Odor Monitoring_integrated Management Of Odor Nuisance In A Tourist Region

Odor Monitoring - Integrated management of odor nuisance in a tourist region Paper: 2009-A-326-AWMA Philippe Micone Odotech France, 20 rue de la Villette, 69328 Lyon Cedex, France Thierry Pagé Odotech Inc., 3333 Queen Mary Road, Suite 301, Montreal (QC) Canada, H3V 1A2 Jean-Michel Martin SAUR, 1, avenue Lavoisier, 78064 La roche sur Yon, France Jacques Baud Sivos des 60 Bornes, President, Mairie de St-Hilaire de Riez, France

ABSTRACT A real-time odor impact monitoring system was installed at the Sivos des 60 Bornes waste water treatment plant (WWTP), which treats the waste water of two towns, St-Hilaire-de-Riez and StJean-de-Monts, on the French coast in the region of Vendée. The users’ objective was to minimize odor nuisances generated by pumping stations of the waste-water system and by various components of the WWTP. The system (called OdoScan®) enabled the user to identify the main source of odor emissions, in this case the outlet of the deodorization process of the pretreatment building of the WWTP. Continuous odor monitoring by site operators allowed immediate and direct action on the source responsible for nuisances, thereby appreciably reducing the number of complaints received by the Sivos during the 2008 summer season. Finally, an analysis of the average impact on the surrounding area helped the user to identify that pumping stations are a potential weak spot in terms of nuisances experienced, but more importantly, that generally, very low levels of odor were experienced by residents.

INTRODUCTION Odor nuisances generated by industrial activities on the French coastline have been regularly identified by vacationers as bothersome over the past few years. These nuisances create a new set of problems between tourists, owners and managers of tourist sites, elected community officials, and local industry. Waste treatment activities in these communities are particularly affected by such problems. During the summer season, the quantity of waste to be treated can often amount to more than 10 times the usual off-season load. This overload causes an increase in odor emissions from treatment sites, and a heightened risk of occurrence of odor nuisances. The installation of odor control equipment on problematic sources is usually the ideal solution. However, to ensure that the investment in such technologies has a real impact on nuisance reduction, it should ideally be accompanied from the outset by an assessment of the odor sources

1

that might generate nuisances, within or outside an industrial site, as well as by monitoring of the improvements achieved. Indeed, while some sources are often perceived as the main culprits in generating odor nuisances, and treated as such, many odor sources with a significant environmental impact are overlooked, so the odor problem persists. This article details the installation of a tool for monitoring nuisance odors in the towns of StHilaire-de-Riez and St-Jean-de-Monts on the French Vendéan coast, a major summer vacation resort area. Together, the two towns host more than 100,000 visitors per week during the summer season, in addition to their usual 15,000 residents. Historically, the main source of nuisance identified by elected officials, residents and tourists was the waste water treatment plant (WWTP). The site of the Sivos des 60 bornes WWTP is in the town of Saint-Hilaire-de-Riez, in the Vendée department. This plant treats all the waste water of the town of Saint-Jean-de-Monts and onethird of the waste water of Saint-Hilaire-de-Riez. The site is near "60 bornes" beach. The coastal road passing in front of the plant links departmental route 123 (D123) to the southern subdivision of the town of Saint-Jean-de-Monts (900 meters as the crow flies). To the north, the plant is separated from this subdivision by an area of sand dunes (the "Pré Salé"), wherein lies a residential zone and a campground. They are the plant's closest neighbors, 180 meters from the nearest odor source on the site. To the southeast, the village of Les Becs borders the beach at a distance of 1 kilometer, but residences further east are even closer (Le Petit Bec at 500 meters). Finally, towards the northeast, besides the residences along the D123, the public forest of Pays de Monts acts as a vegetation buffer prior to the town center of Saint-Jean-de-Monts. The following figure illustrates the above description. Figure 1. Area around the plant.

2

2. METHODOLOGY 2.1 Identifying the odor sources The first step entails identifying the sources that could create olfactory nuisances. An initial examination of the region led to the identification of two potential sources: the WWTP and the pumping stations. The pumping stations are the elements of the sewage system that propel the sewage towards the WWTP. Six pumping stations were identified in areas near residential or vacation resort areas. The WWTP contains several potential emission sources. The purification process uses a biological treatment of waste water, which includes the usual stages: intake, aeration, and settling. The site presently includes the following installations: • • • • •

A waste water intake building (contains a grit chamber and skimmer) A buffer pond An aeration pool, adjustable according to the charge to be treated. Two clarifiers for settling. A sludge collection and treatment building.

Gaseous effluent deodorization systems have been installed on these components. The buffer pond, for instance, is deodorized by a photochemical system. The skimmer has also been equipped with a photochemical unit designed to reduce odor concentrations. Finally, the two sludge intake and treatment buildings employ chemical (acid-base) scrubbers designed to purify the building air prior to atmospheric release. Thus, a total of 14 odor emission sources were identified as potential generators of odor nuisances in the environment. The following figure shows these sources. Figure 2. Sources studied.

3

The inclusion of pumping stations into the assessment of odor problems allows a better understanding of the influence such installations may have on nuisance odor generation. While emissions from these stations are potential nuisance sources, it is entirely possible for the WWTP to be blamed instead, as the odors produced are comparable.

2.2 Odor monitoring system The OdoScan® odor management technology developed by Odotech for continuous odor monitoring was chosen for the site. The system allows operators to monitor the principal odor source(s) of their facilities. The system comprises a computer, software for real-time odor dispersion modeling, and a meteorological station. Samples are taken from the odor sources on the site and analysed by a dynamic dilution olfactometer in accordance with European standard EN 13725. The software parameters are programmed to model the atmospheric odor dispersion in real time and to display the resulting odor plume in odor units per cubic meter of air (o.u./m3). In order to calculate the odor plume, OdoScan®'s atmospheric dispersion model combines realtime data from the meteorological tower with the odor concentration values already measured, and displays the odor plume superimposed on the map of the site. This allows the operator to instantly visualize the impact of the odor, 24/7. The AERMOD dispersion model, a world reference for atmospheric dispersion, is used. The following figure illustrates the concept. Figure 3. The OdoScan® Concept.

4

The dispersion model incorporates the characteristics of emission sources and receptor points or surfaces, as well as the real-time meteorological data in order to calculate the odor concentration in ambient air at several user-defined locations.

2.3 Olfactometry Dynamic dilution olfactometry entails presenting a panel with sample dilutions created by a calibrated olfactometer for measuring the gas flow. The air-odor mixes are presented to panelists at odor sniffing ports. The objective is to determine the odor perception threshold for a gas sample. The odor perception threshold is defined as the odor level at which 50% of an odor panel perceives the odor, while the other 50% does not. By definition, the odor perception threshold is equivalent to one odor unit per cubic meter of air (o.u./m3). The number of dilutions of the odor sample necessary to obtain 1 o.u./m3 is equal to the odor concentration of the sample in odor units per cubic meter of air (o.u./m3). Figure 4. Dynamic Dilution Olfactometer.

Odor Panel

Sniffing Ports

The olfactometric analyses were performed using the ODILE® Dynamic Dilution Olfactometer, in compliance with European standard EN13725. The measurements obtained are objective and relate directly to the individual perception of the odor.

3. RESULTS AND DISCUSSION 3.1 Odor diagnosis The odor diagnosis consists in assessing the relative magnitude of each odor source. This initial analysis pinpoints the most significant sources in terms of odor flow rates. Table I shows the odor flow rates measured at the sources of the WWTP.

5

Table I: Site odor flow rates. Source Building Pretreatment deodorization Buffer deodorization Aeration Clarifier West Clarifier East Sludge deodorization

Odor flow rate (o.u./hr) 0.9 41 1.5 3.2 1.6 1 2.7

The following figure shows these flow rates as a percentage of the overall emissions of the site. Figure 5. Percentage of the overall odor flow rate due to each WWTP source.

The total odor flow rate of the Sivos des 60 bornes WWTP is estimated at 50 million odor units (peak), with all components operating. This number is at the low end of the range for this type and size of facility. Of this overall emission, it appears that the open surfaces (the three ponds) constitute minority emission sources at less than 11%. Inversely, pretreatment deodorization accounts for more than three quarters of all emissions. This predominance is due to two independent causes: firstly, the exhaust concentration was relatively high relative to the buffer pond and sludge deodorizations,

6

and secondly, the volume flow rate for the outlet is the largest on the site (10,000 m3/hr, nearly 40% of the site volume flow rate). This situation is typical. The deodorization of the pretreatment had never been identified as a potential source of nuisance odor. First of all, owing to its very nature, a deodorization outlet is rarely suspected to be the cause of a nuisance, as the deodorizing equipment is usually considered effective. In addition, the deodorization outlet is nearly 15 meters above ground, so that the odors it releases seldom cause a nuisance on the site itself, where operators might potentially recognize the odor. Indeed, the odor plume disperses high up, over long distances, and only hits the ground outside the site property limits. Operators have a natural tendency to identify ground installations as potential nuisance odor sources, since these sources make up their daily olfactory environment. From an overall perspective, the site produces 50 million odor units per hour at peak times, for a volume flow rate of just under 25,600 m3/hr. In terms of emission average, this yields 2,000 o.u. for each cubic meter of gas effluent, a relatively low output for a facility of this size. The following table shows the odor flow rates of the different pumping stations. Table II: Odor flow rates of the six pumping stations studied. Pumping station Les Demoiselles Top Loisir Les Sauges Les Immortelles La Chaussee Les Salins

Odor flow rate (o.u./hr) 1.40 0.53 2.13 0.15 0.05 0.17

The following observations can be made: • The total odor emissions of the stations come to 4.42 million odor units per hour. • Les Sauges, Les Demoiselles and Top Loisir are the greatest odor producers, accounting for 92% of the emissions of the six stations and a peak flow rate of 36%. • Plant elements with odor control equipment account for 8% of the odor emissions at a peak flow rate of 60%. The following figure summarizes these proportions.

7

Figure 6. Waste water flow rates (blue) vs. the associated odor flow rates (orange).

Comparison with the pumped water flows shows that the quantity of odor generated is not a function of the quantity of effluent treated. Indeed, the Immortelles pumping station pumps the largest quantity of waste water but produces one of the lowest odor emissions. Odor flow rate is primarily a function of the time the pumped waste water spends in the system, as well as the nature of the effluent. Depending on the origin of the effluent, it may be more or less charged with odor-producing molecules. The odor diagnosis results show that the pumping stations are a potential source of environmental nuisance. Since the odors from the pumping stations may be readily associated to those from the WWTP, they must be included in the analysis of odor complaints, and thus in the continuous odor monitoring system. The odor flow rates obtained from the odor diagnosis were used as reference values in the atmospheric dispersion modeling software of the OdoScan® system.

3.2 Real-time odor assessment The real-time monitoring of odors was used primarily for identifying the origin of odors when complaints were reported. The use of this system allowed operators of the WWTP, as well as municipal employees in charge of pumping stations, to react as soon as a risk of odor nuisances was detected, responding directly at the source responsible for the problem. As a result, a significant reduction in odor complaints was observed during the 2008 summer season. The following table shows the maximum values at different locations of interest within the study zone. Note that the des Demoiselles campground entrance may be subjected to high odor concentrations.

8

Table III: Maximum odor concentration values estimated by simulation. Les Les La Location Le Petit Bec Les Becs Demoiselles Begonias Caillauderie Max. Conc. 14.5 4.5 5.7 7.7 4.5 (o.u./m3) Time and 17:00 1:00 21:00 2:00 17:00 date 04/12/2008 09/13/2008 08/28/2008 04/13/2008 08/29/2008 The values were nonetheless relatively low. While the odor perception threshold is by definition 1 o.u./m3, the odor recognition threshold is usually set at 5 o.u./m3. At this threshold, people can identify the nature of the odor as well as its source, if they know about it. Thus, the residential zones of des Becs and des Begonias never reached the odor recognition threshold over a oneyear period. The modeling of odors is largely dependent on meteorological parameters. The following figure shows the degree to which different meteorological parameters can have a direct impact on the dispersion of odor, and therefore on its perception by those concerned. Figure 7. Odor plumes as a function of different meteorological conditions.

The wind direction is comparable but the extent of the plume is completely different for the two scenarios. Note that in the left-hand diagram, the odor plumes are very narrow. The complaint risk is much lower than for the right-hand diagram, where the odor plumes are very wide. As a result, a long stretch of the beach is subject to odors. The usefulness of this type of equipment lies in enabling the start-up of specific odor control measures when problems arise (chemicals in the sewer system, boosting the scrubber treatment, increasing fresh airflow in buildings, etc.). The on-demand reduction of odors allows the operator to avoid the costs of continuous odor treatment while ensuring that the quality of life of residents near the sources remains unaffected.

9

3.3 IMPACT STUDY The impact study consists in consolidating the entire real-time monitoring data set to perform analyses of the exposure of nearby residents to odors over a longer period.

3.3.1 Hourly maxima This situation refers to the worst combination in terms of meteorological conditions and emissions of odorous effluent. Examination of the Figure reveals that this situation raises the odor concentration within a radius of 1000 meters to a level of 4 odor units per cubic meter. Figure 8. Maximum concentrations reached in the vicinity of the site.

This means that it is possible to detect and recognize the plant site odors within a 1 kilometer radius in the event of particularly unfavorable meteorological conditions. The most unfavorable impact thus appears to be substantial under maximum operating conditions, at the des Demoiselles campground, the point nearest the site. In the event of poor weather conditions and intense plant operation, all of the other locations monitored could be subject to complaints. The following figure shows the most unfavorable situation for the pumping stations only.

10

Figure 9. Maxima reached for the six pumping stations surrounding the site.

The des Sauges and des Demoiselles pumping stations are the sources with the strongest impact. While this was predictable from the estimated odor flow rates, it yields a perception threshold (1 o.u./m3) range of up to 300 meters for the two points. The other two locations were negligible over their range (<100 meters).

3.3.2 Threshold breaches Two threshold breaches were investigated. The first concerns the breach of the perception threshold, i.e. the percentage of time for which more than one-half of the population detects an odor coming from the site. See Figure 10.

11

Figure 10. Perception threshold (1 o.u./m3) breach as a percentage of time.

Mapping the perception threshold (1 o.u./m3) boundary yields 1% of the time, i.e. approximately 90 hours over the whole year, 900 meters from the site. Note that identification of the odor is not possible at this concentration level. The 5 o.u./m3 threshold -- the concentration necessary for odor recognition -- is mapped in the following figure. Its extent does not exceed 1% of the time (i.e. 90 hours per year) at 300 meters. Figure 11. Recognition threshold (5 o.u./m3) breach as a percentage of time.

12

The following figure deals only with the pumping stations. Only Les Sauges et Les Demoiselles are potentially detectable by their odors at least 2% of the time within a radius of a few dozen meters. Figure 12. Perception threshold (1 o.u./m3) breach as a percentage of time.

The following table shows the percentage of time where these thresholds were exceeded for each location studied. Table IV : Threshold breaches as a percentage of time Location Perception threshold Recognition threshold

Les Demoiselles

Les Begonias

La Caillauderie

Le Petit Bec

Les Becs

11.4

1

0.9

3.3

0.5

2.7

0

0

0

0

The threshold breach numbers reveal a limited impact on the vicinity of the site. In spite of its relative proximity to the site, the Petit Bec subdivision shows a very low potential complaint frequency. On the other hand, that frequency rises to 3% of the time for the neighboring des Demoiselles location, i.e. 262 hours per year.

13

SUMMARY - CONCLUSION The odor monitoring system provides for three types of analyses: 1. The ranking of the odor sources responsible for nuisances. 2. Real-time assessment of where nuisances are coming from. 3. Assessment of the degree of nuisance generated over time. This triple analysis allows industrial site operators to better understand the nature and the causes of odor nuisances for which they are responsible in the vicinity and to implement proper short and long-term solutions. As a result of the monitoring, it was decided to implement odor control solutions at the des Sauges and des Demoiselles pumping stations, and to optimize the odor control system already in place at the pretreatement facility. These improvements resulted in a significant reduction in odor nuisances. In addition, the daily use of the OdoScan® system makes it possible to minimize the number of complaints received by the Town Halls and the WWTP operator.

REFERENCES 1. COMITÉ EUROPÉEN DE NORMALISATION - CEN 13725, Air Quality -Determination of Odour Concentration by Dynamic Olfactometry. 2. CUM. Mesure du nombre d'unités d'odeur (olfactométrie dynamique). Montréal, Communauté Urbaine de Montréal, Service de l'environnement, Direction de l'assainissement de l'air et de l'eau. 1994. 3. PAGÉ, T. et GUY, C.; Odor dispersion modeling. Air & Waste Management Association’s 90th Annual, Toronto, ON, 1997.

KEYWORDS Odor Monitoring, Waste Water Treatment Plant, Olfactometry, Odor Diagnosis, Odor Emission, OdoScan, Odotech, Odor Flow Rate, Real-time odor assessment, Odor Plume, Impact Study

14