Accurate predictions of weather and climate have enormous social and economic values but remain to have significant uncertainties at different time and spatial scales. The difficulties and challenges may come from uncertainties in the initial conditions and/or errors in the forecast models. Moreover, nonlinear processes, leading to a chaotic behavior, may lead to error growth and then the loss of predictability. No forecast is complete without an estimate of the prediction error.
This session will focus on the studies of error growth related to predictability. Contributions related to theoretical, numerical (from idealized models to comprehensive earth system models), and/or observational research are greatly encouraged. The main topics could include (i) exploring approaches to estimate error growth that limits predictability for weather and climate forecasts; (ii) quantifying practical predictabilities of both deterministic and probabilistic approaches for different weather and climate phenomena from thunderstorms, torrential rains, tropical cyclones and typhoons to ENSO and climate change with time scales ranging from a few hours to centuries.; (iii) exploring approaches to reduce/control the above uncertainties such as ensemble forecast, targeted (i.e., adaptive) observations, and data assimilation.