Underrepresentation of African surveillance data in global arbovirus modeling

Recent expansion of arboviruses across Europe has intensified efforts to understand the ecological and environmental drivers of pathogen emergence. However, global interpretations of arbovirus dynamics remain disproportionately informed. Global interpretations of arbovirus dynamics remain disproportionately informed by a small number of well-surveilled countries and regions, including Thailand, Brazil, Singapore, North America and Europe. However, large regions of Africa, despite hosting high mosquito diversity and extensive human-vector contact, remain substantially underrepresented. In this perspective, we argue that this imbalance constitutes a critical blind spot in global arbovirus surveillance, with implications for both risk assessment and predictive modeling. We suggest that apparent emergence in data-rich regions may partly reflect better detection, occurring against a background of poorly characterized transmission in underrepresented regions in parts of Africa. We explore opportunities for the inclusion of other pathogen-vector systems into arbovirus surveillance, such as malaria in Anopheline mosquitoes, that are outside of the scope of routine arbovirus data collection due to assumptions about vector competence. Furthermore, the exclusion of underrepresented vector systems, including Anopheles mosquitoes, may constrain understanding of broader transmission networks. Anopheles are highly represented in malaria research but underrepresented in arbovirus surveillance and vector competence datasets, especially in African arbovirus modeling inputs. We highlight how fragmented surveillance can influence ecological inference, disease forecasting, and global risk assessment, particularly in increasingly interconnected transmission systems. Finally, we argue that integrating arbovirus surveillance into existing malaria monitoring infrastructure across Africa could provide a scalable and cost-effective strategy for improving global surveillance coverage. Addressing structural surveillance biases will be essential for developing more representative and predictive frameworks of arbovirus emergence.