Learning by Doing: Seasonal and Diurnal Features of Tropical Precipitation in a Global‐Coupled Storm‐Resolving Model

Segura, H. ORCIDiD
Hohenegger, C. ORCIDiD
Wengel, C.
Stevens, B. ORCIDiD

DOI: https://doi.org/10.1029/2022GL101796
Persistent URL: http://resolver.sub.uni-goettingen.de/purl?gldocs-11858/11594
Segura, H.; Hohenegger, C.; Wengel, C.; Stevens, B., 2022: Learning by Doing: Seasonal and Diurnal Features of Tropical Precipitation in a Global‐Coupled Storm‐Resolving Model. In: Geophysical Research Letters, 49, 24, DOI: https://doi.org/10.1029/2022GL101796. 
 
Hohenegger, C.; 1 Max Planck Institute for Meteorology Hamburg Germany
Wengel, C.; 1 Max Planck Institute for Meteorology Hamburg Germany
Stevens, B.; 1 Max Planck Institute for Meteorology Hamburg Germany

Abstract

Using the global and coupled ICOsahedral Nonhydrostatic model with the Sapphire configuration (ICON‐S) and a grid spacing of 5 km, we describe seasonal and diurnal features of the tropical rainbelt and assess the limits of ICON‐S in representing tropical precipitation. ICON‐S shows that, by resolving meso‐beta scale process, the rainbelt structure and its seasonality (zonal and meridional migration and enlargement) is reproduced, with better performance over land than over ocean and with a very high degree of agreement to observations. ICON‐S especially struggles in capturing the seasonal features of the tropical rainbelt over the oceans of the Eastern Hemisphere, an issue associated with a cold sea surface temperature (SST) bias at the equator. ICON‐S also shows that a perfect representation of the diurnal cycle of precipitation over land is not a requirement to capture the seasonal features of the rainbelt over land, while over the ocean, 5 km is sufficient to adequately represent the diurnal cycle of precipitation.


Plain Language Summary: Over the tropics, precipitation falls in distinct bands, that span the circumference of the Earth. These bands migrate from the Northern to the Southern Hemisphere and vice versa following the seasonal migration of the sun. Their center of mass also varies east‐west, as well as their area. Where rain ends up falling is of key importance but conventional climate models relying on statistical approaches to simulate convection cannot represent these characteristics. Here we report on the results of simulations on a global domain and, to our knowledge, for the first time integrated with an atmosphere‐ocean coupled over a full seasonal cycle and with a grid spacing fine enough to explicitly represent convection and Mesoscale Ocean eddies. We show that such simulations can reproduce many aspects of the seasonal migration of the rainbelt over land. For instance, the north‐south and east‐west migration of the rainbelt as well as its expansion during the summer season are well captured. This is also the case for the rainbelt in the eastern Pacific and the Atlantic, but not in the Eastern Hemisphere, where the poor representation of the sea surface temperature pattern distorts the representation of the rainbelt and its seasonal characteristics.


Key Points:

In one year of simulation, the ICOsahedral Nonhydrostatic model with the Sapphire configuration (ICON‐S) reproduces the seasonal features of the tropical rainbelt over land with high agreement with observations.

In the eastern Pacific and Atlantic, the seasonal structure and movement of the rainbelt are also reproduced by ICON‐S.

Biases in sea surface temperature explain the struggles of ICON‐S in simulating the oceanic rainbelt of the Eastern Hemisphere.