Integrating broad and deep multiple-stressor research: A framework for translating across scales and disciplines

Abstract

Despite the intense hazard interactions in the Anthropocene, risk research is often limited by disciplinary approaches and single-sector or scale analyses, skewing policy advice toward biased, misguided, and unfair outcomes. Research has been locked in a trade-off between reductionism, ignoring the often-conflictive local contexts, and the holistic imperative, which has been a complex and intractable problem. Here, we provide a framework that embraces the complexities of integrating mixed methods, societal sectors, and analytical scales by using a translator agent-based model. This approach innovates by treating the informational transfers explicitly and dialoguing with different disciplines. We implement it to analyze COVID-19 in Brazil, and our mixed top-down and bottom-up evidence markedly differentiates exposure and vulnerability across social classes. This framework overcomes disciplinary siloing, accounts for cross-sectoral losses, and tracks feedback between environmental and social factors. These innovations are key for promoting evidence-based and context-sensitive policies essential for fairer and more effective adaptation.