East African Digital Solutions to Air Quality Network

Air quality in most East African cities has declined dramatically over the last decades and it air pollution is now the leading environmental risk factor for human health. There is a critical lack of data to assess air quality in East Africa, and therefore to quantify its effect upon human health. Air quality networks in East Africa are still in their early days, with the long term and systematic measurement of air pollutants only available at less than a handful of sites. Large spatial and temporal gaps in data exist. From a historical perspective, very little is known of air pollution concentrations before 2010. The lack of historical data makes it extremely difficult to assess the deleterious effects of air pollution upon human health. It also poses challenges for assessing the efficacy of air quality interventions. Hence informed decisions about infrastructure, which take air quality into account are difficult to make.

This proposal forms a new network to co-create strategy and protocols to bring together data that relate to air pollution in East African Urban areas. It targets the capitals of Ethiopia (Addis Ababa), Kenya (Nairobi) and Uganda (Kampala). New data science techniques will be developed to synthesize disparate data streams into spatially and temporally coherent outputs, which can be used to understand historic, contemporary and future air quality. The proposal will provide a road map to harness the power of new data analytics and big data technologies.

To design this roadmap, three high intensity workshops and interspersed virtual meetings will be undertaken in Stage 1. Each workshop will tackle a key knowledge gap or development challenge:

- Workshop 1: Parameterizing the data problem in East Africa for assessing the causes and effects of air pollution (Kampala)

- Workshop 2: Big data approaches to improve East Africa air quality prediction (Addis Ababa)

- Workshop 3: Creating greater capacity and capability in analytic air quality science (Nairobi)

The Stage 1 research outcomes will enable the development of tailored mitigation strategies for improving air quality. The methodologies developed in the proposal will be translatable and scalable throughout urban East Africa. Hence, the proposal will help realise multiple sustainable development goals (SDGs), including SDG3: Good health and well-being, SDG11: Sustainable cities and communities, and SDG17: Partnerships for the goals.

To ensure the project reaches its maximum potential, it includes an extensive array of research translation activities: workshops with academic and non-academic stakeholders; a professionally designed website, which will hold both academic and non-academic outputs including open source academic papers and presentations; briefing notes directed at a range of external stakeholders, including top down governance and bottom up grassroots organizations.

Project partners from business, academia, governance and public engagement with science are involved and will attend the workshops. They are Uber, Amazon Web Services, PA Consulting, Kampala Capital City Authority, African Population Health Research Centre, Birmingham Open Media, GCRF Multi-Hazard Urban Disaster Risk Transitions Hub, and the Alan Turing Institute. They offer an additional £102,951 of in-kind contributions to the project. Their incorporation widens the available skillsets and will help deliver long-term impact in the East African region.

The project has an extensive array of research translation activities planned:

1. Project website. This will be professionally designed and sufficient budget is requested to ensure the project has a lasting digital legacy for at least five years. Success metrics: (1) high East African engagement rates across all countries and regions within countries, as measured by East African visitor statistics, via website analytics. (2) Generation of business opportunities within air pollution monitoring and prediction within the East African region.

2. Academic outputs. The incorporation of data science techniques into air quality will generate significant impact in multiple academic fields, including atmospheric science, public health and data science. Success metrics: (1) acceptance of at least one perspective piece article on the network outputs in a high impact journal (IF > 5) co-authored by the network with East African representation prominent on the author list. (2) Attainment of Stage 2 DIDA proposal.

3. Comms. The team will liaise with the communications team at the University of Birmingham, African and US CoI organizations and project partner institutions including the Alan Turing Institute, Amazon Web Services (AWS), Birmingham Open Media (BOM), APHRC, Kampala Capital City Authority (KCCA), PA Consulting, Uber to generate media reports on the project outputs and activities to maximise the prominence of the project to multiple stakeholders (governance, practitioners, public, academia). Success metrics: a minimum of two articles published in a major East African newspaper (e.g. the Nation in Kenya, New Vision in Uganda and Capital Ethiopia) during the life course of the project (n.b. UK media outlets will also be targeted).

4. Workshops. Three workshops are the major engines of impact generation in the project.
In addition to their stated aims (see Case for Support) the workshops will provide opportunities for stakeholders (government officials, academics, businesses, and representatives of civil society and the international development community) to share practical experiences of researching air pollution and developing air quality interventions. Success metrics: (1) three workshops will be organised and metrics concerning attendance and media activities will be collated.

5. Briefing notes. The project will produce at least three briefing notes targeting different stakeholder audiences. An indicative series has been identified and will be refined in the webinars prior to the workshops: 'Available data sources for predicting urban air quality in East Africa: the usual and the unusual'; 'The use of data science to generate better temporal and spatial air pollution data'; 'Translation of data science outputs to governance stakeholders'. Success metrics: (1) briefing notes are accessed and used by key stakeholders with references to briefing notes by stakeholders or requests for support collated.

6. Project Studentship. A summer studentship will be funded by project partner PA Consulting to undertake a a business orientated project that is cognate with the aims of the proposal. Success metrics: (1) Successful completion summer placement for student, who will gain a greater appreciation of the role of business in science; (2) Briefing note publication; (3) knowledge translation of air quality solutions to PA Consulting.