Configuration and hindcast quality assessment of a Brazilian global sub‐seasonal prediction system

Abstract

This article presents the Centre for Weather Forecast and Climate Studies (CPTEC) developments for configuring a global sub-seasonal prediction system and assessing its ability in producing retrospective predictions (hindcasts) for meteorological conditions of the following 4 weeks. Six Brazilian Global Atmospheric Model version 1.2 (BAM-1.2) configurations were tested in terms of vertical resolution, deep convection and boundary-layer parametrizations, as well as soil moisture initialization. The aim was to identify the configuration with best performance when predicting weekly accumulated precipitation, weekly mean 2 m temperature (T2M) and the Madden-Julian Oscillation (MJO) daily evolution. Hindcasts assessment was performed for 12 extended austral summers (November-March, 1999/2000) with two start dates for each month for the six configurations and two ensemble approaches. The first approach, referred to as Multiple Configurations Ensemble (MCEN), was formed of one ensemble member from each of the six configurations. The second, referred to as Initial Condition Ensemble (ICEN), was composed of six ensemble members produced with the chosen configuration as the best using an empirical orthogonal function (EOF) perturbation methodology. The chosen configuration presented high correlation and low root-mean-squared error (RMSE) for precipitation and T2M anomaly predictions at the first week and these indices degraded as lead time increased, maintaining moderate performance up to week-4 over the tropical Pacific and northern South America. For MJO predictions, this configuration crossed the 0.5 bivariate correlation threshold in 18 days. The ensemble approaches improved the correlation and RMSE of precipitation and T2M anomalies. ICEN improved precipitation and T2M predictions performance over eastern South America at week-3 and over northern South America at week-4. Improvements were also noticed for MJO predictions. The time to cross the above-mentioned threshold increased to 21 days for MCEN and to 20 days for ICEN. K E Y W O R D Sforecast verification, intraseasonal variability, MJO Q J R Meteorol Soc. 2020;146:1067-1084.wileyonlinelibrary.com/journal/qj

Evaluation of climate simulations produced with the Brazilian global atmospheric model version 1.2

“…For short-wave (SW) and long-wave (LW) radiation, the Rapid Radiative Transfer Model for General Circulation Models (RRTMG, Iacono et al 2008) scheme, the long-wave radiation scheme (Chou et al 2001, CLIRAD-LW), and the short-wave radiation scheme developed by Chou and Suarez (1999) (CLIRAD-SW), the latter modified by Tarasova and Fomin (2000), were tested. After performing these tests, the chosen model configuration to be used in this study is similar to the configuration described in Guimarães et al (2020). The BAM-1.2 physical processes components used for performing the simulations here evaluated are indicated in Table 1: microphysics from Morrison et al (2005Morrison et al ( , 2009; the IBIS-CPTEC surface model (Kubota, 2012); the long-wave radiation scheme developed by Chou et al (2001) (CLIRAD-LW); the short-wave radiation scheme developed by Chou and Suarez (1999) (CLIRAD-SW), modified by Tarasova and Fomin (2000); the Bretherton-Park moist diffusion scheme (Bretherton and Park, 2009) for the planetary boundary layer (PBL), which is referred to as moist-PBL; and the revised version of the simplified Arakawa-Shubert deep convection scheme (Han and Pan.…”

“…Cavalcanti and Raia (2017) and Cavalcanti et al (2020) investigated the ability of a predecessor BAM version with simplified and fast physical parameterizations (known as BAM version 0.0, BAM-0.0) in simulating the lifecycle of the South American monsoon system and climate variability over South America, respectively. Guimarães et al (2020) defined a configuration and performed the first assessment of BAM-1.2 for sub-seasonal predictions, which is the same version currently used at CPTEC for global operational numerical weather prediction. However, the performance of BAM-1.2 climate simulations is yet to be documented.…”

Subseasonal prediction performance for South American land–atmosphere coupling in extended austral summer

“…Skill for ET (Figure 5g-l) is worse than the climatological reference forecast for all models except ECMWF from Week 1 over the six regions, except for NCEP for NDE compared against ERA5L (Figure 5g-l). Forecast skill for T2 (Figure 5m-r) in all models is generally better than the climatological reference forecast over all six regions for Weeks 1 and 2, with ECMWF showing the highest and CPTEC showing the lowest skill, likely due to underdispersion of the CPTEC ensemble (Guimarães et al, 2020). The models have better skill than the climatological reference forecast over NSA even at Week 5, perhaps due to lower T2 variability in the tropics.…”

“…We evaluate hindcasts (reforecasts) of three models from the S2S project database (Vitart et al, 2017) and from a new Brazilian model (Guimarães et al, 2020) over the common period of 1999-2010. The four hindcast datasets and associated modelling systems are described below and in with 11 ensemble members (Guimarães et al, 2020). CPTEC performs a "fixed" (frozen) hindcast set.…”

“…CPTEC performs a "fixed" (frozen) hindcast set. The initialization dates can be found in Table 1 of Guimarães et al (2020).…”