In the field of Numerical Weather Prediction (NWP), understanding the system's sensitivity to initial conditions can unlock a world of possibilities, including effective control mechanisms. Join us in this intriguing episode as we host Dr. Miyoshi and Dr. Sun, who delve into this chaotic yet fascinating world of NWP.

The duo discusses the Observing Systems Simulation Experiment (OSSE), a widely recognized approach for studying predictability where an independent NWP model run synthesizes "nature." They talk about their extension of the OSSE to design the control simulation experiment (CSE), where they apply a small signal to control "nature."

Through the lens of idealized experiments with the Lorenz-63 three-variable system, Dr. Miyoshi and Dr. Sun reveal that it's possible to control "nature" to remain in a chosen regime, thereby avoiding a shift to another. This happens by adding small perturbations to "nature." Even more fascinating is the fact that they achieve more effective control with a perturbation size of less than only 3 % of the observation error when using longer-lead-time forecasts.

The episode culminates in an insightful discussion of the possible applications of CSE in real-world NWP systems, including potential reductions in weather disaster risks. The applicability of CSE to other chaotic systems beyond NWP is also discussed.

Join us as we navigate the complex yet rewarding world of numerical weather prediction with Dr. Miyoshi and Dr. Sun.

Keywords: Dr. Miyoshi, Dr. Sun, Numerical Weather Prediction, NWP, Observing Systems Simulation Experiment, OSSE, Control Simulation Experiment, CSE, Lorenz-63, Chaotic Systems, Weather Disaster Risks.

https://doi.org/10.5194/npg-29-133-2022 Control simulation experiment with Lorenz’s butterfly attractor