Influence of the variability of the odour emission rate on its impact range: a case study of the selected industrial source

Odour concentration measurements in a chosen industrial source were made in this study using the method of dynamic olfactometry. The two different scenarios considered the variation of the odour emission rate as input for the dispersion model were compared for the period 2017 (before installation of the equipment for gas treatment) and 2018 (after implementation of purifying technologies). In this paper the odour impact range was determined by applying model calculations conducted in the Polish reference dispersion model – OPERAT FB software for the grid size 2 x 2 km. The conducted research shows a significant improvement in the odour impact range of chosen industrial source in year 2018 compared to 2017.


Introduction
The odour nuisance is related to the emission of odours into the atmosphere. The three main groups of emission sources of odours can be distinguished. They include agriculture and animal husbandry, municipal waste management and industry [1]. Analysis of problems and citizens' complaints in Poland, related to the odour impact assessment, identified: 16% of the complaints lodged against industrial activities, 34% -about agriculture and breeding, 39% -as result of municipal management and 11% related to other sources [2]. Industrial sources of odour nuisance are characterized by the greatest variety. In this group, a wide range of industrial facilities such as agriculture and food processing industry or chemical industry can be distinguished. Due to the large diversity of products and raw materials, such as dry products as plastics, pharmaceuticals, paints, varnishes, rubber products, fertilizers, acids, products, and semi-finished products from crude oil, the odour nuisance associated with the activity of the chemical industry belongs to the most diverse [1].
In general, each of the industrial sources of odours differs both in terms of emitted odorant types and their amount. The amount of odours emission from production operations depends on a number of factors and remedies used to prevent odour nuisance. Industrial facilities are often located in the immediate vicinity of residential areas; therefore, it is necessary to use appropriate techniques to prevent odour emissions into the atmosphere. Among the techniques allowing for the elimination of odorous pollution, the following techniques can be distinguished: combustion, adsorption, absorption, biological methods or deodorization [1]. The use of an appropriate deodorization system allows for effective reduction of the odour emissions of the facility, as well as to improve the quality of life of the local residents.

Purpose and methodology of measurements
The odour emission measurements are carried out in accordance with the methodology described in VDI 3880:2011 [3] and PN-EN 13725:2007 [4] guidelines. Sampling campaigns are mainly performed during rainless weather that will guarantee the efficiency of measurements and repeatability of results. A vacuum sampler (marked CSD30 sampler) and PET-bags, made of chemically inert material, are generally used during samplings. All elements of collecting set are made of odourless materials that do not absorb odorants. According to the recommendations, the bags are previously conditioned. The measurements of weather data including temperature, humidity, and pressure in the sampling day are carried out using a TESTO 435-2 probe system. During the sampling time, the study facilities must be worked without any deflections.
After collection, without delay, the samples of odours are transported to the Olfactory Laboratory. Then, the odour samples are evaluated and quantified to determine odour concentration. Odour concentration measurements are made using the dynamic olfactometry method in accordance with the PN-EN 13725:2007 standards: 'Air quality -determination of odour concentration by dynamic olfactometry' [4]. The odour measuring equipment includes a dilution device -the four-station olfactometer TO8 with the necessary attachments. The obtained data is calculated on the basis of sensitivity tests determined from the geometric mean of all individual measurements. The unit of the odour concentration is reported as the European odour unit per cubic meter (ouE/m 3 ).

Study site
The industrial facilities investigated in this study are located on the outskirts of a compact urban agglomeration in Poland. In the immediate vicinity, on the northern side of the Plant, an undeveloped area and further the railway line is located. On the eastern side, the Plant is adjacent to another industrial facility. Approximately 1 kilometer to the south of the Plant, there are residential area and services, industrial development and inbuilt areas. There are housing developments at a distance of about 30 m from the borders on the west side of the Plant. At the distance of the emission impact range up to 850 m (equal to fifty times of the emitter height) [5] there are no protected objects: national parks, health resorts, monuments/historical treasures or other areas being the subject of the protection in accordance to Polish Nature Conservation Low [6]. Waste gases from plant production activity are discharged to the main emitter that height is 17 m; the release cross-section of the chimney is particularly 2.5 x 1.2 m. After the system modification (after measurements from 2017), the exhaust air from the production part is centrally purified using a multi-stage adsorb vessel system.

Odour dispersion methods and models
The mechanisms of dispersion of odorants in the atmosphere are the same as the air based transport of other pollutants. Mathematical models of odour dispersion in the air are used to evaluate the environmental impact of odour-causing substances for various emission release scenarios. The accuracy of the odour dispersion model prediction depends on the multiple input parameters: odour emission data (exit velocity, plume rise, temperature, etc.), sources and surface characteristics (surface roughness, local topography, nearby buildings), meteorological conditions and pollution transformation in the atmosphere (wind speed, stability, mixing height, wind direction) [7]. Based on the dispersion model, it is possible to calculate the odour emission distribution from individual sources as well as area sources. The odour concentrations at receptor points in the study area can be evaluated by using available dispersion models. Gaussian-type models are the most common dispersion models used in the atmospheric simulation of pollution transport [8]. In general, the most widely used models for simulation of odorous compounds are Gaussian plume (e.g. AERMOD) and Gaussian puffs (e.g. CALPUFF) distribution software. Each one has its own advantages and disadvantages associated with the steady-state approximations and the vertical particle movement due to gravity during the travel time [9]. The Operat FB model developed based on Gaussian plume, is mainly used in Poland for predicting odour concentration and calculating odour emission rates [5]. The model is based on knowledge and the assumption of the atmospheric transport of pollution, especially relating to the constant and uniform emission rates; constant parameters of the wind and speed direction; processes of vertical and crosswind diffusion; and the terrain domain.

Emission calculations
The odour emission from the installation was calculated as mean odour concentration and exhaust gas flow rate on the main/cetral emitter (CE). The gas flow was specified for the  Table 1.
The results of odour concentration measurements from 2018 show the high decreasing in odour level by 11896 ouE/m 3 as compared with the samplings applied in 2017. The reduction of the odour concentration at the outlet of the emitter is caused by the introduction of modifications in the deodorization installation of waste gases from the Plant.   [5].
Modelling criteria used in the study included the basic values of emission source parameters (emitter height and diameter, odour emission rate), as well as the current dispersion conditions (meteorological data and topographical features).
Modelling of odours dispersion was performed based on the following scenarios: • the analysis performed in normal mode at the maximum system capacity before installing the gas deodorization system (on the year 2017).  Calculations of the odour impact range were carried out in grid format receptors with selected size and step of each grid cells and coefficient of aerodynamic roughness. The grid size was determined to be equal to 2000 m x 2100 m with a grid step at 50 m. The coefficient of aerodynamic roughness identified in the area within 50 times radius of the emitter height [5] was 0.353 m.

Results of the odour dispersion modelling
The OPERAT FB -a Polish reference dispersion model (according to Regulation of the Minister of the Environment of 26 January 2010 on reference values for certain substances in the air [5]) was applied in the analysis of the odour impact range. The environmental detection threshold for odours as a mixture for residential areas under the Polish conditions  The results of the modelling of maximum odour concentration values outside the Plant border at additional points marked as residential buildings are compared in Table 4  The frequency of the exceeding of a given one-hour odour concentration threshold in the year 2017 and 2018 is shown in figure 1. Comparing the left and the right panel in fig. 1, the large differences in the impact distances of odour emission before and after modification purifying system can be seen. The exceedance level of 1 ouE/m 3 and 3% is much more pronounced at calculation grid for 2017, in distance towards the north exceeds 400 m and approximately 500 m to the east of the Plant boundaries. In 2018, the frequency of exceedances of one-hour odour concentration has not occurred. Modelling of odour  In case of the odour dispersion modelling the predicted concentrations in the environment were used for designating areas and populations affected by the odour nuisance. At the selected points within the study grid, where the residential buildings were located, the significant improvement in the air quality was observed in 2018 compared to 2017. The highest value of average annual maximum concentrations of odours was determined in the point B4, at the level of 37.821 ouE/m 3 and 0.358 ouE/m 3 in 2017 and 2018, respectively. According to the average concentrations of odours, the highest value was evaluated in the point B5. The average odour levels were found to be 0.3229 ouE/m 3 in 2017, while in 2018 -0.0075 ouE/m 3 . According to the emission ranges, results show that the development of solutions for the improvement of the existing production and deodorization conditions may prove helpful in limiting the odour dispersion.

Summary and conclusions
Most complaints related to chemical odours are associated with industrial facilities that should continue to improve its odour control performance. The emission of odours into the atmosphere from the industrial plants, especially in urban areas, can cause problems related to odour nuisance, and thus even negative health effects. In order to asses the odour impact before and after an industrial plant upgrade, the odour concentration measurements by dynamic olfactometry method was used in this study. The prediction of odour concentrations determined by the Polish reference dispersion model pointed to the direct impact of the modernization of production lines and the deodorization system in the analysed Plant. The model calculations of the odour impact range with respect to the emission values show that the distance of odour dispersion decreases with the emission decrease. The research results indicate that odour evaluation and dispersion modelling are valid tools in the evaluation of odour control improvements implemented by the operators of industrial facilities. The study was co-financed within the framework of the project No. 0402/0095/18 with the specific subsidy granted for the Faculty of Environmental Engineering Wroclaw University of Science and Technology (W7) by the Minister of Science and Higher Education to conduct research and development work and related tasks contributing to the development of young scientists and doctoral students in 2018/2019.