Consequences of systematic model drift in DYNAMO MJO hindcasts with SP‐CAM and CAM5

摘要: Hindcast simulations of MJO events during the dynamics (DYNAMO) field campaign are conducted with two models, one conventional parameterization (CAM5) and a comparable model that utilizes superparameterization (SP-CAM). SP-CAM is shown to produce qualitatively better reproduction fluctuations precipitation low-level zonal wind associated first DYNAMO compared CAM5. Interestingly, skill metrics using real-time multivariate index (RMM) suggest opposite conclusion CAM5 has more than SP-CAM. This inconsistency can be explained by systematic increase RMM amplitude lead time, which results from drift large-scale in projects strongly onto index. hindcasts exhibit contraction moisture distribution, extreme wet dry conditions become less frequent time. reproduce observed but also have stronger patterns budget terms, such as an drying meridional advection tendency appears enhanced off-equatorial synoptic eddy activity Systematic tendencies similar magnitude intraseasonal tendencies, therefore important for understanding prediction skill.