COAMPS AND NAVY ESPC FORECAST SEASONAL AND DIURNAL VARIABILITY IN PISTON 2018-2019

Abstract:

Principal Component Analyses (PCA) was used to better understand variability in amplitude and time evolution of the diurnal cycle over the Maritime Continent (MC) and how this is modulated by: (1) regionally-varying air-land-sea interaction; and (2) the passage of larger scale conditions associated with the Madden Julian Oscillation (MJO) and Boreal Summer Intraseasonal Oscillation (BSISO). The nonlinear effects of these multi-scale physical processes are not well understood because they work in tandem with each other. The ONR sponsored Propagation of Intra-Seasonal Tropical Oscillations (PISTON) field campaign combined intensive observation periods during Boreal Fall of 2018 and 2019 in the Western Pacific (WP) with numerical studies to better understand these interactions and improve the model representation. A variety of dynamic conditions were observed during PISTON including BSISO and MJO passage, monsoon transition, tropical cyclone genesis and transit, equatorial Mixed-Rossby and Rossby wave generation, mesoscale convective complexes, and shallow, altocumulus, and cumulus congestus clouds. We will highlight results from the Navy’s Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS) and global Earth System Prediction Capability (ESPC) real-time forecasts during the PISTON 2018 and 2019 field campaigns. Comparisons of model data with observational data from PISTON, NASA’s Cloud, Aerosol, and Monsoon Processes Philippines Experiment (CAMP2Ex), and Taiwan’s South China Sea Twin Island Monsoon Experiment (SCSTIMX) reveal that, despite some model bias, both models were able to capture essential elements of the atmosphere. Assessment of the large-scale environment showed a strong correlation between typhoon cyclogenesis and intensification and BSISO phase 6-8 but not necessarily with MJO phase 6-8. The Navy ESPC forecast was found to have skill to about 17 day lead times in capturing the changes in convection extent associated with BSISO and MJO phase. However, the predictability on the northward propagation of convection is only 7 days. We attribute this increased model error to inaccurate multiscale influence of typhoon and equatorial waves at the spatial scale of the model because the northward propagation of BSISO is influenced by the westward moving convection. Over the MC region, Navy ESPC’s prediction bias in diurnal cycle leads to excess land locking convection which then projects to ESPC’s intraseasonal time scale model biases.