Link Between the Time‐Space Behavior of Rainfall and 3D Dynamical Structures of Equatorial Waves in Global Convection‐Permitting Simulations. Issue 2 (18th January 2023)

• Record Type: Journal Article
• Authors: Jung, Hyunju; Knippertz, Peter
• Is Part Of: Geophysical research letters. Volume 50:Issue 2(2023)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2023-01-18
• Subjects: equatorial waves -- wave identification -- DYAMOND -- representation of convection -- space‐time filtering -- spatial projection; Geophysics -- Periodicals; Planets -- Periodicals; Lunar geology -- Periodicals; 550
• DOI: 10.1029/2022GL100973
• Languages: English
• ISSNs: 0094-8276

Abstract

Abstract: Equatorial waves (EWs) control a considerable portion of tropical rainfall variability but numerical models often struggle to capture them. Increased computing power now enables global simulations with resolved deep convection, which is believed to produce more realistic EWs. Here we identify EWs in global ICON simulations with varying horizontal resolution by (a) filtering rainfall based on space‐time spectral analysis and (b) projecting wind and geopotential onto theoretical wave patterns. The simulations demonstrate that Kelvin, mixed‐Rossby gravity and equatorial Rossby waves are consistently represented, regardless of model resolution and convective treatment. For smaller‐scale inertio‐gravity waves, however, explicit convection appears to be a prerequisite. Surprisingly, the associated rainfall signals are not accompanied by corresponding wind patterns but appear to be connected to mesoscale convective systems. This demonstrates the importance of analyzing rainfall and dynamical aspects of EWs jointly for a robust assessment. Plain Language Summary: Equatorial waves (EWs) are one of the most important weather features in the tropics. They propagate east‐ or westward, and substantially modulate rainfall on spatial scales from hundreds to ten thousands of kilometers and on temporal scales from a couple of days to several weeks. However, weather forecasting and climate models often fail to accurately capture them, and it is a challenge to objectively identify them in observational and model data. Here we investigate EWs in novel high‐resolution global simulations, which represent the Earth's atmosphere as realistically as possible, by employing two different wave‐identification methods, one that characterizes the time‐space behavior of rainfall associated with waves and the other their spatial pattern in wind and pressure. Our results show that some types of EWs exhibit robust and realistic behaviors with little sensitivity to model configurations. However, small‐scale wave types show significant propagating signals in rainfall but no corresponding wind patterns as we would expect from theory. We find that the time‐space filter isolates large clusters of thunderstorms as small‐scale waves, leading to a misinterpretation. Thus, a complementary analysis of wind pattern is required to objectively detect EWs. Key Points: Statistics of equatorial waves (EWs) differ markedly when using wavenumber‐frequency filtering and spatial projection for wave identification Phase speed and variance of large‐scale EWs are weakly dependent on resolution and convective treatment in ICON Mesoscale convective systems project onto westward inertio‐gravity wave rainfall signatures without corresponding dynamical patterns