Three years have passed since late January 2020, when a new coronavirus broke out and reports began to come in about infected people in the country. 2020 was also the year when the decarbonization movement took off. Increased awareness of weather and climate science predictions and a widely shared global climate change crisis drove the move toward decarbonization. Pandemic modeling studies may also draw lessons from the process by which weather and climate science has improved its methods and insights to prepare for future outbreaks.

Recently, the journal Proceedings of the National Academy of Sciences (PNAS) published an article titled Learning from weather and climate science to prepare for a future pandemic.

The paper emphasizes the need to acknowledge and incorporate uncertainty in the development of forecast models. For example, the accuracy of weather forecasts depends on how accurately atmospheric conditions are predicted, but errors are introduced by observation methods and representativeness. Observations from stations have a ground-based bias, while satellite data, while having smaller deviations, suffer from the disadvantage of poor vertical resolution. Similarly, in SARS-CoV-2, the detection of virus levels in domestic wastewater, false positives and false negatives in testing, changes in testing strategies etc. all involve errors. Even under the assumption that the current situation is perfectly estimated, future predictions are not immune to error due to the shortcomings of the model. In weather and climate models, there is also the challenge of limited data at precise spatial and temporal resolution and the inability to simplify the effects of processes that are not well understood, such as the impact of clouds on convection. Modeling of infectious diseases also has the challenge of simplifying them, even though the exacerbation of symptoms is strongly influenced by age, gender, behavioral patterns, environmental conditions, and medical history.

Small uncertainties can occur due to incomplete observations or model flaws that produce very different predictions. In weather forecasting, this problem is addressed by running the model multiple times with slightly different initial conditions. On the other hand, in infectious disease forecasting, which differs from weather forecasting, it is difficult to make a complete analogy for the spread of the SARS-CoV-2 mutation with changes in infection routes and disease status. Daily adjustments to pandemic models are necessary for the transmission of new mutant strains and pathogens.

The authors emphasized the importance of lessons learned from weather and climate change science, particularly in the areas of model development, international comparisons, data exchange, and risk communication. For global modelling, there are international comparison projects under the World Climate Research Programme to support the Intergovernmental Panel on Climate Change. Regarding data sharing, a data sharing program has been in place since 1950 under the World Meteorological Organization, which has been a role model for long-term international collaboration and cooperation that did not falter during the Cold War era. The discussion proposed international data sharing for infectious diseases as well, including monitoring of domestic wastewater to predict future outbreaks of pandemics. Lessons can also be learned from the climate change experience regarding communication about risk and uncertainty. More frequent and accurate dissemination of scientific information does not necessarily lead to rational decision making. The lesson from climate change science is that it is sometimes effective to take situation-specific communication measures based on social science methods.

The paper emphasizes that interdisciplinary collaboration is more important than ever when facing complex crises. It was not that weather and climate research is superior to other fields of study, but suggested that infectious disease modeling research has much to learn from weather and climate scientists with half a century of experience with uncertainties, forecasting, global data exchange, and public debates.

Reference
Schemm S. et al. (2023) Learning from weather and climate science to prepare for a future pandemic. January 17, 2023 PNAS 120 (4) e2209091120 https ://www.pnas.org/doi/10.1073/pnas.2209091120

Contributor: IIYAMA Miyuki (Information Program)