Quantitative Assessment of Diurnal Variation of Tropical Convection Simulated by a Global Nonhydrostatic Model without Cumulus Parameterization

Abstract

This study investigated the resolution dependence of diurnal variation in tropical convective systems represented by a global nonhydrostatic model without cumulus parameterization. This paper describes the detailed characteristics of diurnal variation in surface precipitation based on three-dimensional data, with the aim of explicitly clarifying the mechanism that underlies the variation. The study particularly focused on the evolution in the size of the precipitation area for deep convective systems with an analysis of the vertical structure of thermodynamic fields. This analysis compares the results of simulations with horizontal grid sizes of 14 and 7 km (R14 and R7, respectively). Over land, the phase delay of diurnal variations in R7 is about 3 h less than that in R14. R7 produces a pronounced diurnal variation in the size distributions of precipitating area(s), especially for areas with a radius of 0-100 km; this characteristic is not found for R14. Such areas actively evolve between noon and evening, leading to the smooth development of larger-scale precipitating areas having a radius of 100-150 km. The maximum surface precipitation in R7 over land occurs at around 2000 local time throughout the tropics, approximately 2 h prior to the development of nighttime deep convection. Deep convective regimes are important as agents of vertical heat transport in the tropics. The present results suggest that precipitating areas with a radius ,100 km make a strong contribution to the total amount of precipitation and to mass transport.

Resolution Dependence of the Diurnal Cycle of Precipitation Simulated by a Global Cloud-System Resolving Model

“…The precipitation peaks appear around 2400 LST in Δ14 and Δ7. Previous studies using NICAM (Sato et al 2009;Noda et al 2012) also reported a midnight peak of precipitation at these resolutions. The peak time changes to around 2000 LST in Δ3.5.…”

“…Global simulations with 14 km and 7 km grid spacing without CP schemes have effectively reproduced the diurnal precipitation cycle in tropical regions (Sato et al 2009;Noda et al 2012;Kodama et al 2015), especially over the ocean. Nevertheless, the precipitation peak in the diurnal cycle over land occurs later than in TRMM observations.…”

Cold and Warm Rain Simulated Using a Global Nonhydrostatic Model without Cumulus Parameterization, and their Responses to Global Warming

“…The configuration of the nonhydrostatic icosahedral atmospheric model, (NICAM) (Tomita and Satoh 2004;Satoh et al 2008Satoh et al , 2014 used in this study was the same as that described by Noda et al (2012). Cloud microphysics processes were computed using a one-moment scheme that predicts the mixing ratios of rainwater, graupel, and snow and also diagnoses cloud water and cloud ice (Tomita 2008).…”

“…For CTL, the SST boundary conditions and sea ice concentrations (SICs) were the same as those used in the boreal summer 2004 experiment of Noda et al (2012). The conditions for GW were developed using the dataset of the World Climate Research Program Coupled Model Inter-comparison Project phase 3 (CMIP3) and the method of Mizuta et al (2008).…”

“…However, using a high-resolution nonhydrostatic GCM without cumulus parameterization, which computes tropical convection explicitly, we can expect to reduce this ambiguity. For example, previous studies show realistic simulations of atmospheric disturbances over the scales of several thousand kilometers as well as diurnal variations (Miura et al 2007;Nasuno et al 2009;Sato et al 2009;Noda et al 2012Noda et al , 2014.…”