27 June 2014 - Since the launch of Vmerge, a research consortium to better understand Rift Valley Fever (RVF) and other vector borne diseases, the Food and Agriculture Organization of the United Nations (FAO) has been working towards increasing knowledge of RVF occurrence linked to the distribution and dynamics of competent vector populations in West Africa. This follows on from similar work conducted by FAO and other institutions in East Africa.
With the use of risk modelling tools, Vmerge aims to predict RVF outbreaks in order to prevent their emergence and mitigate their consequences. Two statistical models using remote sensing tools were proposed by FAO and the Centre de coopération international en recherche agronomique pour le développement (CIRAD). Functioning as early warning tools, both models produce risk maps. Risk maps help to predict relative vector abundance based on landscape and climatic parameters, including rainfall and vegetation indices. By identifying an increased risk of vector amplification in areas where hosts are present, it is possible to identify the areas that are most vulnerable to an incursion of RVF. Such calibrated risk modelling tools will help focus efforts to prevent and control an epidemic.
The main difference between FAO and CIRAD’s proposed risk models regards scale. Temporal and spatial scaling is typically an important issue in remote sensing and Geographic Information Systems (GIS). FAO's model uses satellite images that are updated on a daily basis but moderate in resolution (1 km). This means that the risk maps produced can give an accurate overview of a landscape and can monitor ecological phenomena daily but might be missing detailed information on local climatic variations. CIRAD's model, on the other hand, uses very fine resolution maps, allowing the viewer to accurately identify areas and herds at risk but over a limited terrain, making it hard to extrapolate information about a larger area.
By combining these different scales and sharing information on RVF occurrence and drivers, Vmerge can develop a useful and cost-effective tool capable of identifying the areas with the highest probability of an epidemic outbreak. FAO has initiated and facilitated the development of a Google database where each Vmerge partner has listed the data obtainable in their institute in order to increase transparency and create a platform of all available data for risk modelling research. FAO and CIRAD aim to create compatible and complementary risk models in order to avoid duplications and to increase information exchange amongst Vmerge partners within the spirit of encouraged data sharing.
In view of harmonizing and integrating both risk models, a technical meeting was held at FAO headquarters on 19 and 20 June 2014 to share FAO and CIRAD's findings as well as their preliminary analyses. The Central Veterinary Institute of Wageningen University & Research Centre (CVI-WUR), the Istituto Zooprofilattico Sperimentale dell’Abruzzo e Molise “G. Caporale” (IZSAM) and Euro-AEGIS actively contributed to the discussion by presenting previous case-studies on risk models and scaling issues, reviewing RVF expert knowledge as well as RVF drivers other than climate (e.g. livestock distribution and transhumance routes). The meeting was organised to collaboratively devise a strategic plan for integrating FAO and CIRAD’s risk modelling tools in Senegal and Mauritania. This is the first key step towards identifying as early as possible the initial signs of a potential RVF epidemic in West Africa.