Overcoming Citizen Science Challenges for Cleaner Air
It is time for change. Despite the EU exceeding its 20% cut in GreenHouse Gas (GHG) emissions since 2009, many argue that scientists and politicians are not doing enough.
The EU’s 40% GHG reduction target by 2030 is lauded as being too soft, and its plans for a carbon neutral economy by 2050 is said to be too long. .
Even the UN Secretary General, Antony Gutierrez, argued for a worldwide climate emergency to be called this week (December 2020).
Even though governments are making an effort to cut emissions - for instance, by introducing tax breaks for electric cars and home grants for solar energy - more needs to be done in the short term as well as in the long run to achieve a sustainable future. It's not just the responsibility of government and science to fight pollution and climate change, it is the responsibility of every single person on the planet. As Dr Alison Doig, from the Energy Climate & Intelligence Unit said, reducing GHG is ‘what you do at home’. But, for people to understand how to play their part beyond buying energy friendly appliances, and for governments to make effective incentives for change, they need to know what the local environmental problems are, especially in cities where people’s health is severely affected by toxic air.
Official measuring stations can provide some data on urban air pollutants, general weather conditions etc., but these are on a macroscale and do not necessarily paint a complete picture of a whole city. However, an opportunity to supplement city data for public sector decision making, in a manner that gives ownership of the data back to the public for actionable change, presents itself in the form of citizen science performed by the people living in the very areas that require change.
Citizen science involves the engagement of non-scientists in scientific research. Citizen science projects have long been reported to be useful in policy development but there is surprisingly little detail of how projects have contributed to real decisions, behaviour change and public impact. In fact, research institution Rand recently noted there is a lack of research on models, such as community citizen science, that focus on using data to inform policy or community actions. Successful open citizen science-related platforms like sensor.community (former luftdaten), smart citizen kit and telraam offer very valuable data. Still, despite everyone’s efforts, the data isn’t used for scientific research and policymaking in its full potential.
A recent study (Hecker, Wicke et al.) emphasized the many benefits that citizen science may provide for science, society, and policy. They include boosting the spatio-temporal data collection through volunteers, tapping into distributed knowledge domains, increasing public interest and engagement in research, enhancing societal relevance of the respective research, and increasing scientific literacy and individual skills. Whilst the study recognises the importance of citizen science, it noted there were difficulties in the uptake of citizen science results into actual policy implementation due to a lack of citizen science alignment with current policy structures and agendas. Establishing a fully-fledged, well-aligned relationship between government and citizen science initiatives is easier said than done. Research has identified three recurring barriers that impede effective cooperation:
Lack of participation: Citizen Science initiatives launched after specific incidents noted difficulties in engaging a broad base of volunteers for data-collection efforts following their respective disaster events. Difficulties engaging consistent participation means initiatives struggle to scale and achieve wider impact. Reasons for participation dropping off can include lack of time, technical issues, perception that the effort was unnecessary or no longer relevant. The key is to have an approach which fosters openness and diversity in representation. For best results, the scientific process should be made more participatory by including multiple stakeholders, in particular SES groups, so that the benefits are more reciprocal to all those involved. It has also been found that paying attention to the user experience can significantly improve data harvesting and continued engagement.
Lack of skills/tools: Citizen Science projects often use resources and techniques for their experiments because they are freely available, even though they might not fully fit with the intended purpose of an activity. Using inappropriate technology or approaches means the resulting outputs are questionable. On the other hand, research shows that laypersons can collect data of same quality to experts, if familiarised with the methods.
Lack of trust: Governments and scientists are sceptical about data quality and participatory efforts observed in citizen science initiatives. Vulnerable people, those from lower SES groups, are cited as being poorly represented. The further away citizen science project is from traditional academic research, the more questions are raised about the abilities of citizens to use sound scientific methods. This goes to the extreme, in that data from volunteers is considered undesirable by experts or policy makers and may even be prohibited for official use.
COMPAIR tackles the challenges of trust, skills, and participation in Citizen Science to improve air quality by adopting a quadruple helix approach (involving science, policy, industry, and society). COMPAIR will redefine how different actors interact in order to unlock effective innovation. Despite the quadruple helix concept being around for the last few years, cities tend to fall back on the triple helix and not truly involve society unless in voting or official consultations. However, tackling climate change requires the involvement of the full urban value chain. COMPAIR aims to enable the quadruple helix relationships through both technology and design thinking processes and together will empower the collective to deliver new solutions for enhancing air quality and mitigating climate change in cities.
The initiative envisages the following roles for the four stakeholder groups:
Government: Public authorities are potentially the main beneficiaries of CS activities. They can use the results to inform short- and long-term policy decisions, or to improve existing services by leveraging the outputs of co-innovation resulting from COMPAIR Ideathons and Data Jams. Impact can therefore be on both operational and policy levels provided that collected data is of high quality, something that will be achieved with the help of Academia.
Academia: Researchers lead the experiments and provide best practice examples from the world of science so that CS data, analysis and outcomes meet the necessary quality standards and can be considered robust enough for policy making. Whilst the aim is for the citizens to support and help each other based on their own experiences, there will be some needs that require a more professional researcher’s touch to move the experiments forward. Also, as stated in the objectives, we expect at least three research organisations to use our findings/data in future research.
Business: Businesses can help recruit participants for data collection amongst their members/customers; provide the much-needed industry’s perspective on sustainability; prototype new solutions; and change their own practices in accordance with final results and recommendations.
Society/citizens: Citizens provide detailed data coverage through their citizen science experiments. Being able to visualise and analyse data on both a personal and community level, individuals will be able to make more environmentally friendly decisions on a daily basis and have their voice heard in Government.
As well as taking the quadruple helix approach, COMPAIR sets out to address the three core issues hindering the widespread adoption of CS as follows:
1. Increasing representativeness in participation through trusted relationships
Not only will the project focus on engaging society in problem solving on a local level, it will also deploy special tactics for the involvement of hard-to-reach groups. Studies show that participants in environmental Citizen Science activities tend to be older, white and educated (NASEM 2018), yet environmental problems affect everybody. In fact those most burdened by poor air quality are the least likely to participate in research about it. This includes young people, and those living near industrial areas. As a result, research outcomes are skewed by the lack of participation from those that are historically disenfranchised. COMPAIR aims to increase participation in Citizen Science for more accurate air quality and environmental data and make it more representative through a multi-pronged engagement strategy. COMPAIR will encourage hard-to-reach people to participate in citizen science through (a) the recruitment of existing, trusted community champions working for or with the city council who will engage and support citizen scientists (b) through evidence of local problems using a state-of-the-art Augmented Reality app, and (c) targeted media campaigns both social and traditional.
2. Bolstering skills and science capacity through easy-to-use tools and data visualizations. COMPAIR makes it easier for all citizens to participate in improving air quality and helping to meet Green Deal targets through availability of easy-to-build-and-use air quality sensors; visualisation dashboards that provides a comprehensive environmental map of a city using data on air quality and energy consumption; a simulation dashboard that enables citizens to interact with specific targets and actions while instantly informing them about changes that must happen (on an individual, community, industry level) to make cities more liveable in the future. In pilots with existing Digital Twin platforms (Flanders, Athens), an integration attempt will be made to connect COMPAIR with advanced IoT platforms to provide even better insights for policy making. The achieved results based on the DIY sensors will be actively promoted in the science and CS communities and on CS Sensor platforms like sensor.community and Telraam as best practices to stimulate participants to set up similar projects or re-use the ideas for new innovations.
3. Increasing trust in CSby professionalising approaches to produce action ready data. To help overcome the challenge of data quality, COMPAIR aims to professionalise and at the same time simplify the process of collecting and processing CS data, making it open so its value can be scaled from the local level (for behavioural change) to the city level (policy action) and be reused by professional researchers, businesses and citizens themselves. Trust will be increased in three fundamental ways:
The first is to make data more accessible and usable by improving its publication and availability through the central Information Manager (discussed later). This approach involves aligning the data models with existing international and EU data and meta-data standards (e.g. ISA², OGC, WRC, ISO). Further alignment may be needed with sectoral data models. In the case of air quality, this would be with standards bodies like the European Environment Agency. In the case of transport and mobility, with EU ITS standards.
The second is to improve the data quality itself, by utilising expert calibration algorithms for automated quality assessment and validation (citizens can turn that on and off), which will improve the accuracy of IoT sensors. This approach results in useful information about data quality, allowing one to make conclusions such as "this data point is 80% reliable.”
The third is to broaden the flexibility of the API and make it more tailored for policy use. For instance, this can be done by changing the output from DIY devices from hourly results to a more flexible time window (e.g., 15 or 5 min basis). This would make CS results more relevant for operational decision making.
As trust in CS data increases, so do the opportunities for its use. Generally, citizen science is a good conversation starter to discuss local issues, while CS data that commands trust can also be collated and easily ingested by existing infrastructures (e.g., Digital Twins) to provide continued support for decision makers who are increasingly challenged to look for new data sources to make better policies.