Spatial Analysis of Air Quality Assessment in Two Cities in Nigeria: A Comparison of Perceptions with Instrument-Based Methods

However, there remains a dearth of such comparative studies from urban areas of the developing world. Hence, the aim of the research set out in this paper was to explore the link between perceptual indicators of AQ and measurements from instruments in various locations within two cities in Nigeria (Abuja and Enugu) along with whether various demographic factors influenced perceptions of AQ in those locations. The two cities were chosen for their contrasting geographical and socio-economic situations (summarised below), although the intention was not to compare the two cities per se. Instead, the choice of such contrasting urban contexts was intended to provide the basis for a robust comparison between perceptual and instrument-based analysis of AQ.

Some studies have suggested that perceived AQ can match well with instrument-measured and model-based results, although many of these have been conducted in urban contexts in the developed world [ 24 ]. A study by [ 25 ] noted that smells reported by individuals were significantly associated with modelled NOand SOconcentrations, and in their study of public perceptions of AQ in Chinese cities, ref. [ 26 ] noted that perceptions can match results based on reference equipment. However, others have noted that perceptual indicators of poor AQ may not necessarily tally with instrument-based measurements [ 27 29 ]. For example, ref. [ 30 ] found an insignificant relationship between perceived and measured AQ in Seoul, South Korea, and in their work in the USA [ 31 ] found perceived AQ to be getting worse while instrumentally obtained information showed there had been an improvement. In Texas, United States of America, ref. [ 32 ] found only a weak relationship between perceived AQ made by locals and that measured via instruments.

One of the issues with using perceptual indicators is that much is likely to depend on the biological ability of humans to detect pollutants. For the particulate matter (PM)-based indicators, such as smoke and dust, the detection may be relatively straightforward as people can see smoke as well as dirt on skin, clothes and surfaces, and indeed this may well explain their relatively high ranking in Table 1 . Gaseous pollutants such as SOand NOare typically detected by people through odour and taste. The colourless gas SOcan be detected by taste at concentrations of 0.35 to 1.05 ppm and has an irritating odour with a threshold in the range of 0.67–4.75 ppm, depending upon the individual [ 22 ]. An odour threshold range of 0.1 to 0.4 ppm for NOis cited by [ 23 ]. On the other hand, CO is both a colourless and odourless gas that humans cannot so readily detect. Hence, while odour and even taste are important perceptual indicators, as noted in Table 1 , some pollutants may not be detectable if concentrations are below a threshold, while others may not be detectable unless levels are so high that they cause effects such as difficulty in breathing. It is also conceivable that the ability to detect pollutants may vary depending on factors such as age, health and education, and there is also the possibility that experience of exposure to poor AQ could be important. Finally, it is also likely that people’s perceptions of poor AQ may be framed by assumptions based on the prevalence of causes such as traffic or indeed influenced by the views of others. Therefore, while public perceptions of AQ have been explored by researchers, albeit to a much lesser extent in urban areas of the developing than the developed world, there are important questions to ask about how well such perceptions match results obtained using instruments.

A different approach to assessing and managing AQ is to seek the views and perceptions of those people directly experiencing it. For example, an investigation by [ 16 ] found that people perceived AQ to be very poor and the major environmental problem in Ljubljana, Slovenia, while [ 17 ] found that urban residents, especially those suffering from asthma, are more concerned about and tend to check the AQ in their vicinity. In their study at Richards Bay and its surroundings in uMhlathuze Municipality, KwaZulu, South Africa, ref. [ 18 ] found that residents were concerned about AQ, and most perceive it to be fair or poor, with industrial emissions regarded by them as the major cause of poor AQ. According to [ 19 ], residents in Accra, Ghana, were aware that their AQ is poor and does influence their health, although they also found that some demographic groups, such as the less educated and the elderly, were less aware of poor AQ in the city. Similar work was conducted in Nigeria among people living near a cement manufacturing area of Ewokoro and Reno-North Local Government Areas of Ogun State by [ 20 ] and noted that people are aware of poor AQ and its adverse health effects. Recently, ref. [ 21 ] using an online survey of residents’ perception of AQ in Abuja and Enugu cities, Nigeria, stated that a variety of ‘perceptual’ indicators of poor AQ were employed, and these varied in terms of their relative importance ( Table 1 ). Respondents were aware of their exposure to poor AQ, and they used smoke, dust, and odour as the main perceptual indicators of poor AQ. However, while the use of perceptual indicators of poor AQ has value, according to [ 16 ], the use of such tools should be seen as providing a “supplementary” approach alongside data from monitoring instruments.

Conversely, many cities in developing countries have an unstable electricity supply, inadequate or no laboratories or air quality monitoring stations, little or no financial and human resources, and weak telecommunications systems, resulting in a dearth of reliable and timely information on AQ [ 13 14 ]. Finding and using the best monitoring equipment may be unique to each context and may also depend on levels of precision required, standards and regulatory framework. For example, ref. [ 7 15 ] note that a sensor’s ability to measure accurately can be compromised by chemical and physical interference, and improvements designed to tackle the abnormalities need to be validated against reference measurements. Low-cost alternative sensors have been developed, but the measurements can fluctuate depending upon factors such as temperature, humidity, pressure, and signal instability [ 15 ].

Appropriate means of assessing pollutant levels are needed as the basis for AQ management [ 6 ], and many instruments such as passive air pollutant samplers, air quality sensors, reference instruments [ 7 ], and Earth Observation (EO) satellites have been employed, with data from such instruments often used as the basis for predictive models [ 8 10 ]. According to [ 11 ], nearly all urban areas in Europe and North America have networks of AQ monitoring instruments at a density of about 1 per 100,000 to 600,000 residents, while in developing regions, such as Africa, it is estimated to be around 1 per 4,200,000 residents. Although there has been some improvement in recent years, there are still huge challenges in monitoring AQ, especially in urban areas of developing countries [ 12 ]. For instance, ref. [ 11 ] reported that in the United States of America (USA), federal, state, and tribal agencies constructed the existing AQ monitoring systems over five decades at a cost of millions of dollars for the hardware, human resources, and institutional frameworks to install and maintain monitoring stations and process, analyse, and report the information. Additionally, in developed countries such as the USA, there is a steady electricity supply system, telecommunications, and other structures to ensure uninterrupted collection/processing of data as well as maintenance of hardware.

The burden of disease linked with exposure to poor AQ is significant and rising, and this is especially the case in middle and low-income countries [ 1 ]. The World Health Organization (WHO) has reported that ambient poor AQ is responsible for about 4.2 million global deaths annually, with 99% of the global population breathing air that exceeds the organisation’s guideline limits [ 2 ]. Poor AQ also has several important adverse impacts on climate change and may directly affect natural ecosystems and biodiversity [ 3 ]. Nitrogen oxides (NO), nitrogen monoxide (NO), and ammonia (NH) emissions alter terrestrial and aquatic ecosystems and Tropospheric Ozone (O), while black carbon and Particulate Matter (PM) are short-lived climate forcers that lead directly to global warming. Poor AQ also damages materials, buildings and artwork through corrosion, biodegradation, and weathering and fading of colours. A report by [ 3 ] also stated that the market costs of Poor AQ include reduction in productivity of labour, extra health expenses, and loss in yield of crops and forests. Indeed, despite urbanisation being one of the fundamental features of economic development, concerns about AQ have led to a wide-ranging discussion about the meaning of urban sustainability [ 4 ]. An important consideration here is the need to define what is meant by AQ. A common approach taken by agencies charged with monitoring and managing AQ is to define it in terms of concentrations of pollutants, be they gasses or particles in the air (please see for example [ 5 ]).

To address the issue of varying methods and approaches taken across the published studies, source [ 46 ] was employed wherever possible for Abuja, and thus if multiple sources reported data for a location, then only [ 46 ] was employed. The other sources [ 49 ] were only used to fill locations not covered by [ 46 ]. In Enugu, the same method was applied. Most of the data came from sources [ 51 ] as this is the one with the highest level of coverage across locations, followed by source [ 50 ]. Again, if multiple sources were available for a location, then [ 51 ] was used as the “default”. Other sources with limited data which are not in the graphs are [ 52 54 ] for Abuja ( Table S4 ) and [ 55 56 ] for Enugu ( Table S5 ), and they identified some locations in both cities to be of high concentration of air pollutants instrumentally obtained.

Of the four sources for Abuja, the one with the best coverage in terms of the number of locations was [ 46 ], and the other three sources could be used to fill gaps, such as Gwagwalada. For Enugu, both sources were used to cover the locations. One of the issues here is that all the sources used different methods and approaches to assess AQ, and full details of the methods, including instruments, used can be found in the sources listed in Tables S4 and S5 . However, as an illustration of this variation, a summary is provided here for [ 46 50 ]. The source [ 46 ], the one mostly used for the Abuja locations, measured PM2.5 and PM10 using a handheld China Way CW-HAT200 aerosol sampler or counters, with the instrument held 2 m above ground level. Each location was randomly monitored hourly between 06.00 to 12.00 for dry (11–15 February 2019) and wet (17–21 June 2019) seasons. These data were used to calculate the daily mean levels of PM. For SO, NOand CO, a BOSEAN portable gaseous emission analyser was used to determine the ppm. Source [ 50 ] focussed on measuring AQ in around three major and two minor roads in Enugu. These researchers employed air samples randomly collected and monitored in periods of high vehicle density, 8:30 am–10:30 am and 4:30 pm–6:30 pm. Data collection took place over three days in the week (Monday, Wednesday and Friday) and twice monthly during September and November 2017. Concentrations of CO, NOand SOwere determined using a 350XL Emission Analyzer. These data were used to calculate the mean daily concentrations of the pollutants.

The authors made use of existing published data for instrument measurements of AQ at the locations in the two cities. This was to be consistent with the available existing data and to minimise constraints due to logistical issues during the Covid 19 pandemic. A desk study (using the search engines Surrey Search, Google and Google Scholar) retrieved 40 published studies from a wide range of sources and date periods on air pollutant concentrations covering Abuja and Enugu. The search was further narrowed by only including those reported in referred academic journals spanning the period 2014 to 2020, giving four sources for Abuja [ 46 49 ] and two sources for Enugu [ 50 51 ] (see Tables S4 and S5 (Supplementary Material) ). The period chosen for the journal search (2014–2020) allowed for the best match possible with the time of the survey (end 2020/early 2021).

The survey participants were asked to provide demographic information and to score locations within the two cities in terms of their AQ (see Appendix A for the questionnaire format). The locations included in the questionnaire were based on those identified by some key informants during fieldwork undertaken in May and June 2019 and are listed in Table 3 and Table 4 , along with some brief descriptive notes based on the experience of the authors. The key informants (12 in number) were from the Federal Ministry of Environment (FME), Abuja Environmental Protection Board (AEPB), Nigeria Meteorological Agency (NIMET), National Orientation Agency (NOA), National Environmental Standards and Regulations and Enforcement Agency (NESREA), National Space Research and Development Agency (NASRDA), and a private construction and other individuals from the cities. They were asked to identify the locations ( Table 3 and Table 4 ) they considered to be useful for comparison in terms of AQ within Abuja and Enugu. They would often mention a location they regarded as likely to have poor AQ and suggested other locations they suspected to have better AQ to provide a comparison. These locations were subsequently included in the survey questionnaire.

The gathering of survey data was undertaken between October 2020 and March 2021 during the COVID-19 pandemic. The questionnaires were provided online via the QUALTRICSXMPlatform, although hard copies were also distributed to help some respondents who lacked access to an electronic device. The research was facilitated by carefully chosen, trained, and monitored local field assistants. Incentives were provided to the eight field assistants at 7808 (USD 14) each over the data collection period, and the respondents were paid 1672 (USD 3) each per properly finished survey form. The field assistants helped assure appropriate stratification, exclusion of children due to ethical reasons, and minimisation of the impacts of the COVID-19 pandemic in the demographic categories in Table 2

The research employed a questionnaire-based survey of residents in Abuja (137 respondents) and Enugu (125 respondents), and the sample was stratified to guarantee adequate demographic to reflect the population profile in the two cities ( Table 2 ). The result of the stratification was an approximately 50:50 gender balance in both cities. Age was categorised into two groups: 18–34 years and 35 years or over-covering 45% and 55% of respondents, respectively, in the two cities. The categorisation of age into two groups allowed an alignment with the official classification into adults and youth in Nigeria [ 42 ]. The minimum age of 18 years was designed to broadly comply with the research ethic protocols of the research organisation (University of Surrey). The income level was divided into three categories of no income/low income, mid-income, and higher income. There was a higher proportion of higher-income earners in the Abuja sample (40% of respondents) compared with Enugu (15% of respondents). Additionally, more respondents have higher education (PhD/masters or equivalent) in Abuja (36% of respondents) than in Enugu (18% of respondents). The three categories of income were chosen to allow for adequate numbers within each category to facilitate statistical analysis. The three categories are linked to the levels of income (lower income, middle income, and upper income) used by [ 43 ] in the USA. Similarly, the respondents were divided into three groups based on the highest level of education to allow for adequate numbers in each group to facilitate statistical analysis. The education and income levels employed here are similar to those of [ 19 ] in their study on perceptual AQ in Accra, Ghana.

Enugu city is the capital of the Nigerian south-eastern state of Enugu ( Figure 1 ). The city is located at approximately 223 m above sea level at 6°27′10″ N 7°30′40″ E and has an annual rainfall of approximately 2000 mm and a daily mean temperature of 26.7 °C [ 39 40 ]. Enugu city’s population was estimated to be 773,000 in 2019 [ 41 ]. Unlike Abuja, which is a more recent construct from the 1970s, Enugu existed long before Nigerian independence in 1960 and is not a “planned” city. Without formal zoning, the road system has evolved with the city’s growth, and manufacturing and business activities are intermixed with residential locations.

Abuja city sits at an elevation of 840 m above sea level at 9°4′ N 7°29′ E and has an annual rainfall of between 305 mm to 762 mm and a daily mean temperature of 32.5 °C [ 36 ]. In 2010, the United Nations regarded the city as the world’s fastest-growing, with a 140% increase in 10 years (2000–2010) [ 37 ]. Nigeria’s National Bureau for Statistics 2016 estimated the population of FCT to be 3.5 million using the annual growth rate of 9.3% from the country’s 2006 population census, with Abuja metropolis having a projected population of roughly 1.9 million [ 38 ]. Abuja, which replaced the older and more densely populated city of Lagos as the capital of Nigeria in December 1991, is a planned city with wide roads and districts mapped out for residential houses, governmental buildings, and commercial activities. Abuja’s economy is dominated by the financial service sector, retail and real estate, although there is some manufacturing that takes place in the Idu industrial area.

While the instrument-based measurements of the pollutants matched the perceptions of AQ provided by the respondents, it is important to consider the levels at which humans may detect these pollutants either by odour or taste. Figure 4 presents the instrument-measured concentrations of SOand NOin locations of the two cities alongside the minimum detectable odour and taste levels found in the literature. The SOdata for some locations in Abuja ( Figure 4 a) show recorded concentrations below the odour but above the taste thresholds that can be detected by humans. However, in Enugu ( Figure 4 b), the SOconcentrations are well above the minimum taste and odour levels at Ogbete/Ogbete Main Market locations, suggesting that they may be readily detected by humans. In both cities, it is interesting to note how the locations that have levels above the minimum detectable by humans tend to be those locations that scored worse for AQ by the survey respondents. For NO, the concentrations in Abuja ( Figure 4 c) and to a lesser extent Enugu ( Figure 4 d) are well above the detectable (by odour) levels for some locations towards the right-hand side of the graph—those that scored worse by the respondents. Nonetheless, it does need to be noted that the survey respondents are likely to have used additional clues, such as the presence of smoke, dust, and perhaps the odour of a “cocktail” of pollutants (including hydrocarbons), to frame their scoring rather than just direct sensing via odour or taste of SOand NO

In terms of comparing the overall perceptions of AQ (based on the mean scores) with measurements from instruments, the instrument-measured concentrations of SO, NOand CO for the cities are higher in the locations on the right sides of Figure 3 . As the locations in Figure 3 are ranked from left to right based on the scores of respondents, this suggests that the perceptions of the respondents broadly match the AQ as recorded via instruments. The PM data for Abuja are sparse, but the few that are available also have this pattern. In Enugu, the PM concentrations appear to be consistent across all the locations, but there is a suggestion that the areas deemed by respondents to have the best AQ (left-hand side of Figure 3 b) do have the lowest PM concentrations. For SO, NO, and to a lesser extent CO, the data in Enugu match well with perceived AQ, with the highest concentrations of the pollutants being found in locations where respondents scored the worst.

Using the Hochberg post hoc test, the locations were divided into statistically homogenous groups (< 0.05); six groups from the nineteen locations in Abuja and seven groups for the eighteen locations in Enugu. These are labelled A1 to A6 for Abuja and E1 to E7 for Enugu in Figure 2 . In Abuja ( Figure 2 a), the locations with the worst AQ are Gwagwalada and Wuse Market, while the better AQ locations are those in group A1 (Maitama, Central Business District (Central Area), Wuse 2, and Gwarinpa). Maitama is the only location with an AQ viewed as very good by the respondents ( Figure 2 a). Figure 2 a also illustrates locations in group A6 which are Apo/Apo Bridge, Area 1 Junction, Utako, Berger Junction, Area 3 Junction, Mabushi, Lugbe, and AYA junction in Asokoro, which are rated similarly, with Gwagwalada and Wuse Market being the worst AQ locations. In between the better locations and the worst locations are Wuye, Area 11, Jabi, Kado and Durumi. In Enugu ( Figure 2 b), group E7 which includes Coal Camp, Ogbete/Ogbete Main Market, Old Park and Abakpa/Abakpa Junction are the locations with the worst perceived AQ. The location with the worst AQ in Enugu is Abakpa/Abakpa Junction. Emene, Ugwuaji, and NOWAS Junction are other locations with perceived poorer air quality that also have Nkpokiti, Ogui, Asata, and Awkunanaw in the same group. The locations with the cleanest air in Enugu are Independence Layout and GRA in group E1 with equal scores, followed by New Haven and Trans-Ekulu.

The mean scores and 95% confidence intervals of the AQ in the selected locations in Abuja and Enugu are shown in Figure 2 , with higher number scores meaning poorer perceived AQ. The locations are ordered so that those with the worst perceived AQ scores are on the right-hand end of the x-axes of Figure 2 , while those having the best scores are on the far left. While the perceptions of AQ varied across locations within both cities, it is not possible to use these data to make a more general comparison of AQ between the cities.

4. Discussion

The research described in this paper was designed to explore whether the perceptions of AQ align with those provided by instruments. Based on published results in the literature, it does indeed appear to be the case that the locations in both cities vary significantly in terms of the concentration of some pollutants in the air. However, it does need to be noted that comparisons between the two cities are challenging, as much depends on the choice of location within them as well as differences in methods adopted by the researchers to measure pollutant concentrations. Nonetheless, the findings suggest that perceptions of AQ do appear to be associated with instrument-based measurements as those locations in Abuja and Enugu that people perceive to have the worst AQ are those that instrument-based studies have shown to have the highest concentrations of pollutants such as PM, SO2, NO2, and CO. This intriguing and important finding, however, does have some important caveats.

Firstly, there were significant differences between demographic groups in how they perceived AQ. Gender was not important in either city, but there were differences in terms of the income-education demographics in Abuja (but not Enugu) and age in Enugu (but not Abuja). In both cases, the differences between demographic groups were most marked for locations that had poorer AQ (i.e., locations in the lower parts of Table 5 and Table 6 ). Why should this be so? The differences between age groups in Enugu were noted in previous research [ 21 ], where the younger demographic (18–34 years old) was more positive about measures to control AQ and the need for action by agencies such as state and local government. This may be associated with youth identity and activism in the Igbo (the dominant ethnic group in Enugu) compared with the much more multi-ethnic and multi-cultural nature of the Abuja respondents from Tiv, Yoruba, Igbo, Hausa, Fulani, Edo, Idoma, Igala, Efik, Ibibio, Gbagyi, Ijaw, Eggon, and Berom [ 21 ]. However, the importance of the income-education demographics in Abuja (but not Enugu) is intriguing. One possibility may be linked to the greater wealth and income levels of respondents in the Abuja sample; 58% of Abuja respondents had annual incomes of about 51,000 compared with 46% for Enugu. Differences in income levels can generate segregation between rich and poor neighbourhoods, as seen, in the most extreme cases, with “gated” communities that have been on the rise throughout Sub-Saharan Africa and which are known to generate spatial fragmentation and urban segregation [ 59 ]. Their popularity is partly driven by a desire for better security, and they exist in Abuja and Enugu. Segregation between richer and poorer neighbourhoods, with the former tending to be more distant from causes of pollution [ 60 ], may engender a more polarised view of AQ. A further factor is that people living in low-income neighbourhoods tend to have more negative views of their environment than do those from wealthier neighbourhoods [ 61 ].

2 in Abuja and NO2 in Enugu, the concentrations were indeed higher in locations perceived to have the worst AQ, but these concentrations were below the thresholds at which the pollutants could be detected via the senses of smell and taste. It would appear more likely that respondents were basing their views of AQ on the presence of more visible “clues”, such as smoke and dust, and while the odour is an important indicator (2 and NO2. In addition, respondents may also have framed their scoring of locations, at least in part, on the presence of causes of poor AQ. This possibility resonates with the work of [

Secondly, while there is an apparent link between perceptions of AQ and measurements of pollutant concentrations in both cities, this may not necessarily mean that people can detect the pollutants directly. For SOin Abuja and NOin Enugu, the concentrations were indeed higher in locations perceived to have the worst AQ, but these concentrations were below the thresholds at which the pollutants could be detected via the senses of smell and taste. It would appear more likely that respondents were basing their views of AQ on the presence of more visible “clues”, such as smoke and dust, and while the odour is an important indicator ( Table 1 ), it may be that it reflects a “cocktail” of pollutants in the air including hydrocarbons. These clues may well be positively correlated with concentrations of SOand NO. In addition, respondents may also have framed their scoring of locations, at least in part, on the presence of causes of poor AQ. This possibility resonates with the work of [ 62 ], who analysed data from the Third European Quality of Life Survey undertaken between 2011 and 2012. They concluded that perceived exposure to air pollution is formed based on sensory awareness as well as what they refer to as a “cognitive component”, which is framed using knowledge of exposure to sources of air pollution (e.g., presence of traffic), perceived ability to cope with that exposure, and the perception of health risks. Moreover, an earlier study by [ 63 ] shows that odour annoyance is more perceived to exist in locations that have higher levels of traffic. This study and the study of [ 64 ] illustrated that the detection of air quality by people could be because of the extent or duration of the situation which causes poor air quality. The work of [ 64 ] also found that people perceive traffic as one of the main sources of poor AQ. Social discourse such as views from people via online social sites can also influence perceptions of environmental issues including AQ [ 65 ].

The results, both perceptions and measurements, do indeed point to differences between the AQ of locations in the two cities. The locations for comparison in the survey were selected using the advice provided by key informants, and some locations were known to have issues with AQ. In general, the AQ in government reserved areas (GRA or areas resided by higher-income earners) and administrative locations were better than market locations, busy junctions, and lower-income earners’ residential locations, which are mainly mixed with other functions such as manufacturing and retailing with a high density of inhabitants ( Table 4 ). In Abuja, Maitama is perceived to be the cleanest among the selected locations, and this location is mainly inhabited by prominent people and political office holders, while the AQ at Wuse Market and Gwagwalada is scored more poorly. This result is similar to the work of [ 54 ], where they found that locations with major vehicular traffic junctions, such as AYA Junction, Area 1 Junction, Area 3 Junction, Gwagwalada, Wuse Market, and Mabushi Roundabout, have poorer AQ compared with locations with lower vehicular traffic. For Enugu, the results show that Independence Layout and GRA (mainly higher-income earners residential areas) are jointly ranked as those having the best AQ, while seven locations (Ugwuaji, NOWAS Junction, Emene, Coal Camp, Ogbete/Ogbete Main Market, Old Park, and Abakpa/Abakpa Junction) (mixed-use, commercial and transportation locations) that are rated above neutral are the locations with the worst AQ.


The assessment of AQ, especially within the developing world, has, to date, mainly been achieved using monitoring equipment and models, but issues including inadequate funds and technical know-how have been limiting their use. These limitations could be alleviated by using a participatory approach, such as the one used in the present research to draw upon people’s perceptions of AQ. While perceptions of AQ are framed by perceptual indicators, such as odour and dust, there may well be other factors at play such as the “cognitive component” identified by [ 62 ] and framing via social discourse [ 65 ]. Hence, while the research reported here has shown a link between the perceived AQ of locations in Abuja and Enugu and the measured levels of pollutants, it is perhaps understandable that this may not always be the case as reflected in the spectrum of published evidence that falls in both camps [ 25 32 ]. More research is certainly needed in this important field, especially for urban centres in the developing world. Nonetheless, we suggest that the use of perception in monitoring AQ should be globally accepted as a complementary approach to the use of instruments. Perceptual-based approaches cannot substitute for instrument-based approaches, but the authors are very much in agreement with [ 16 ] that perceptual-based approaches can be a useful supplement. Perceptual indicators and monitoring not only serve as a form of assessment but also as a form of awareness-raising of AQ and related environmental phenomena.

Leave a Comment

Your email address will not be published. Required fields are marked *

%d bloggers like this: