The Madden–Julian Oscillation (MJO), as a key driver of global weather and climate anomalies, is an important source of subseasonal predictability. However, most climate models still struggle to reproduce its fundamental characteristics, posing a critical challenge that urgently needs to be addressed in climate prediction. Previous studies have pointed out that the convective adjustment timescale (tau) is one of the key parameters affecting MJO simulation in climate models, but its sensitivity remains under debate.
A research team led by Professor Lu Wang from Nanjing University of Information Science and Technology, China, and her PhD student Xuan Zhou, conducted a set of sensitivity experiments with refined tau values in CAM6 to clarify the role of tau in MJO simulation. Their paper was recently published in Atmospheric and Oceanic Science Letters , titled "Sensitivity of MJO simulation to the convective adjustment timescale in CAM6".
According to the study, the overall MJO simulation biases tend to decrease as tau values increase, with a critical threshold of 2 hours. Specifically, when tau > 2 hours, the MJO simulation bias clearly responds to changes in tau. In contrast, when tau ≤ 2 hours, the simulation bias is relatively insensitive to tau variations.
The research further revealed the physical mechanism behind this phenomenon: a larger tau value suppresses convective precipitation, allowing more moisture to accumulate in the boundary layer. This leads to larger intraseasonal moisture perturbations in the boundary layer, which is crucial for MJO development and propagation. Conversely, a smaller tau value (≤ 2 hours) results in insufficient boundary layer moisture accumulation, thereby failing to supply the strong intraseasonal moistening required for the MJO.
"Our results reconcile the contradictory conclusions from previous studies regarding the impact of tau variations on MJO simulation biases," says Prof. Lu Wang, the corresponding author. "Research on the physical parameters affecting MJO simulation biases contributes to the targeted optimization and improvement of climate models."
"The next step is to investigate the impact of nonlinear terms among multiple physical parameters on MJO simulation biases", adds Prof. Wang. By employing the PPE (perturbed parameter ensemble) approach, they aim to quantitatively assess the independent and synergistic contributions of physical parameters to MJO simulation biases, identify the key parameters, and reveal the underlying physical mechanisms.
