Novel Integration of Satellite Imagery and Social Science for Anticipating Zoonotic Spillover Events in the Era of Anthropogenic Change

Initially we will carry out a global scale review of the most important zoonotic infections, examining the role that habitat changes play in their occurrence. From this broad approach we will focus on candidate diseases (see example ecological concept below for proof-of-concept case study) and use high-resolution satellite data, advanced image processing and machine learning algorithms to identify zoonotic disease occurrence that is associated with alterations in natural habitats and/or land use change. Having identified characteristics that are indicative of disease outbreaks, case studies will be developed to understand how the changing human-animal interface alters exposure risk in humans via elite interviews with key actors involved in environmental management and health governance directed, for example, through our co-supervisor at ONS. We will identify case studies related to ecological concepts, for example, detecting habitat changes that alter host density. Empirical evidence suggests that changes in animal density underlies disease emergence; with population increases leading to zoonotic outbreaks and population crashes resulting in disease fade-outs. Changes in animal density are ultimately linked to resources. Rodents for example, a taxa notable for their high zoonotic competence, are known to fluctuate in density in response to seed availability. Hantavirus, a zoonotic disease with high human mortality rate peaks after high seedset following significant rainfall. In contrast, mast years (periodic super-abundance of tree seeds) leads to high rodent density that can dilutes vector-borne disease transmission. Seed density can be determined using multi-modal/sensory data including satellite images, environmental measurements, and related to disease prevalence/risk. Focused interviews with key actors in environmental management, for example, pest control could provide insight into change in human-rodent interactions in the face of environmental change. Outcome: Satellite imagery, combined with data analytics offers a non-invasive method for predicting spillover events that are linked to habitat change. The integration of social science to understand how the human-animal interface changes provide insight into the mechanisms underlying disease emergence. This interdisciplinary approach is timely to link with project partner ONS's Pandemic Preparedness Toolkit.