In this project, we will collaborate with researchers from six West African countries (Nigeria, Benin, Togo, Ghana, Côte d'Ivoire and Burkina Faso), which are part of the Bill and Melinda Gates Foundation and the Department for International Development project WAVE (West Africa Virus Epidemiology for Root and Tuber Crops), to design effective control and management strategies for these cassava diseases. Our research aims to assess disease control methods that could maximise yield in a cost-effective manner. Current potential control methods for CMD and CBSD include using resistant or tolerant cultivars, removing infected plants and restricting trade. From our previous work on the control of cassava diseases, we know that implementing these measures may not always be straightforward. For example, trade restrictions limit the dispersal of the disease, but also slow the dispersal of new varieties through the informal trade sector. This suggests that control through a combination of strategies requires careful planning. Recently the Bill and Melinda Gates Foundation and the Department for International Development have awarded the project "West African Virus Epidemiology for Root and Tuber Crops" (WAVE). The WAVE project aims to collect data to underpin the development of disease control strategies. We currently support the WAVE project with sampling guidance, however, within WAVE there is no capacity to use the data to develop models and produce a set of effective control options for multiple diseases simultaneously. Our proposal aims to identify, using modelling in combination with the data from the WAVE project, how best to coordinate a combined approach to controlling these diseases based on the use of resistant and tolerant planting material, which will help decision-makers across the region to plan how best to implement disease control strategies to alleviate the yield loses caused. In order to assess the most effective cassava disease control strategies, we will begin our work by modelling the distribution of cassava in the region, using the most recent satellite population, cropland distribution and cassava production data. We will use this host distribution map to advise on sampling strategies, as well as offering statistical and data management support throughout. We will then develop a model for the spread of CMD and explore factors that deliver robust control of the pathogen. We will adapt a previous model on the spread of CBSD to a West-African context, and will determine both the risk of introduction and the likely rate of spread should it reach West Africa. We will then identify factors that are effective in rapid containment and eradication of the disease. Finally, we will combine the models to consider the dispersal and control of both diseases simultaneously. We will use this to advise WAVE collaborators on the best use of control strategies in order to increase the likelihood of successfully managing CMD while retaining the ability to eradicate CBSD incursions. This will lead to significant reductions in yield losses attributed to both diseases for cassava growers across the region and a subsequent increase in population welfare.

Poor air quality damages the lives and livelihoods of millions of people and is predicted by the World Health Organisation (WHO) to become the world's largest cause of preventable death by 2030. Those living in Low- and Middle-Income Countries (LMICs) and cities are particularly affected, both through short-term acute effects and an accumulated life-long reduction in quality of life and health. There is a major opportunity to co-design and co-produce a highly fault-tolerant system for air pollution measurement, that is fully open-source, and built from easily available low cost and off-the-shelf components. The ambition is that this approach would be scale-able and could be sustained in LMICs by in-country practitioners at modest cost, long-term. New measurements can then be coupled to integrated assessment models developed by in-country agencies with our support to enhance their decision-making capacity on air pollution mitigation. This modelling will use a tool developed by project partners in the University. This new innovation for monitoring and modelling, can catalyse action and support long-term beneficial change, initially in our early adopter partner countries, and then applied to other LMICs. Recent research from the University of York's Wolfson Atmospheric Chemistry Laboratories (WACL) has developed a low power, highly fault tolerant technology based on the clustering of multiple low-cost air pollution sensors to provide high quality measurements of target air pollutants. This approach exploits the simplicity, modest cost and high reliability of state-of-the-art sensors and electronics, but significantly improves the quality of data collected. The real-world use of sensor technologies has been slowed due to issues relating to poor individual sensor data quality. York have developed a technology that uses multiple sensors of the same type to solve the two key outstanding barriers to application in LMICs, that of sensor-to-sensor variability and unexpected sensor failure. The aim is to enable a self-supporting user community that can build and fix its own instruments and help improve on our initial designs. This approach differs fundamentally from the prevailing paradigm of a top-down commercial services model which has for many years failed to function in LMICs. The Stockholm Environment Institute centre (SEI) in the Department of Environment and Geography at the University of York has been working with the Ministries of Environment in Togo and Cote d'Ivoire and the Ghana Environment Protection Agency, and the University of Lomé, Togo and Université Félix Houphouët-Boigny in Cote D'Ivoire to develop national models using LEAP-IBC (developed by SEI), to support national low-emission planning. We will build on this work applying LEAP-IBC to Lomé, Abidjan, Accra, and another Ghanaian city (e.g. Kumasi) where no such tool is available, and there is limited or no regular monitoring. This will allow them to develop emission inventories of key air pollutants, baseline and mitigation emission projections, and to estimate the resulting concentrations of PM2.5 and the associated human health impacts. We will work with local academics and planners to support the development of the analysis, guiding them through the data collection, model design, model validation and extraction of results. Working with the University of Colorado and WACL, we will further develop the GEOS-Chem Adjoint model inputs to LEAP-IBC that converts emissions in LEAP-IBC to concentrations of PM2.5 and ozone in these cities. The inclusion of this modelling, developed by planners in Ghana, Cote d'Ivoire and Togo will also allow for an understanding of how the monitoring and modelling can be mutually beneficial to provide the evidence needed for the further development, implementation and monitoring of air quality plans in these cities and opportunities to achieve ambient air quality standards in cities.