‌4

‌Calibrating the global: How are Ghanaian scientists shifting Africa’s position in global atmospheric science?

‌Jessica Pourraz and Allison Felix Hughes

According to the World Health Organisation (WHO), air pollution – ambient and indoor – is the leading environmental risk to health worldwide. It accounts for 7 million deaths per year and most severely affects populations in the Global South, where nearly 92% of total air pollution–related deaths occur. It also reflects significant social inequalities, as the poorest populations are the most affected (Landrigan et al. 2017). Prospective studies show, for instance, that by 2050, the African population will almost double, and half of its inhabitants will live in cities where they will more likely be exposed to air pollution (Katoto et al. 2019). In African cities, the main contributors to air pollution are vehicle fuel combustion; dust from unpaved roads, brickwork, and construction sites; the combustion of biomass and household waste; and industry.

In 2018, the WHO estimated that in the capital city of Ghana, Accra, 28,000 premature deaths had occurred due to air pollution. Despite the scale and severity of air pollution in Ghana, institutional monitoring systems remain weak, and environmental regulations are poorly enforced. To measure air pollution in the metropolis of Accra, the experts at the Environmental Protection Agency (EPA-Ghana) only have a few outdated mobile devices, which they use along the main roads, and two fixed reference-grade monitors located on the University of Ghana campus and at an elementary school in downtown Accra. The monitoring improvement is made all the more difficult as Global North actors set international standards and rely upon expensive reference-grade monitors. These monitors are high-precision air-quality control instruments that meet specific standards and criteria set by internationally recognised regulatory agencies and scientific organisations. They undergo rigorous testing, calibration, and quality assurance procedures to ensure their accuracy and traceability in compliance with recognised standards. These monitors often involve advanced sensor technology and may incorporate multiple sensors to measure various air pollutants simultaneously. Reference-grade monitors are heavy machinery installed in a fixed location (they are not portable), and they are very costly (US$100,000 on average). In the field of air pollution research, the international scientific community and academic journals consider highly technical and expensive measurement instruments such as fixed reference-grade monitors – not affordable to most scientists in the Global South – as the ‘gold standard’.1 ‘Not only does this virtually create a Global North monopoly on publication, but it also wrongly positions research carried out with such technology as “novel” as if the new research was shedding light on an unknown problem’ (Negi and Ranjan 2020: 2).

In turn, the lack of resources to purchase, operate and maintain the traditional fixed reference-grade monitors largely explains the dearth of regular and systematic air quality monitoring in countries like Ghana – as well as in most of the other Sub-Saharan African (SSA) countries (Giordano et al. 2021; Hagan and Kroll 2020; Morawska et al. 2018). This issue directly affects Ghanaian scientists working on air pollution, but does not mean that air pollution research remains out of reach of Ghanaian scientific institutions: studies of air pollution in countries like Ghana (Arku et al. 2008; Dionisio et al. 2010) and Kenya (deSouza 2020) go back several decades and involve considerable work carried out by local researchers, who adapted their methods to less costly material. How do scientists from these countries manage to build expertise in air pollution?

One key tool used by Ghanaian researchers has been low-cost sensors (LCS), which have become very popular among local academic actors and even among EPA-Ghana experts, because they make it possible to produce air quality data at a much lower cost. LCS are affordable miniature low-tech sensors designed to produce real-time continuous data on particulate matter (PM) – a mixture of solid particles, liquid droplets, and gases – that can be shared through digital platforms. These devices illustrate the emergence of a growing but highly differentiated global market for air monitoring. The air quality monitoring market worldwide accounted for US$5.08 billion in 2024 and is expected to reach US$6.73 billion by 2029, with a growth rate of almost 6% between 2024 and 20292. This market is highly segmented by product type, end user, and geography. It is also deeply hierarchised through norms and standards. It includes both expensive reference-grade monitors and LCS, the latter being mostly dedicated to low-income countries, such as in the Global South, where they form alternative or even complementary networks to the institutional monitoring system.

To a certain extent, the LCS market constitutes a limited and temporary solution for Ghanaian experts, as it implies both a prolonged dependence on gold standards and certain forms of exploitation by the LCS-providing companies. Indeed, some experts question the validity of LCS measurements and point to the fact that, unlike reference-grade monitors, LCS are not subject to international standards and certification procedures. Therefore, there is a need to benchmark LCS against fixed reference-grade monitors (Parasie and Dedieu 2019; Sahu et al. 2020). This work of benchmarking, also called calibration, is ‘defined as the process by which outcomes from LCS are compared against reference monitors and adjusted after comparison’ (Pritchard et al. 2018: 6). In order to make the devices accurate and ensure the validity of the data produced, Ghanaian scientists carry out this calibration by themselves; this calibration work is, in turn, appropriated and valued on the market by LCS companies that sell their devices to other Global South countries. Ghanaian scientists, therefore, not only depend on a scientifically dismissed market, defined by lower standards than the ‘gold standard’ they cannot afford in a knowledge economy dominated by Global North actors, but the calibration work they achieve in this market is not officially recognised nor acknowledged, nor financially compensated despite the added value it brings to the devices for African markets. But at the same time, the LCS market operates as an emancipatory location for Ghanaian scientists. The use and calibration required by LCS play an important part in shaping expertise in the Global South. They help Ghanaian scientists shift Africa’s position within the global atmospheric science community by producing scientific knowledge and publishing it in international journals.

By analysing the assemblages formed by academics from the Global North and African experts, international donors, and private actors to produce scientific knowledge, this chapter explores the complexity of this situation. It shows how Ghanaian scientists are coping with international inequities in air knowledge production by inserting themselves into global technological markets and by appropriating incoming technologies, instruments, and tools from the Global North to develop legitimate expertise and create new ways of conducting atmospheric science. With these issues in mind, we (1) describe the international academic partnerships that have been established over the last fifteen years in Ghana to allow their scientists to conduct research; (2) discuss calibration in practice by examining the invisible work of the Air Quality Research Laboratory of the Department of Physics at the University of Ghana, where environmental and atmospheric physicists carry out all the calibration work; and (3) conclude by analysing how this free and invisible labour participates in shaping Ghanaian scientific expertise and contributes to legitimising Ghanaian scientists in global atmospheric science.

This contribution is based on an ethnographic work carried out at the Air Quality Research Laboratory of the Department of Physics at the University of Ghana (UG), at the laboratory of the Environmental Quality Department of the EPA-Ghana, and during their air quality–monitoring tour of the Accra metropolis between August 2021 and April 2023. These observations were completed by semi-structured interviews conducted in Accra (N=35) with atmospheric physicists and academics from the University of Ghana, EPA experts, members of NGOs, representatives from startups, officials from the Greater Accra Metropolitan Area, and officials from the Ministry of Environment. It was completed by the analysis of grey literature, such as EPA-Ghana activity reports, press articles, and social media posts.3

‘Africa–Global North’ research collaborations on air pollution in Accra

Air pollution stems from a combination of several emission sources, which specific meteorological and geographical conditions may amplify. Particulate matter (PM) (PM2.5 and PM10) is the most commonly used measure of air pollution. PM10 includes particles with a diameter of less than ten micrometres (µm), while PM2.5, or fine particles, are those smaller than 2.5 µm. Epidemiologists have identified the adverse effects of fine particles on human health due to their ability to penetrate deep into the lungs, affecting the respiratory system on a deeper level (Moizard-Lanvin 2021). PM data are, therefore, the most commonly produced to objectify air pollution and the easiest data to collect, as they do not require any laboratory chemicals or organic analysis.

West African cities such as Accra have PM levels well above WHO thresholds. The WHO recommends that PM2.5 should not exceed ten micrograms per cubic metre of air (µg/m3) and that PM10 should not exceed 20 µg/m3 (World Health Organisation 2021). However, average levels in Accra are still two to four times higher than the WHO guideline value (Alli et al. 2021).

The issue of air pollution is not new in Ghana. The first Environmental Protection Act – EPA Act 490 – was passed in 1994. In 1997, the Environmental Quality Department of the EPA-Ghana, under the remit of the Ministry of the Environment, began monitoring air quality as part of a programme funded by the World Bank, which ended in 2001. Several air quality indicators were measured on an ad hoc basis in the western region of Ghana, which was polluted by mining activities, and in some districts of the main Ghanaian cities, including Accra, Kumasi, Takoradi, and Tema. In 2004, the EPA-Ghana experts received support from the US Agency for International Development (USAID) and the US Environmental Protection Agency (US-EPA) to deploy an institutional system for monitoring PM in residential, commercial, and industrial areas of Accra and along the main roads of the Greater Accra Metropolitan Area (Pourraz 2024).

Around this time, in 2006, the now head of the Quality Research Laboratory of the UG Department of Physics began conducting research on air pollution as part of his PhD on the characterisation and source apportionment of airborne particulate matter in certain urban neighbourhoods of Accra.4 He had the opportunity to work under the supervision of Professor Majid Ezzati at the Harvard School of Public Health in Boston. He thus benefited from the material and instrumental resources of this prestigious university. Prior to that, academic research in this area in Ghana was non-existent.

This work has led to a number of scientific collaborations, enabling Ghanaian academics from the Department of Physics and the Department of Geography at UG to conduct studies in Accra with the support of Professor Ezzati of Harvard University. Thus, for about a month in 2006, these academics collected continuous data on PM2.5, PM10, and concentrations of sulphur dioxide (SO₂) and nitrogen dioxide (NO₂) – which are indicative of fine variations in vehicle pollution – in two poor neighbourhoods of Accra: James Town and Nima. The first results of this pioneering research highlighted vehicles’ contribution to air pollution. The academics’ next study, carried out in 2007, further investigated the trends in these same neighbourhoods, with stationary and mobile data collected over three months. In September 2007, two additional neighbourhoods – Asylum Down (middle class) and East Legon (upper class) – were added to the study and monitored for a year. The results showed that biomass particles, road dust, and vehicle emissions were the main contributors to PM (Arku et al. 2008; Dionisio et al. 2010a; Dionisio et al. 2010b; Rooney et al. 2012; Zhou et al. 2013). For all these studies, the monitors used to measure PM were considered not as LCS but as reference sensors, even though they were not fixed reference-grade monitors.

These academic collaborations initiated in 2006 have continued as part of the global partnership Pathways to Equitable Healthy Cities, coordinated since 2018 by Professor Ezzati, who is now working at the School of Public Health of Imperial College London. Within the framework of this global partnership, several students from the UG have been recruited to conduct their MPhil research, or even as research assistants, to collect weekly data (PM, weather variables, pictures) and maintain the database. These collaborations have been reinforced by the cooperation of US-EPA experts and scholars from Columbia University, York University, and the Kigali Collaborative Research Centre, through which Ghanaian academics were able to obtain a significant number of LCS. Fifteen years of this ‘Africa–Global North’ scientific collaboration have enabled Ghanaian scientists to make their mark on the global academic landscape by performing measurements of the spatial, socioeconomic, and temporal patterns of ambient and indoor air pollution in Accra, and by differentiating pollutants and their sources, using mainly LCS.

In January 2019, the EPA-Ghana experts, with the help of Ghanaian academics and financial support from the World Bank, finally installed the first two fixed reference-grade monitors in Accra, including one at the UG Department of Physics. Fixed reference-grade monitors play an instrumental role, as they are designed to provide accurate and reliable measurements of various air pollutants. They are also used as benchmark instruments for comparing or calibrating other air quality–monitoring devices such as LCS. By that point, EPA-Ghana experts had already deployed ten LCS in Accra, including four received from the US-EPA to measure PM on a continuous basis. Given the lack of sufficient financial resources, these LCS allowed for the expansion of the institutional monitoring network and the characterisation of the spatial variability of PM levels in different areas of Accra.

Because Ghana is located close to the equator, with extreme temperatures and other particular weather parameters, the Air Quality Research Laboratory of the UG Department of Physics is striving to become a centre of excellence for LCS testing. Its unique technological setup, including a fixed reference-grade monitor, should encourage more academic collaborators from abroad to have their LCS tested by the UG Department of Physics and make them suitable for use elsewhere in Africa. This whole process is part of a wider project conducted by the UG Department of Physics to replicate the Air Quality Sensor Performance Evaluation Center (AQ-SPEC) programme established by the regulatory agency responsible for improving air quality in California and funded by the US-EPA. As part of an effort to inform the general public, the AQ-SPEC programme aims to conduct a thorough characterisation of currently available LCS under ambient (field) and controlled (laboratory) conditions.5 Ghanaian environmental and atmospheric physicists are therefore seeking not only to create a hub to test LCS and publish the results of their evaluations online but also to train more academics and experts from Africa to use and calibrate LCS.

In this first section, we have shown how, over the last 15 years, Africa–Global North research and institutional collaborations surrounding air pollution have enabled the training of Ghanaian scientists, the development of local capacities, and the acquisition of instruments. Transnational technological, financial, and human flows to Ghana, mainly from the United States and later from the United Kingdom, highlight the dominance of the anglosphere and the Global North in international research on air pollution, which is also reflected in the financial support provided by the World Bank and USAID and technical assistance from the US-EPA to the EPA-Ghana experts. Having said that, by appropriating financial resources, instruments, and knowledge from the Global North, Ghanaian scientists are trying to counterbalance this dominance and become key players in atmospheric science in Africa, for instance by replicating the AQ-SPEC.

In the following sections, we show how Ghanaian scientists, by adapting LCS to local settings via calibration work, are instrumental in producing knowledge surrounding air quality and are contributing to shifting Africa’s position in global science.

Reconfiguring air quality monitoring in the Global South with LCS

The need for more air quality data for African cities has severely hampered efforts to characterise and understand the patterns of air pollution concentrations and to promote policy initiatives to control and regulate such pollution. Until 2010, only actors such as governments and research organisations that could afford very expensive instruments and had qualified personnel could routinely perform PM measurements similar to those taken in the Global North.

Over the last decade, however, air quality monitoring has seen radical changes, driven by the emergence of LCS, which provide a way to bridge the gaps in official air quality monitoring. LCS were originally intended to democratise environmental monitoring practices, such as air quality sensing, that might ordinarily be the preserve of expert scientists. ‘Citizen science’, which refers to this alternative production of knowledge around toxic exposure by a wide range of actors without resources – such as citizen groups, activists, and vulnerable populations exposed to toxic sites – has been studied extensively by social science academics in the United States, in the context of petrochemical industrial areas in Louisiana and Pennsylvania (Allen 2018; Gabrys et al. 2016; Ottinger 2010; Pritchard and Gabrys 2016). These works have demonstrated that most of the data produced by LCS are intended for information sharing, awareness raising, and lobbying the authorities, and not yet for any legal purposes.

In resource-constrained countries around the world, such as those in Sub-Saharan Africa, the use of LCS for air quality monitoring offers significant opportunities to expand air quality data acquisition. This has led to a reconfiguration of the instruments and actors able to perform such monitoring – with the latter now including academics in the Global South – and of ‘the temporal and spatial scales on which we think about air quality in general’ (Giordano et al. 2021: 158). LCS can provide measurements on the level of a street or a neighbourhood, that is, at a scale that previously eluded monitoring institutions. These sensors also enable continuous monitoring in real-time, which can show pollution peaks and not just average pollution levels, as is often the case with institutional air monitoring (Ottinger 2010). ‘These techniques also are reported to neglect the prospect of being able to correlate the variations in short-term intra-day atmospheric parameters’ (Sahu et al. 2020: 1347). In other words, low-cost technologies have allowed for the atmospheric sciences to become more accurate and precise.

In Accra, academics and EPA experts are deploying a wide range of LCS, all portable direct-reading PM2.5 monitors, such as the Chinese ZeFan continuous monitor and the US monitors Clarity Node, PurpleAir, and QuantAQ Modulair. These LCS are all based on light-scattering techniques to measure mostly PM2.5. As Figure 4.1 shows, ‘particles flow through the measurement cavity, the light intensity of the infrared/red light reaching the phototransistor is modulated by the presence of particles in the light path’ (Giordano et al. 2021: 3). Put differently, the laser detects the particles and counts them (as can be seen in Figure 4.1, representing the resulting output for the photodiode signal in a low-cost PM sensor with particles absent (above) and present (below) in the light path).

These LCS are also integrated into digital infrastructures (computer servers, mobile applications, websites) that make the circulation, processing, and formatting of data possible (Parasie and Dedieu 2019). LCS require infrastructure comprising an electrical network, Wi-Fi, and data-management centres. Start-up companies are stepping up to produce affordable, easy-to-use, and portable wireless PM sensors to monitor air pollution (Sahu et al. 2020). These developments are shaping a new market for air data.

Clarity Node and PurpleAir are the LCS most commonly used by academics working in Accra to measure PM continuously and in real-time. Both companies sell their LCS at the same price and under the same conditions, irrespective of who the customer is and where they are located in the Global North or Global South. Beyond this common trait and the fact that they are both American, the two companies have differing market models and strategies.

PurpleAir is considered to be the pioneer in the field of LCS. Its device requires Wi-Fi access to upload the data to the cloud and a power source. This can be highly challenging in Africa, where power cuts and connectivity issues are common. Nevertheless, there is the option to store data on an SD card. As this is an open data system, there is no need for a license to use the LCS. The cost of the device is around US$250, which is very cheap in comparison to the price of the reference-grade monitors, but still represents a significant amount of money for citizens of the Global South and at the margins. PurpleAir sells sensors to a wide range of actors, such as private citizens, academics, and NGOs working to develop citizen science. On its website, the company claims that it ‘makes sensors that empower community scientists who collect hyper-local air quality data and share it with the public’.6

Clarity was created in 2014 in collaboration with UC Berkeley.7 It focuses on PM2.5 and NO₂, and its devices are manufactured in Taiwan for economically advantageous reasons related to contract manufacturing conditions. Clarity Node devices are all solar-battery powered and have a cellular connection, which makes them significantly easier to use in places where it can be challenging to maintain infrastructure, both in developed economies and in lower- to middle-income countries.

One of the key differences, compared to PurpleAir, is that Clarity does not sell sensors to private individuals and, therefore, does not support individual citizen scientists. Clarity works directly with government agencies, researchers, and organised community groups, selling them its devices. This is because Clarity’s experts want direct access to the reference data produced by regulatory authorities, for instance, with the two fixed reference-grade monitors belonging to the EPA-Ghana. Accumulating data from these monitors would enable Clarity technicians to calibrate their low-cost devices and to ensure that their data are accurate or, if necessary, to apply correction factors. This would give them total control over the system and allow the company to create even more value.

According to Clarity’s official statement, citizen science–type models cannot achieve this level of quality assurance and quality control around data correction and calibration. This is partly why Clarity likes to partner with government agencies. The company can then design and deploy LCS networks in connection with specific sources, policies, and interventions to objectively support certain operational decisions. Clarity considers that this is where LCS are most useful: used together with regulatory monitoring as an extension or a supplement.

This approach explains why, contrary to PurpleAir, the Clarity Node LCS are not the property of their users. Customers have to pay an annual service subscription fee (licence fee) of US$1,200 – a relatively high cost for academics in the Global South – which includes hardware, hardware replacements, software, cloud services, and full-time project support. This license has to be renewed every year, which can be a financial obstacle to its use by academics, among other actors. The data are the property of the clients, although they can be made available to everyone, including the company, as open access data on a non-profit platform, Open AQ (https://openaq.org/#/), and on Clarity Open Map, a simplified platform. The idea, with these platforms, is to pool all the data and standardise them, with a view to supporting global or regional data sharing.

Clarity acts as the custodian of the data that are kept on its servers and does not monetise the data in the sense of selling them. The company supports Open Data, which it feels constitutes a powerful tool for improving air quality management and fostering community engagement. However, air quality data and the data collected on how customers use the sensors help the company improve its LCS operation. The company can thus monetise these improvements by increasing their licence fees and thereby creating market value. For example, Clarity has fitted its LCS with mini-solar panels to ensure their energy self-sufficiency, as well as an SD card slot to alleviate internet connectivity problems.

As part of the Pathways to Equitable Healthy Cities partnership, Ghanaian and US academics have also deployed a ZeFan device, a Chinese low-cost real-time continuous monitor that measures PM2.5 concentrations at one-minute intervals, which is extremely basic (Alli et al. 2021). Minute-by-minute measurements of weather variables are also recorded by a portable weather meter that tracks several weather parameters, including temperature, relative humidity (RH), and the heat index, which can affect the accuracy of air pollution data. Results show that at all sites, PM2.5 peaks at dawn and dusk, coinciding with rush-hour commuting, and that nitrogen oxide (NO_x) levels, mainly from vehicles, are increasing.

Although the forms and presentations of these LCS vary, from the simplest device, such as ZeFan, to the most elaborate one – such as the Clarity Node with its solar panel – all contain the same components. The main one, and the most important, is the Plantower sensor, which is made in China.8 It consists of a small rectangular metal case containing a miniature fan, a printed circuit board, and a laser (as shown in Figure 4.3).9 All companies producing LCS around the world, whether American or Indian, depend on China to procure this sensor.

The performance of these LCS in the ambient atmospheric environments in which they are being used, however, has yet to be thoroughly evaluated (Lewis et al. 2016). Some of the LCS are far from efficient, even with correction factors applied, and most of them deteriorate very quickly when exposed to humidity and very high temperatures. The performance and accuracy of LCS are significantly affected by particle-size distribution, which changes depending on the source of the pollution and by the variability of meteorological parameters like temperature and humidity, which can be extreme in Global South countries such as Ghana. Data from uncalibrated LCS are insufficient to inform policy decisions. Thus, in addition to testing sensors in situ in order to adjust them to the local environment (Pritchard et al. 2018), calibration work is necessary. In the last section, we present and analyse this calibration work in practice, and the role of Ghanaian scientists, by examining the Air Quality Research Laboratory of the UG Department of Physics.

From free and invisible scientific labour to legitimised Ghanaian science

One of the authors of this paper carries out the LCS calibration process himself, as observed by the other author. Calibration is a fairly straightforward process, which involves positioning LCS alongside fixed reference-grade monitors for a certain amount of time – between three weeks and four months – in order to ensure that the data they measure are of high quality. The ultimate goal is to identify correction factors that can be applied to correct these LCS when they are over-recording or under-recording. It also helps to test their robustness in such extreme conditions.

The data collected from the LCS and reference-grade monitor are then downloaded onto the UG Physics Department computers. The computers are equipped with specific software used to analyse the data and convert them into a graph made up of curves. The curves showing the respective PM data from the LCS and the fixed reference-grade monitor are displayed on the screen and compared. Even if they do not overlap exactly, the curves should at least follow the same trend. The correction process is then carried out to reduce the discrepancy between the two curves (the blue and black in Figure 4.4b), often due to under- or overestimation by the LCS. Calibration allows for different types of knowledge to be held alongside each other.

In 2021, one of the authors of this paper carried out one of the first LCS intercomparison studies in Africa at the UG Department of Physics. For four months, he and his team collocated twenty LCS – two QuantAQ Modulair, two PurpleAir, and sixteen Clarity Node devices – with the fixed reference-grade monitor on the department’s rooftop. The LCS were also collocated with a meteorological station measuring temperature, relative humidity, and wind speed and direction.

The collocation site on the university campus is an urban area with low-density housing, few trees, and relatively limited traffic flows. The nearest road is 500 metres away. There are no known large combustion facilities or other emission sources near the site. The objectives of the study were to assess the LCS performance in terms of precision and reliability, to evaluate how each type of LCS correlated with the fixed reference-grade monitor, and to compare the use of four machine learning models to correct the LCS data. Machine learning is a branch of artificial intelligence (AI) that can learn from feedback (McFarlane et al. 2021). The Ghanaian scientists have used the data produced by the reference monitor and the LCS to develop the most accurate correction factors for calibrating the LCS data. Doing so, these LCS could be deployed at other locations, and the correction-factor model would correct the data to reflect the values that a reference monitor would have produced at that location.

Each LCS model requires its correction factor, which explains why academics developed four different correction factors for each brand of LCS. The four correction factor models that were developed for PurpleAir, Clarity, and QuantAQ Modulair were then applied to correct PM2.5 data collected using a network of LCS situated around Accra. Thus, if the LCS were overestimating or underestimating values, those biases would be removed by the correction factor equation developed by the academics at UG. In order to stabilise the data, two LCS remain permanently collocated with the fixed reference-grade monitor at UG. As pollution sources change with the seasons, the LCS regularly need to be re-corrected. If the LCS data are not corrected, or if the devices are not used or tuned properly, they can be very misleading, which can generate inaccurate data.

The private companies producing the LCS do not, however, financially compensate for all the calibration work described above and performed by Ghanaian scientists. The literature has already provided critical insights into the free labour exploited through do-it-yourself projects and the maker movement (Davies 2017). In our research, we found that this was an issue not only of free labour, but also of invisible labour (Bangham et al. 2022) by Global South scientists for the benefit of LCS manufacturers located in the Global North. Ghanaian academics carry out, free of charge, all the testing and validation work for LCS manufacturers, completely anonymously, and invisibly for the other customers who benefit from it:

So we actually have full-time project partners working with the Clarity networks, working with the Ghana EPA, helping establish the correction factors, diagnosing technical challenges, replacing faulty devices, providing siting guidance, etc. And it’s very helpful in terms of providing at fixed costs for technical deployments with kind of reducing risks around network failures, connectivity issues. We try to take all these little costs and challenges associated with scaling air quality data and make it very predictable and easy to fund in a budget. So to speak. (VP, Business Development and Partnerships, Clarity Movement Co).

In other words, all the free and invisible labour provided, and the knowledge produced, by Ghanaian academics and scientists is leveraged by the private company to put together a package of services, which it will then sell to these same experts and scientists when they buy their annual service subscription (license) at a cost of US$1,200. Ghanaian scientists are, therefore, part of what some scholars have called ‘polycentric innovation’, a term which ‘designates the global integration of specialised research and development capabilities across multiple regions to create novel solutions that no single region or company could have completely developed on its own’ (Bhaduri 2016: 8). However, they don’t appear as co-developers of innovation, let alone benefit from the economic spin-offs generated by the sale of LCS.

At the same time, since 2006, access to and the calibration of these new technologies has allowed Ghanaian researchers to produce and publish scientific data on air quality (Alli et al. 2021; Arku et al. 2008; Clark et al. 2020; Dionisio et al. 2010a; Dionisio et al. 2010b; Jack et al. 2015; McFarlane et al. 2021; Rooney et al. 2012; Zhou et al. 2013). Ghanaians indeed use insights from data produced by low-cost technologies from the Global North and China, which enable them to develop ‘legitimate’ expertise. By doing so, they exist – and do not perish (Harzing 2010) – in an academic landscape largely dominated by researchers from the Global North who can afford fixed reference-grade monitors for their research. However, this legitimacy is still from the perspective of Global North actors who define what legitimate data are and are not (Garrocq and Parasie 2022). The balance of power remains with Global North knowledge production, despite the active role of Ghanaian scientists, who are still dependent on the values and perspectives of the Global North. Indeed, ‘since the academy evaluates knowledge based on Western standards of reliability and validity, non-Western paradigms will still have to be altered to fit the criteria of Western frameworks’ (Thambinathan and Kinsella 2021: 2). Nevertheless, this allows for Africa’s position in global atmospheric science to shift and has, for instance, led to the organisation of the first Air Sensors International Conference (ASIC) in Africa, held in Accra in October 2023.

Ghanaian researchers’ knowledge, control, and improvement of algorithms to make LCS work properly in African settings have led them to consider developing their own technology so as to be autonomous and no longer have to pay annual licence fees. This has led the Ghanaian physicists at UG to engage in the development and assembly of their own locally made LCS in order to reduce costs and have control of the technology. As we have clearly shown previously, all companies producing LCS around the world, whether American or Indian, depend on China to procure the main component. Therefore, Ghanaian scientists also plan to import the Plantower sensor from China and assemble it locally with a printed control board and a few cables (as shown in Figures 4.5a–c). Ghanaian scientists could, therefore, move beyond the pattern of invisible labour in science and innovation to redesign products and processes, which would mark self-determination and independence of their values and perspectives from Global North academics and start-up companies (Thambinathan and Kinsella 2021).

Conclusion: How are Ghanaian scientists shifting Africa’s position in global atmospheric science?

Ghana is one example in this collective book about how techno-scientific globalisation in the Global South is characterised in the form of serious constraints linked to international inequities. Indeed, Ghanaian atmospheric scientists and experts have always depended on international collaboration, external funds, and Global North instruments and tools to produce air pollution data and knowledge. Despite these hurdles and challenges, we have shown how the Ghanaian Physics Department’s scientific team have managed to shape international collaborations since the mid-2000s, which have resulted in calibrating a fairly large number of LCS on which they depend to conduct atmospheric science research. This important validation work has resulted in a sufficiently robust and large body of data, offsetting the fact that LCS still need to be scientifically evaluated and certified by regulatory agencies.

The Department of Physics at UG is thus becoming an African hub for the calibration and validation of LCS, which allows Ghanaian academics to produce scientific knowledge on air pollution. By appropriating tools and instruments from the Global North, they are creating alternative ways of doing air science to cope with constraints triggered by international inequities. Calibration technologies from the Global North and China enable them to use insights from the data they produce to develop legitimate expertise, which is, however, still determined from the perspective of the hegemonic academic actors. Nevertheless, Ghanaian scientists valorise their data by publishing their research results in internationally renowned scientific journals and making them visible to the global scientific community. Thus, Ghanaian scientists are coping with knowledge production inequities and reconfiguring the research ecosystem by shifting Africa’s position in global atmospheric science.

They are also participating in innovation in a context of scarcity by adapting incoming technologies to local settings and appropriating these technologies with the ultimate goal of being able to assemble locally and control their own technology, which would be a marker of independence from the epistemological and commercial frameworks and ideologies coming from Europe and the United States.

These efforts, however, have taken almost two decades, and local scientists still have a long way to go before they are fully emancipated from the global academic hierarchy and international inequities. In the absence of a national research and innovation policy, Ghanaian researchers are still largely dependent on funding from academic bodies in the Global North. The case study we have presented in this chapter is certainly an encouraging jolt that is shaking up the established order in global socio-academic and techno-industrial structures. However, it also shows that the North American and Chinese companies that produce LCS and their components benefit from these power hierarchies by taking advantage of the free and invisible calibration labour carried out ‘from below’ by the little helpers – meaning local scientists – of global techno-scientific capitalism.

Endnotes‌

References

Alli, A. S., S. N. Clark, A. Hughes, and others, ‘Spatial-Temporal Patterns of Ambient Fine Particulate Matter (PM2.5) and Black Carbon (BC) Pollution in Accra’, Environmental Research Letters, 16 (2021): 074013 https://doi.org/10.1088/1748-9326/ac074a.

Allen, B., ‘Strongly Participatory Science and Knowledge Justice in an Environmentally Contested Region’, Science, Technology, & Human Values, 43 (2018): 947–71 https://doi.org/10.1177/0162243918758380.

Arku, R. E., J. Vallarino, K. L. Dionisio, and others, ‘Characterizing Air Pollution in Two Low-Income Neighbourhoods in Accra, Ghana’, Science of the Total Environment, 402 (2008): 217–31 https://doi.org/10.1016/j.scitotenv.2008.04.042.

Bangham, J., X. Chacko, and J. Kaplan, Invisible Labour in Modern Science (Rowman & Littlefield, 2022).

Bhaduri, S., ‘Frugal Innovation by “The Small and the Marginal”: An Alternative Discourse on Innovation and Development’, [Conference session], Prince Claus Chair Inaugural Lecture, ISS, The Hague, the Netherlands, 23 May 2016 https://doi.org/10.13140/RG.2.1.1290.9682.

Clark, S. N., A. S. Alli, M. Brauer, and others, ‘High-Resolution Spatiotemporal Measurement of Air and Environmental Noise Pollution in Sub-Saharan African Cities: Pathways to Equitable Health Cities Study Protocol for Accra, Ghana’, BMJ Open, 10 (2020): e035798 https://doi.org/10.1136/bmjopen-2019-035798.

Davies, S. R., Hackerspaces: Making the Maker Movement (Polity Press, 2017).

deSouza, P., ‘Air Pollution in Kenya: A Review’, Air Quality, Atmosphere & Health, 13 (2020): 1487–95 https://doi.org/10.1007/s11869-020-00902-x.

Dionisio, K. L., R. E. Arku, A. F. Hughes, and others, ‘Air Pollution in Accra Neighbourhoods: Spatial, Socioeconomic, and Temporal Patterns’, Environmental Science & Technology, 44 (2010a): 2270–76 https://doi.org/10.1021/es903276s.

Dionisio, K. L., M. S. Rooney, R. E. Arku, and others, ‘Within-Neighbourhood Patterns and Sources of Particle Pollution: Mobile Monitoring and Geographic Information System Analysis in Four Communities in Accra, Ghana’, Environmental Health Perspectives (Online), 118 (2010b): 607–13 https://doi.org/10.1289/ehp.0901365.

Gabrys, J., H. Pritchard, and B. Barratt, ‘Just Good Enough Data: Figuring Data Citizenships through Air Pollution Sensing and Data Stories’, Big Data & Society, 3 (2016) https://doi.org/10.1177/2053951716679677.

Garrocq, J. B., and S. Parasie, ‘Peut-On Redistribuer la Surveillance de la Qualité de l’Air? Une Enquête sur les Métrologies à l’Oeuvre dans un Concours Institutionnel d’Évaluation de Micro-Capteurs’, Revue d’Anthropologie des Connaissances, 16 (2022) https://doi.org/10.4000/rac.29150.

Giordano, M. R., C. Malings, S. N. Pandis, and others, ‘From Low-Cost Sensors to High-Quality Data: A Summary of Challenges and Best Practices for Effectively Calibrating Low-Cost Particulate Matter Mass Sensors’, Journal of Aerosol Science, 158 (2021): 105833 https://doi.org/10.1016/j.jaerosci.2021.105833.

Hagan, D. H., and J. H. Kroll, ‘Assessing the Accuracy of Low-Cost Optical Particle Sensors Using a Physics-Based Approach’, Atmospheric Measurement Techniques, 13 (2020): 6343–55 https://doi.org/10.5194/amt-13-6343-2020.

Harzing, A., The Publish or Perish Book: Your Guide to Effective and Responsible Citation Analysis (Tarma Software Research PTY Ltd., 2010).

Jack, D. W., K. P. Asante, B. J. Wylie, and others, ‘Ghana Randomised Air Pollution and Health Study (GRAPHS): Study Protocol for a Randomised Controlled Trial’, Trials, 16 (2015): 420 https://doi.org/10.1186/s13063-015-0930-8.

Katoto, P., L. Byamungu, A. S. Brand, and others, ‘Ambient Air Pollution and Health in Sub-Saharan Africa: Current Evidence, Perspectives and a Call to Action’, Environmental Research, 173 (2019): 174–88 https://doi.org/10.1016/j.envres.2019.03.029.

Landrigan, P. J., R. Fuller, N. J. R. Acosta, and others, ‘The Lancet Commission on Pollution and Health’, The Lancet, 391 (2017): 462–512 https://doi.org/10.1016/S0140-6736(17)32345-0.

Lewis, A. C., J. D. Lee, P. M. Edwards, and others, ‘Evaluating the Performance of Low-Cost Chemical Sensors for Air Pollution Research’, Faraday Discussions, 189 (2016): 85–103 https://doi.org/10.1039/C5FD00201J.

McFarlane, C., R. Garima, C. Malings, and others, ‘Application of Gaussian Mixture Regression for the Correction of Low-Cost PM2.5 Monitoring Data in Accra, Ghana’, ACS Earth Space Chem, 5 (2021): 2268–79 https://doi.org/10.1021/acsearthspacechem.1c00217.

Moizard-Lanvin, J., ‘Sélectionner et Agréger les Ignorances’, Revue d’Anthropologie des Connaissances, 15 (2021) https://doi.org/10.4000/rac.25114.

Morawska, L., P. K. Thai, X. Liu, and others, ‘Applications of Low-Cost Sensing Technologies for Air Quality Monitoring and Exposure Assessment: How Far Have They Gone?’, Environment International, 116 (2018): 286–99 https://doi.org/10.1016/j.envint.2018.04.018.

Negi, R., and A. Ranjan, ‘Delhi’s Fight Against Air Pollution Could Get a Boost if Science Was Decolonised’, Scroll.in, 11 November 2020 https://scroll.in/article/976361/delhis-fight-against-air-pollution-could-get-a-boost-if-science-was-decolonised.

Ottinger, G., ‘Buckets of Resistance: Standards and the Effectiveness of Citizen Science’, Science, Technology, & Human Values, 35 (2010): 244–70 https://doi.org/10.1177/016224390933712.

Parasie, S., and F. Dedieu, ‘À Quoi Tient la Crédibilité des Données Citoyennes? L’Institutionnalisation des Capteurs Citoyens de Pollution de l’Air en Californie’, Revue d’Anthropologie des Connaissances, 13 (2019): 1035–62 https://doi.org/10.3917/rac.045.1035.

Pourraz, J., ‘Production and Circulation of Local Knowledge About Air Pollution and Health Effects in Ghana’, in K. W. Fomba, B. Tchanche Fankam, A. Mellouki, D. M. Westervelt, and M. R. Giordano, eds., Advances in Air Quality Research in Africa (Cham: Springer, 2024), pp. 421–39 https://doi.org/10.1007/978-3-031-53525-3_24.

Pritchard, H., and J. Gabrys, ‘From Citizen Sensing to Collective Monitoring: Working Through the Perceptive and Affective Problematics of Environmental Pollution’, GeoHumanities, 2 (2016): 354–71 https://doi.org/10.1080/2373566X.2016.1234355.

Pritchard, H., J. Gabrys, and L. Houston, ‘Re-Calibrating DIY: Testing Digital Participation Across Dust Sensors, Fry Pans and Environmental Pollution’, New Media & Society, 20 (2018): 1–20 https://doi.org/10.1177/1461444818777473.

Rooney, M. S., R. E. Arku, K. L. Dionisio, and others, ‘Spatial and Temporal Patterns of Particulate Matter Sources and Pollution in Four Communities in Accra, Ghana’, Science of the Total Environment, 435–36 (2012): 107–14 https://doi.org/10.1016/j.scitotenv.2012.06.077.

Rueda, E., E. Carter, C. L’Orange, C. Quinn, and J. Volckens, ‘Size-Resolved Field Performance of Low-Cost Sensors for Particulate Matter Air Pollution’, Environmental Science & Technology Letters, 10(3) (2023): 247–53 https://doi.org/10.1021/acs.estlett.3c00030.

Sahu, R., K. K. Dixit, S. Mishra, and others, ‘Validation of Low-Cost Sensors in Measuring Real-Time PM10 Concentrations at Two Sites in Delhi National Capital Region’, Sensors, 20 (2020): 1347 https://doi.org/10.3390/s20051347.

Thambinathan, V., and E. A. Kinsella, ‘Decolonising Methodologies in Qualitative Research: Creating Spaces for Transformative Praxis’, International Journal of Qualitative Methods, 20 (2021): 1–9 https://doi.org/10.1177/16094069211014766.

Wooldridge, A., ‘The World Turned Upside Down: A Special Report on Innovation in Emerging Markets’, The Economist, 15 April 2010 http://www.economist.com/node/15879369.

World Health Organization, WHO Global Air Quality Guidelines: Particulate Matter (PM2.5 and PM10), Ozone, Nitrogen Dioxide, Sulfur Dioxide, and Carbon Monoxide (World Health Organization, 2021).

Zhou, Z., K. L. Dionisio, T. G. Verissimo, and others, ‘Chemical Composition and Sources of Particle Pollution in Affluent and Poor Neighbourhoods of Accra, Ghana’, Environmental Research Letters, 8 (2013): 044025 https://doi.org/10.1088/1748-9326/8/4/044025.