Abstract

Weather forecasting plays a crucial role in modern society, serving as an essential basis for daily life, disaster response, and policy decision making. However, the development of numerical weather prediction (NWP) is often understood primarily as an advancement in computational technology. This paper explains the evolution of NWP from the first numerical weather prediction experiment using Electronic Numerical Integrator and Computer (ENIAC) to modern NWP, digital twins, and artificial intelligence–based prediction, with an emphasis on making these developments accessible to secondary school teachers. The ENIAC model, modern NWP, and digital twins are compared in terms of their complexity in model, observation and data assimilation systems, approaches to uncertainty treatment, and social application. Through this comparison, NWP is shown to represent not only technological progress toward higher forecast accuracy, but also a gradual transformation in how uncertain future information has been integrated into social decision-making. By presenting NWP as a case in which scientific knowledge, technology, and social consensus have co-evolved, this study suggests that numerical weather prediction can be effectively used by school teachers as an educational topic for explaining the relationship between science and society.