The Impact of Objects with a Potential Odour Nuisance on the Life Comfort of the Urban Agglomeration Inhabitants

Abstract

Odour nuisance is one of the main causes of environmental complaints. People exposed to long-term odorants may experience headaches, nausea, difficulty concentrating, loss of appetite, stress, insomnia, and discomfort. Some chemical compounds, besides unpleasant odours, can cause adverse symptoms, diseases, and even death in human bodies. One of these compounds that make up BTEX (benzene, toluene, ethylbenzene, xylene) is benzene, which is present in the environment, mainly in the air, because of emissions from traffic, the petrochemical industry, and combustion processes. Factories, such as refineries and petrochemicals, that form part of some urban agglomerations constitute extensive industrial facilities. This paper presents the survey research results in an urban agglomeration, which enabled, among others, an indication of significant sources of odour nuisance and the areas most exposed to this nuisance. In addition, an analysis of residents’ complaints about odour nuisance over a 10-year period was carried out, which showed, on the one hand, the variability of the number of complaints and, on the other hand, the areas of the city where there were the most complaints. This work aims to evaluate the problem of odour nuisance in an urban agglomeration and to identify its causes through the analysis of residents’ complaints and the results of surveys.

1. Introduction

1.1. Literature Review

Odorants are chemical compounds that stimulate the human olfactory system and, as a consequence, cause odour nuisance, thus affecting the comfort of life [1], the quality of work, and the health of workers [2] as well as the health of those exposed [3]. The emission of odorants is significant for people living near municipal, industrial, and agricultural facilities, which are potential sources of odorant emissions [4,5].

Odour nuisance is one of the leading causes of environmental complaints [6,7,8,9]. It generally does not threaten health and life. Still, it can be unpleasant and annoying, which is very important from the viewpoint of environmental and social indicators of the circular economy concept [10,11,12,13]. People exposed to long-term odorants may experience headaches, nausea, difficulty concentrating, loss of appetite, stress, insomnia, and discomfort [14,15,16,17]. Some chemical compounds, in addition to unpleasant odours, can cause adverse symptoms, diseases, and even death in human bodies. One of these compounds that make up BTEX (abbreviation for the aromatic compounds benzene, toluene, ethylbenzene, and xylene) is benzene, which is present in the environment, mainly in the air, because of emissions from traffic, the petrochemical industry, and combustion processes. The total metabolism of benzene lasts for many hours from the moment it is introduced into the body by the respiratory tract, and benzene remains in the body for a long time [18]. Benzene is one of the most used chemicals in many countries, including China [19]. Chinese medical literature is full of reports of benzene overexposure and benzene poisoning [20].

Factories, such as refineries and petrochemicals, that are part of some urban agglomerations constitute extensive industrial facilities [21]. Their functioning is associated with the emission of various types of volatile organic compounds into the atmosphere, mainly from production processes, tanks for storing products and intermediate products, and waste storage areas [22]. In the petrochemical industry, most organic compounds come from petroleum fractions, specifically from several basic hydrocarbons, such as methane, ethane, propane, benzene, toluene, and xylene [23]. The process of refining crude oil involves the physical, thermal, and chemical separation of crude oil into major distillation fractions. The primary industrial products, according to the EPA [24], are divided into three main categories:

• Fuels (motor gasoline, diesel and distillates, liquefied petroleum gas, jet fuel, fuel oil residues, naphtha, and coke);
• Finished non-fuel products (solvents, lubricating oils, greases, kerosene wax, jelly, asphalt, and coke);
• Chemical raw materials (diesel, ethane, propane, butane, ethylene, propylene, butylene, butadiene, butadiene, toluene, xylene).

The sources of emissions at refineries are leakage, combustion of fuels in process heaters, and all other refinery processes [24]. In addition to VOC emissions, oil refineries are also a source of emissions of such compounds as sulphur dioxide, reduced sulphur compounds, hydrogen sulphide, thiols, carbon monoxide and nitrogen oxides, as well as particulate matter [25].

In their work, Song et al. [26] proposed and developed a modelling system based on multi-source detection of air pollution in a transport agglomeration and its heterogeneous features, which can be used to monitor air condition.

Research on odour nuisance shows that people’s adverse reactions to odours in their surroundings are a very complex problem. This is related to factors such as residents’ current state of health, environmental commitment, and social activity [27,28,29]. To mitigate the negative impact of potential odour nuisances on the urban environment, it is necessary to reliably assess odour emissions and determine their size and long-term effects [30].

Despite advances in assessment techniques [31,32,33] and studies on strategies to minimise odour emissions [34,35], this issue remains unresolved.

Lubinaah et al. [36] published preliminary results of studies on the health effects of the refinery located in Oakville, Ontario. The results showed that, in the summer of 1992, 56% of survey respondents noticed refinery odours at least once a month, while in the summer of 1997, fewer people said so, i.e., 47%. In addition, 35% of respondents who noticed odours during the summer months of 1997 indicated that the odour effect had improved over the past year, and 47% reported an improvement over the past five years. Despite the reduction in adverse effects, reports of cardinal and general symptoms in adults and children remained virtually unchanged.

Song et al. [37] developed a solution using machine learning based on pollution measurements and urban features. This solution may help find a correlation between odour exposure and annoyance.

Two types of studies can be conducted to assess the impact of odours on the living comfort and well-being of people living in areas close to sites that emit odours: survey studies [38,39,40] and field studies [38,41,42,43,44,45,46].

Prior to the internet age, data were primarily gathered through methods like questionnaires, interviews, panel surveys, observational research, and telephone surveys. Telephone surveys are an objective method of data collection, but this trend is constantly changing [47]. The transition from telephone surveys to self-managed surveys is primarily dictated by the ability to control the costs of conducting them and to maintain or improve quality, or both. With the rise of the internet and advancements in technology, people transitioned entirely to using online or modern survey methods, such as using mobile applications as well as online platforms.

1.2. Bibliometric Analysis

The initial literature search conducted on 3 August 2023 was based on a current search for the terms “Odour” AND “complaint” AND “survey” in titles, abstracts, and keywords in the Web of Science academic database. The data range was set to “published all years to the present”. The document type was set to “scientific articles and review articles”, and the language was set to “English”.

Figure 1 contains a chart of times cited and publications over time—Odour AND complaint AND survey (Web of Science database, field: all).

As many as 35 articles fell into the subject area of environmental science compared with 20 in environmental engineering, 18 in public environmental occupational health, 9 in water resources, 5 in civil engineering, 4 in green sustainable science technology, 4 in environmental studies, and 3 in construction building technology. Among the articles that are most cited are the following:

• Frasnelli and Hummel (2005) [48]—cited by 226 authors; paper and pencil questionnaire.
• Bartha et al. (1999) [49]—cited by 179 authors; paper and pencil questionnaire.
• Hillert et al. (2002) [50]—cited by 119 authors; e-mail questionnaire.
• Brancher et al. (2002) [51]—cited by 112 authors.
• Hayes et al. (2014) [52]—cited by 84 authors.
• Merkonidis et al. (2015) [53]—cited by 47 authors; paper and pencil questionnaire.
• Ames et al. (1993) [54]—cited by 36 authors; paper and pencil questionnaire.

The connections between the analysed documents and highly cited articles are presented in Figure 2, while the top 100 similar works are presented in Figure 3. Among them, the most cited article strictly focused on the survey of odour complaints is Hayes et al. [53].

According to the publications on olfactometry-related surveys, those methodologies are widely utilised to assess a broad range of environmental impacts; however, challenges include designing appropriate questions that address health effects, sensory irritation, and community-level awareness regarding odour issues. The authors highlight that designing surveys can be challenging in terms of minimising bias, and careful attention to sampling procedures is essential to accurately represent the population. Surveys enable a more structured and extensive analysis than is typically possible with most qualitative research methods—face to face or telephone interviews. The surveys conducted in the most highly cited articles were primarily administered using the pen-and-paper method, generally without accounting for detailed locational aspects or the variability of meteorological conditions.

The aim of this work is to evaluate the problem of odour nuisance in an urban agglomeration and to identify its causes through the analysis of residents’ complaints and the results of surveys.

The scope of this study was to conduct a survey of odour impacts in an urban agglomeration, with an indication of potential sources of these impacts. The purpose of analysing the results of these surveys was, among other things, to identify potential sources as well as to determine the areas of the urban agglomeration most vulnerable to odour nuisance.

In addition, an analysis of odour nuisance complaints over the past 10 years was conducted to characterise the impact zones.

The innovation in this research is the synergy of two types of odour impact evaluation—a survey and a set of complaints—combined with GIS analysis. During the present research, an elaborate, proprietary questionnaire on the study of odour impact in a large urban agglomeration was developed, along with linking it to the results of ten years of complaints from residents of the agglomeration.

2. Materials and Methods

2.1. Characteristics of the Agglomeration Under Consideration

This study covered the town of Plock, located in Poland, and its neighbouring towns. As part of the research, a survey was conducted among the residents as well as an analysis of complaints of the residents living in Plock and neighbouring towns about odour nuisance. Figure 4 shows the location of these areas.

Plock is a town with country rights located in the centre of Poland, approximately 110 km west of Warsaw, the capital of Poland, in the Masovian Voivodeship. The area of the city is approx. 88 km². At present, Plock has approximately 120 thousand inhabitants (as of 2019), of which almost one-fifth are under the age of 19 years [56].

The fuel, energy, and chemical industries dominate the city. The Polish Oil Company ORLEN S. A, Płock, Poland. is headquartered here, which is a powerhouse on the fuel market in Poland and one of the largest companies of this type in Central and Eastern Europe; PERN—Oil Pipeline Operation Company, Płock, Poland is located here; Basel Orlen Polyolefins, Płock, Poland is also active in the chemical industry. The food, machinery, clothing, shipbuilding, and construction and assembly industries are also developing well in Plock [57].

The field survey carried out in Plock, the results of which are presented in the work by Wiśniewska and Kulig [58], allowed for the identification of three objects with a potential odour: refinery, sewage treatment plant, and landfill and mechanical–biological treatment plant of municipal waste located approx. 13 km from the city centre.

2.2. Previous Studies on Air Pollution

Since 2018, measurements of gaseous and dust pollutants have been carried out in the city of Plock. The reason for initiating this study was a series of complaints from city residents about odour nuisance. As part of the measurements, a network of samplers for measuring gaseous pollutants and low-cost sensors for measuring dust pollutants was designed in twelve locations [59]. BTEX substances were collected using active and passive samplers. Each sample consisted of a stainless-steel tube filled with a solid absorbent polymer. The tube was finished with brass swagelock caps. The type of sorbent used in the sampler was selected appropriately for the specific type of chemical compounds being determined. In the case of passive sampling, a diffuser with a mesh protecting against the ingress of solid particles was installed instead of the plug. Adsorbed hydrocarbons were recovered by a desorption system connected online to a chromatograph, and their concentrations were analysed using GC/FID or GC/MS. The measurement results were related to the amount of valid data determined in accordance with the Guidance on the Annexes to Decision 97/101/EC on Exchange of Information as revised by Decision 2001/752/EC [60].

2.3. Surveys

As part of this study, a survey was conducted among the residents of one of the urban agglomerations in Poland, Plock.

The survey was developed in the Google Forms application and was available to anyone with a link leading to the form. The link was published on the City Hall website. The information campaign was implemented through the social networks of the city of Plock (Facebook, Linkedin) and local and city residents’ groups. The announcements to the above respondents included full information about the purpose of the survey as well as a link leading to an electronic form containing 11 questions. Three of them were open-answer questions, two of them were yes/no answers, and the rest required the selection of one of at least three answers (

Table 1. Survey on odour nuisance in urban agglomerations.

No.	Question	Answer 1	Answer 2	Answer 3	Answer 4	Answer 5	Answer 6	Answer 7	Answer 8
1.	Do you have an odour nuisance in Płock or its surroundings?	YES	NO	-	-	-	-	-	-
2.	In which area of Płock or the surrounding village do you feel an odour nuisance? Enter the street name in Płock or the name of the neighbouring villages.	OPEN ANSWER
3.	At what time of day/night do you feel the odour nuisance? You can tick more than one answer.	12:00 a.m.–3:00 a.m.	3:00 a.m.–6:00 a.m.	6:00 a.m.–9:00 a.m.	9:00–12:00 p.m.	12:00 p.m.–3:00 p.m.	3:00 p.m.–6:00 p.m.	6:00 p.m.–9:00 p.m.	9:00 p.m.–12:00 a.m.
4.	How to estimate the intensity of the perceived fragrance?	I do not feel the odour	Very weak	Weak	Clearly	Strong	Very strong	Extremely strong	-
5.	How do you assess the hedonic quality of the perceived fragrance?	Unpleasant	Indifferent	Pleasant	-	-	-	-	-
6.	How do you estimate the intensity of the odour nuisance?	No odour	No nuisance	Small nuisance	Nuisance	Large nuisance	Extreme nuisance	-	-
7.	Can you determine what is causing your odour nuisance?	OPEN ANSWER
8.	In which wind direction do you most often experience an odour nuisance?	N	NE	E	SE	S	SW	W	NW
9.	How often do you experience odour nuisance in Płock or near Płock?	Never	At least once a month	At least once a week	At least once a day	Almost constantly	-	-	-
10.	Have you ever complained about an odour nuisance?	YES	NO
11.	If you have complained about odour nuisance, please write to whom.	OPEN ANSWER

). The form was available to respondents for three months.

In survey studies addressing odour nuisance in large urban agglomerations, several potential sources of uncertainty can be identified, especially when the questions focus on subjective perceptions, such as odour intensity and hedonic quality. Key factors introducing uncertainties include the following:

1.

Subjectivity in odour perception and response

• Individual differences: People respond differently to the same odours due to variations in olfactory sensitivity, personal preferences, or past experiences with certain odours.
• Diverse interpretive standards: Terms like “intensity” or “hedonic quality” may be interpreted differently by respondents, leading to inconsistent responses.

2.

Timing and location of survey administration

• Variable atmospheric conditions: Wind speed and direction, temperature, and humidity can impact odour dispersion and, therefore, influence perceived odour intensity.
• Temporal comparability: Conducting surveys on different days or at various times of day can lead to varying responses due to changes in odour intensity. Odor conditions may differ in the morning, evening, or on weekends when urban traffic patterns vary.

3.

Respondent location within the agglomeration

• Proximity to odour sources: Respondents situated at different distances from odour sources may perceive varying levels of intensity, potentially affecting survey outcomes.
• Local environmental conditions: Buildings, terrain topography, and local infrastructure can alter airflow and odour distribution within the city.

4.

Social and cultural factors

• Influence of social opinion: Respondents may be swayed by the views of neighbours, media reports, or local discussions, impacting their odour assessments.
• Social norms and expectations: In large agglomerations, tolerance for industrial or urban odours may vary depending on social status or quality of life expectations.

5.

Question formulation in the survey

• Ambiguous wording: Questions may be interpreted differently by respondents, particularly when addressing subjective concepts such as “hedonic quality”.
• Suggestion effect: Question phrasing or placement within the survey may affect responses by implying an expected range of odour intensity or quality.

Accounting for these potential sources of uncertainty in the analysis will enable a more accurate interpretation of the data and consideration of limitations related to the subjectivity and variability of odour perception in large urban agglomerations. To minimise these uncertainties, we applied the following methods to enhance the accuracy and reliability of the results:

• Standardisation of the questionnaire and instructions—clear definitions of terms: we explained terms such as “intensity” and “hedonic quality” of odour to ensure that all respondents interpreted them consistently. We provided clear guidelines for odour assessment to minimise individual interpretive differences.
• Use of subjective rating scales and quality indicators with descriptive, well-defined scales, which are easier to understand and allow for response standardisation.
• Control of survey location: questions about weather conditions as well as location of odour nuisance.
• Minimising the impact of social and cultural factors: ensuring response anonymity to reduce the influence of social norms and social pressure on respondents’ answers.

Within the framework of this study, 120 responses were obtained from residents from the entire agglomeration, making the results representative for a confidence level of 90% and a maximum error of 8%.

2.4. Residents’ Complaints

To obtain accurate information on the occurrence of complaints about odour nuisance in the city of Plock, an application was sent to the City Hall with a request for access to this environmental information covering the period from January 2012 to 2021. The information received included an indication of the region of the city concerned by the complaint. Visualisation of the spatial distribution of the number of complaints in the study area was made using ESRI ArcMap 9.3 software. The maps show the cumulative values—for the years 2012–2021—and the spatial distribution of the complaints in each year. The cumulative number of complaints was also visualised using a histogram, while the number of complaints in each year was visualised using a box-and-whisker plot. Both charts were also generated in ArcMap 9.3. software.

2.5. Statistical Analysis of the Results Obtained

The results of the survey and the quantitative and area data obtained from the City Hall within the framework of public information were subjected to static analysis using R 4.3.0 software.

As part of descriptive statistics, the following parameters were calculated: median, mean, and standard deviation; in addition, minimum and maximum values were indicated. When analysing the results of the questionnaires, the frequencies of the obtained values were calculated. To compare the means of the two groups—odour annoyance and odour intensity—a paired sample t-test with 95.0% confidence interval was performed. To verify whether there were statistically significant differences in the mean values of odour intensity and odour annoyance, with wind direction as a fixed factor, an ANOVA test was used. To check the heterogeneity of the variances, the Levene test was performed, while to determine which mean was statistically significantly different from the others, we used post hoc, Dunnett, and Games–Howell tests. In cases where basic ANOVA tests were not met, the non-parametric Kruskal–Wallis test was used.

3. Results and Discussion

3.1. Analysis of Previous Study Results

Measurements carried out using the reference method have shown that the most frequently identified cause of smog episodes is unfavourable meteorological conditions, i.e., low- or very-low-speed winds from the northern sector, causing the accumulation of emissions of toluene, benzene, and other VOCs, coming mainly from the refinery but also from other much smaller industrial plants located in the city, traffic pollution, and, during the heating season, also emissions from the combustion of solid fuels. For benzene, there is a limit value of 5 µg/m³ for the annual average, in accordance with the Directive on ambient air quality and cleaner air for Europe (2008/50/EC) [61], as well as a reference value set for an hourly average (30 µg/m³), according to the Polish Regulation of the Environment Minister on reference values for certain substances in the air [62]. The presented results of the 2021 study [59] showed six episodes of higher one-hour concentration than the reference value and two in the case of benzene.

3.2. Analysis of Residents’ Complaints

Figure 5 and Figure 6 present the spatial distribution of complaints reported to the City Hall in each time. Figure 5 contains the spatial distribution of the number of complaints in the study area in the years 2012–2021, and Figure 6 containes the spatial distribution of complaints in individual years.

The analysis of Figure 6 shows that the largest number of complaints was recorded in the western (Scarpa housing estate) and northwestern (Lukasiewicz housing estate) parts of the city (south of the refinery). In the case of Figure 7, it can be observed that the highest number of complaints was reported in 2018, and the lowest number was reported in 2012. Considering the situation in 2018 and 2020, between October and November, the refinery carried out works during which one of the chimneys was shut down and a reserve emitter was activated. Any malfunctions and repairs (modernisation) may cause a stronger odour effect than under normal operating conditions [63].

Figure 7 shows the histograms of the cumulative number of complaints per year. The lines in the bar graph correspond to the streets/locations where the complaints were found each year.

Figure 8 shows morning-moustache graphs—number of complaints per year.

Figure 7 and Figure 8 show that the highest number of complaints was recorded in 2016 (160) and 2018 (228).

3.3. Analysis of Survey Results

Figure 9 and Figure 10 contain distribution plots of odour annoyance and odour intensity.

Figure 11 shows a polar chart showing the intensity and nuisance of the odour effect as a function of the wind direction.

Table 2. Descriptive statistics of odour annoyance and odour intensity splinted by wind direction.

	Odour Annoyance								Odour Intensity
	E	N	NE	NW	S	SE	SW	W	E	N	NE	NW	S	SE	SW	W
Valid	39	155	107	127	56	71	40	50	39	155	107	127	56	71	40	50
Median	4.00	4.00	4.00	4.00	4.00	4.00	4.00	4.00	5.00	5.00	5.00	4.00	4.00	5.00	5.00	5.00
Mean	3.87	4.07	4.20	4.04	4.00	4.21	4.25	4.06	4.46	4.61	4.75	4.40	4.32	4.74	5.02	4.58
SD	0.80	0.74	0.72	0.72	0.76	0.67	0.67	0.71	1.21	1.03	1.15	1.04	1.01	1.09	0.73	0.99
Min.	1.00	0.00	2.00	2.00	2.00	3.00	2.00	2.00	0.00	3.00	1.00	2.00	3.00	3.00	3.00	3.00
Max.	5.00	5.00	5.00	5.00	5.00	5.00	5.00	5.00	6.00	6.00	6.00	6.00	6.00	6.00	6.00	6.00

contains descriptive statistics of odour annoyance and odour intensity splinted by wind direction.

Figure 12 provides a summary of the answers to the question “How often do you experience odour nuisance?” according to the mean values of odour annoyance and odour intensity, broken down by wind direction.

Table 3. Responses to questions on the frequency of the experience of odour nuisance.

How Often Do You Experience Odour Nuisance?	Annoyance	Frequency	Percent
At least once a day	3	9	6.12
	4	90	61.2
	5	48	32.6
	Total	147	100.0
At least once a month	2	6	7.50
	3	28	35.0
	4	43	53.7
	5	3	3.75
	Total	80	100.0
At least once a week	0	1	0.249
	2	8	1.99
	3	78	19.4
	4	238	59.3
	5	76	18.9
	Total	401	100.0
Never	0	4	66.6
	1	2	33.3
	Total	6	100.0
Almost always	3	9	5.84
	4	58	37.6
	5	87	56.4
	Total	154	100.0

provides a summary of the answers to the question about frequency of odour nuisance.

The analysis of Figure 10 shows that most respondents indicated a north wind, N (32%), a northwest wind, NW (26%), and northeast wind, NE (22%), for both odour annoyance and odour intensity. The analysis of Figure 13 shows that the odour nuisance felt by the inhabitants of the agglomeration is related to the occurrence of winds mainly from the N, NW, and NE directions. Those who reported high odour nuisance and significant intensity pointed to the refinery as the source of the nuisance or described the odour as “chemical,” “suffocating”, “oil”, or “like a gas station”. The direction of the wind indicated that the refinery may have been the source of these complaints. On the other hand, with the wind blowing from the direction of the wastewater treatment plant, respondents spoke of a “sew-age”, “sludge”, and “rotten” odour. These characteristics, described in question 7, “Can you determine what is causing your odour nuisance?”, make it possible to distinguish between these sources.

Based on the obtained results, it can be concluded that the reason for the odour influence in the studied agglomeration is the refinery operating there. In the event of the appearing southerly winds, in the southwest of the city, there is a wastewater treatment plant, which is also a potential source of odour nuisance.

3.4. Statistical Analysis of Survey Results

The survey results were statistically analysed. To compare the means of the two groups—odour annoyance and odour intensity—a paired sample t-test with 95.0% confidence interval was performed. The t-value is −10.11, while the degrees of freedom is 787, Cohen’s d effect size is −0.717, and p-value is < 0.001. The mean difference is −0.510, while the standard error SE difference = 0.025. The alternative hypothesis was that the difference is not 0, so the null hypothesis was rejected of no difference; with a high degree of confidence, the true difference in means is not equal to zero (Figure 13).

Table 4. Descriptive statistics of odour annoyance and intensity with wind direction as a fixed factor.

Descriptives—Odour Annoyance
Wind Direction	Mean	SD	N
E	3.872	0.801	39
N	4.065	0.744	155
NE	4.196	0.720	107
NW	4.039	0.717	127
S	4.000	0.763	56
SE	4.211	0.674	71
SW	4.250	0.670	40
W	4.060	0.712	50
Descriptives—Odour Intensity
Wind direction	Mean	SD	N
E	4.462	1.211	39
N	4.613	1.034	155
NE	4.748	1.150	107
NW	4.409	1.042	127
S	4.321	1.011	56
SE	4.746	1.092	71
SW	5.025	0.733	40
W	4.580	0.992	50

contains descriptive statistics of odour annoyance and odour intensity (as dependent variables) and wind direction as a fixed factor, while Figure 14 contains descriptive plots of those parameters.

In Figure 14, there are presented odour annoyance and odour intensity grouped by wind direction (abbreviations: SD—standard deviation, N—number of observations).

To verify whether there are statistically significant differences between the mean values of odour intensity and odour annoyance, with wind direction as a fixed factor, the ANOVA test was used. The results of the ANOVA tests for odour intensity with wind direction as a fixed factor are the following:

-

No homogeneity correction: F factor = 2.684, p = 0.010, size of effect: η² = 0.029, ϖ² = 0.018;

-

Brown–Forsythe homogeneity correction: F factor = 2.752, p = 0.008, size of effect: η² = 0.029, ϖ² = 0.018;

-

Welch homogeneity correction: F factor = 3.544, p = 0.001, size of effect: η² = 0.029, ϖ² = 0.018.

The results of the ANOVA tests for odour annoyance with wind direction as a fixed factor are the following:

-

No homogeneity correction: F factor = 1.643, p = 0.121, size of effect: η² = 0.018, ϖ² = 0.007;

-

Brown–Forsythe homogeneity correction: F factor = 1.640, p = 0.122, size of effect: η² = 0.018, ϖ² = 0.007;

-

Welch homogeneity correction: F factor = 1.597, p = 0.138, size of effect: η² = 0.018, ϖ² = 0.007.

For odour annoyance, a one-factor analysis of variance showed no statistically significant differences between group means. On the other hand, the one-factor analysis of variance showed statistically significant differences between group means for odour intensity (F (7.00) = 2.68, p = 0.01, size of effect: η² = 0.029, ϖ² = 0.018). This means that at least one average is statistically significantly different from the others. To determine which is the average, post hoc tests were performed. Since the significance of the Levene variance non-homogeneity test was less than 0.05 (p = 0.003), variances were considered heterogeneous, i.e., differences between variances in the compared groups. Due to the lack of homogeneity of variance, the Dunnett test and the Games–Howell test were chosen. The most significant statistically significant differences were observed between odour intensity values between wind direction groups: SW and NW (pDunnett = 0.009, pGames–Howell = 0.002) and SW and S (pDunnett = 0.009, pGames–Howell = 0.004). The results are presented in Figure 15.

The non-parametric Kruskal–Wallis test was used to verify these results due to a failure to meet the basic assumptions of the ANOVA test. The results of this test confirmed the ANOVA analysis—the p-value was 0.006, indicating significant differences between group means for odour intensity. The multiple (two-sided) comparison test showed statistically significant differences between group mean values of odour intensity (grouping variable: wind direction) for the same pairs of wind direction values (SW and NW and S and SW) as the Dunnett and Games–Howell tests. The Kruskal–Wallis test also confirmed the hypothesis that there were no statistically significant differences between the group mean values of odour annoyance for the group variable wind direction.

Previous studies focusing on pollution measurements did not take into account information from residents and their perception of the odour impact. Survey research allows for a broader look at the problem of odour nuisance and ensuring the comfort of life in an urban agglomeration. The research shows the main odour source as the refinery.

4. Conclusions

As mentioned in the introduction, the field survey carried out in Płock allowed for us to identify three objects with a potential odour: refinery, sewage treatment plant, and landfill and mechanical–biological treatment plant of municipal waste located approx. 13 km from the city centre. The present study confirmed that the refinery is the main source of odour nuisance. However, an additional potential source of odour impact could also be the sewage treatment plant, whose odour nuisance was also reported by the respondents.

The intensity and nuisance of the odour effect were a function of the wind direction. Most respondents indicated a north wind, N (32%), a northwest wind, NW (26%), and northeast wind, NE (22%) for odour annoyance and odour intensity.

The largest number of complaints were recorded in the western and northwestern parts of the city (south of the refinery), especially in 2018 when a reserve emitter was activated. The increased odour nuisance may have been caused by malfunctions and repairs to the main chimney, which would have caused a stronger odour effect. The presented study results concern an important issue, odour nuisance, as confirmed by the surveys conducted, the analysis of residents’ complaints, and the bibliometric analysis. This is the first stage of research aimed at developing a methodology for identifying and characterising sources of odour nuisance in an urban agglomeration using in situ measurement devices. The next stage of the research will be field studies to perform the odour characterisation of individual potential sources of odour nuisance and their odour interaction. The results may have potential application in all cities where there are potential sources of odour.