The Atlantic Ocean's current system stabilizes global climate, but scientists fear its collapse. Climate assessments often overlook how uncertainty in the model parameters affect predictions. This study addresses this gap and provides a framework for this issue. Our findings show that sea surface salinity  reduces prediction uncertainty of the occurrence of current slow down, whereas the remaining uncertainty is huge, which implies importance of considering the uncertainty in model parameters.

Many cities around Northeast Indian Ocean lack long records needed to estimate rare storm surges. Furthermore, estimating rare surges with only one statistical method can hide uncertainty. We analyzed 1950–2024 reanalysis surge data for 11 cities using an ensemble approach. Results show that rare events became more frequent in some central basin cities but less frequent in western and outer-basin cities. The study shows that using ensemble approach gives a clearer picture of present surge risk.

We studied how best to handle uncertainty in hydrological models that simulate how rain becomes river flow. We tested a well-known Bayesian way to estimate uncertainty in model settings as well as in model design. This approach helps judge which models are better considering their uncertainty. However, considering uncertainty in model settings is not helpful to provide river flow forecasting, so costly Bayesian tuning is rarely justified for practical purposes.

How drought impacts communities is complex and not yet fully understood. We examined a disaster dataset and compared various drought measures to pinpoint affected regions. Our new combined drought indicator (CDI) was found to be the most effective in identifying drought events compared to other traditional drought indices. This underscores the CDI's importance in evaluating drought risks and directing attention to the most impacted areas.

It is generally difficult to control large-scale and complex systems, such as Earth systems, using small forces. In this paper, a new method to control such systems is proposed. The new method is inspired by the similarity between simulation-observation integration methods in geoscience and model predictive control theory in control engineering. The proposed method is particularly suitable to find the efficient strategies of weather modification.