaint was recorded at that residence that was deemed meteorologicallyimprobable, based on the wind direction and wind speeds.


33/42

{00235104.DOC /3} - 33 -

FIGURE 8: Odor Complaints Morning of June 4

As Figure 8 indicates, two odor complaints were recorded during the time reviewed from residences inthe City of Everett. Both of these were determined to be meteorologically improbable because thelocations were all to the south of Cedar Grove. Before and during the time of the complaints, winds wereblowing from the SW towards the NE. This wind direction would indicate that the odor experienced bythe complainants in Everett was generated at a source located to the SW of the residences. There areseveral shipyards to the SW of the complainants as well as the Kimberly Clark paper mill.

One odor complaint was located to the SE of Cedar Grove at 6620 Sunnyside Boulevard and wasdetermined to be meteorologically improbable. At the time of the complaint the winds at that locationwere from the SW. Cedar Grove is located NW of this residence. The proximity of this residence to an

agricultural area, which is located to the S and SW of the residence, indicates that this may be the sourceof the odor. In June, agricultural areas frequently spread manure and other fertilizers on their fields,which would have an odor similar to compost. Feed that has been fermented, such as silage, balage orhaylage, can also have strong odors some associate with compost.11

11See for example, From Harvest to Feed: Understanding Silage Management, C.M. Jones, A.J. Heinrichs. G.W. Roth and

V.A. Ishler College of Agricultural Sciences, The Pennsylvania State University, 2004


34/42

{00235104.DOC /3} - 34 -

The remaining 3 complainants were located to the NE of Cedar Grove in Marysville. The odorcomplaints were determined to be meteorologically possible, because Cedar Grove is located upwind ofthese residences.

At this point it should be noted again that a Snohomish County Health Inspector visited Cedar Grove on

June 4, and did not detect any negative odors, or odors that might indicate a problem with the process. Inaddition, as discussed above in the section on cumulative impacts, when the wind is blowing from theSW towards Marysville, the air parcel that moves over Cedar Grove will also potentially move overPacific Topsoils, Everett Bark, Seacrest Marina, Hansen Boat, Gedddes Marine, the Marysville WWTPlagoons headworks and lift station, as well as at least one agricultural field before it reaches residencesin Marysville. The odors experience by the residence could originate from any one or more of thesesources, and as was indicated by the complaints recorded in Everett which were meteorologicallyimprobable, odors from other types of operations can be confused with the odor generated by compost.

11.4 Analysis of Complaints for Everett Area

In 2009, there were a total of 93 complaints received residential areas in Everett. Based on an analysisof the wind direction and strength at the time of the complaint, 41 of those complaints (44%) weredetermined to be meteorologically possible, 51 (55%) were determined to be meteorologicallyimprobable, and 2 were unknown because no address was provided. The meteorologically improbablecomplaints indicate that a majority of the odors may be coming from the SW or SE areas of the City ofEverett. There are also some indications that there may be small sources of odor located near theresidences of certain complainants. The example odor complaint analysis below provides an indicationof some of those potential odor sources.

The day on which the highest number of meteorologically possible complaints was received fromEverett was June 11. On the Evening of June 11 from 20:27 to 22:15, four odor complaints werereceived; all were from the Everett area. During this time period the winds were calm, less than 1 mphand emanating from the N. Therefore the complaints were determined to be meteorologically possible.Figure 9 indicates the locations of the odor complaints and wind directions.


35/42

{00235104.DOC /3} - 35 -

FIGURE 9: Odor Complaints Evening of June 11

A Snohomish Health District Inspector was on-site at Cedar Grove on June 11, 2009. In that inspectionhe noted the following:

Driving to the facility I detected a faint sour odor that is different from the odors associated with anormal compost operation, however Im not certain it came from the compost facility. In walking thesite I noted the facility was operating in compliance with the solid waste handling permit.As discussed above in the section on Cumulative Impacts, when the winds are calm and coming fromthe N it is possible for a fairly intact parcel of air to move from Smith and Spencer Islands over theEverett area. That parcel of air would contain the odors emitted from all of the sources on the twoislands. However, it is also possible on very calm days, for any odors generated in the Everett area toremain trapped in a stagnant air parcel over the City. The following, review of the odor complaints for

the morning of June 11, indicates that this may have been the situation.

On the morning of June 11 the winds were from the N or NNW, the wind speeds were recorded as < 1mph with no wind gusts. The winds changed to a generally WSW direction at around 08:46. An odorcomplaint was recorded for the residence at 1110 Poplar Street at 08:36. The complaint was deemed tobe meteorologically possible because the winds were from the north before and during the time of thecomplaint. When the winds shifted at 08:46, the wind speed also increased to a steady light wind withspeeds of 3 to 8 mph. The light winds should have resulted in displacing any stagnant air parcel that was


36/42

{00235104.DOC /3} - 36 -

sitting over the Everett area. However, two more odor complaints were recorded in Everett at 11:30 and13:04. Since the wind was now blowing steadily from the WSW, these complaints were considered tobe meteorologically improbable. In addition, during this same time period, two odor complaints wererecorded in Marysville, one at 10:48 and another at 11:06, which were deemed meteorologicallypossible since the residences were now downwind of Cedar Grove. The coincidence of these complaints

being received on the same day and at roughly the same time means that the complaints in Marysvilleand Everett could not be caused by the same odor source, unless that source was located to the SW ofthe city of Everett. Figure 10 indicates the locations of the odor complaints and wind directions duringthe morning hours.

FIGURE 10: Odor Complaints Morning of June 11

As discussed above, the meteorological conditions in Everett on the morning of June 11 indicate thesource of the odor is most likely located somewhere in Everett and possibly to the SW of the complaints.Harbor Marine, TC System, Everett Navel Station, Everett Shipyards and Kimberly Clark Pulp andPaper Mill are all located within the City of Everett and downwind of the complainants.

The review of the odor complaints for Everett also indicates that some individuals may be living rightnext to an odor source. One example involves a frequent complainant who lives in the City of Everett.


37/42

{00235104.DOC /3} - 37 -

Cedar Grove recorded 22 odor complaints from this person in 2009, 12 complaints (55 percent) of theodor complaints were determined to be meteorologically possible. The remaining complaints occurredduring wide ranges of wind speeds and wind directions, which would indicate either multiple sourceswith similar odors or one source very nearby. The individual was invited to tour Cedar Grove to help thefacility locate the source of the odor. After touring the facility, the person was unable to detect the

particular odor she was experiencing at her residence from any of the processes at the facility. It waslater determined that the individual lives approximately 1 block from a crematorium, which has thepotential to be the source of the odors.

The morning odor event also brings into question the odor complaints recorded during the evening hourson June 11. Perhaps the odors generated during the morning hours were from a source located withinthe City of Everett. This would also explain why the complainants described the odor as chemical likeor burning the eyes, nose and throat. This is not the typical description for odors from a compostingfacility, but would instead indicate that the chemicals causing the odor may be solvents, likenaphthalene, xylene, and toluene. Any of the before mentioned facilities, Harbor Marine, TC System,Everett Navel Station, Everett Shipyards and Kimberly Clark Pulp and Paper Mill, could use solvents in

their operation, which would produce a chemical like smell.

11.5 Analysis of Complaints for Lake Stevens and Areas

Located Southeast of Cedar Grove

In 2009, there were a total of 8 complaints recorded for the area SE of Cedar Grove including LakeStevens.. Of those complaints 3 (38%) were determined to be meteorologically possible and 5 (62%)were determined to be meteorologically improbable. The day with the highest number ofmeteorologically possible complaints recorded was June 8. On June 8th from 18:30 to 20:02, two odorcomplaints were received and both were determined to be meteorologically possible. One odorcomplaint was received at 09:27 in Marysville, but was determined to be meteorologically improbable,since the winds at Cedar Grove were from the NE during and before that time period.

From 11:46 to the time of the complaints the winds at Cedar Grove varied from the WSW to the WNW,with consistent light winds from 1 to 11 mph. Figure 11 indicates the locations of the odor complaintsand wind directions.

The two Lake Stevens residences are approximately 2.7 miles from Cedar Grove. During stableatmospheric condition it may be possible for a plume or puff of odorous air to stay intact enough overthat distance to create an odor impact. However at the time of the complaint, winds were light butconsistent, indicating a more neutral or even slightly unstable condition. Under these conditions, aplume or puff is less likely to stay intact over that distance, indicating that a much stronger odor source

or one located closer to the complainants house would be more likely to generate a complaint.

The conclusion that Cedar Grove would be unlikely to be the source is also indicated by the 62 percentof the complaints that were determined to be meteorologically improbable. The improbable complaintsall have meteorological conditions similar to the condition seen in Figure 8 above for 6620 SunnysideBoulevard on June 4. The winds were generally coming from the SW. There are agricultural areaslocated nearby and upwind of the location of the complaints. Many farms apply fertilizer, often in theform of animal manure, to their fields which would generate odors similar to compost. Feed that has


38/42

{00235104.DOC /3} - 38 -

been fermented, such as silage, balage or haylage, can also have strong odors some associate withcompost. It is likely these agricultural sources are the source or contribute to the odors in this area.

FIGURE 11: Odor Complaints Evening of June 8

11.6 Analysis of Complaints for Marysville AreaIn 2009, there were a total of 209 complaints received from the Marysville area. Based on an analysis ofthe wind direction and strength at the time of the complaint, 120 of those complaints (57%) weredetermined to be meteorologically possible, 89 (43%) were determined to be meteorologicallyimprobable.

The analysis of the odor complaints for Marysville indicate a strong possibility that other sources maybe involved in many of the odors experienced by residence in Maryville. Not only do the meteorologicalconditions indicate the possibility, but the descriptions of the odors provided by the complainants oftendont match the type of odors generated by compost. The odor descriptors used by many complainantsin Marysville include dead corpses, a rendering plant, sour garbage, sewage and manure.Compost generally has a woody smell with a slight sweet note or a slightly sour note like vinegar.

Figure 12 indicates specific areas in Marysville were clusters of complaints have occurred. The areas 1,2, 4 and 6 accounted for 82 perc

Cedar Grove Composting Odor

Documents