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

Abstract

This paper presents an evaluation of climate simulations produced by the Brazilian Global AtmosphericModel version 1.2 (BAM-1.2) of the Center for Weather Forecast and Climate Studies (CPTEC). The model was run over the 1975-2017 period at two spatial resolutions, corresponding to ~180 and ~100 km, both with 42 vertical levels, following most of the Atmospheric Model Intercomparison Project (AMIP) protocol. In this protocol, observed sea surface temperatures (SSTs) are used as boundary conditions for the atmospheric model. Four ensemble members were run for each of the two resolutions. A series of diagnostics was computed for assessing the model´s ability to represent the top of the atmosphere (TOA) radiation, atmospheric temperature, circulation and precipitation climatological features. The representation of precipitation interannual variability, El Niño-Southern Oscillation (ENSO) precipitation teleconnections, the Madden and Julian Oscillation (MJO) and daily precipitation characteristics was also assessed. The model at both resolutions reproduced many observed temperature, atmospheric circulation and precipitation climatological features, despite several identified biases. The model atmosphere was found to be more transparent than the observations, leading to misrepresentation of cloud-radiation interactions. The net cloud radiative forcing, which produces a cooling effect on the global mean climate at the TOA, was well represented by the model. This was found to be due to the compensation between both weaker longwave cloud radiative forcing (LWCRF) and shortwave cloud radiative forcing (SWCRF) in the model compared to the observations. The model capability to represent inter-annual precipitation variability at both resolutions was found to be linked to the adequate representation of ENSO teleconnections. However, the model produced weaker than observed convective activity associated with the MJO. Light daily precipitation over the southeast of South America and other climatologically similar regions was diagnosed to be overestimated, and heavy daily precipitation underestimated by the model.Increasing spatial resolution helped to slightly reduce some of the diagnosed biases. The performed evaluation identified model aspects that need to be improved. These include the representation of polar continental surface and sea ice albedo, stratospheric ozone, low marine clouds, and daily precipitation features, which were found to be larger and last longer than the observed features.

An Assessment of Land–Atmosphere Interactions over South America Using Satellites, Reanalysis, and Two Global Climate Models

“…We conducted a detailed assessment of land-atmosphere interactions over South America using satellite products, a novel reanalysis (ERA5-Land) and simulations from two global climate models, HadGEM3 and BAM-1.2. The climatology and variability of these models have previously been evaluated at the global scale (Figueroa et al 2016;Williams et al 2018;Kuhlbrodt et al 2018;Coelho et al 2021), and landatmosphere interactions over South America were analyzed in similar versions of these models at subseasonal time scales (Chevuturi et al 2021, manuscript submitted to Climate Resilience Sustainability). Here, we focused on assessing landatmosphere interactions over longer, climate-relevant time scales, with the aim of comparing results from different FIG.…”

“…In addition, we obtained monthly AMIP-type runs from BAM-1.2 (Coelho et al 2021) (Figueroa et al 2016;Guimarães et al 2020). In this study, BAM-1.2 was run with triangular 126wave truncation (TQ126), corresponding to a horizontal grid of approximately 1.08 3 1.08 at the equator, 42 sigma vertical levels (32 levels in the troposphere and 10 in the stratosphere), and the model top was at 2 hPa.…”

“…BAM-1.2 uses the land surface scheme IBIS-CPTEC (Kubota 2012). A full description of BAM-1.2 and its physical components is provided by Coelho et al (2021). A summary of the characteristics of ERA5-Land, HadGEM3, and BAM-1.2 is presented in Table S2.…”

“…There is a clear need to evaluate the representation of landatmosphere interactions in climate models in order to determine where model improvements are needed, and ultimately to reduce the uncertainty in future climate projections. In this study, we selected two global climate models representing different stages of model development for a detailed assessment of land-atmosphere interactions over South America: the U.K. Hadley Centre Global Environment Model version 3 (HadGEM3), which has been developed over several decades and has been included in multiple CMIP phases (Murphy 1995;Johns et al 1997;Pope et al 2000;Collins et al 2001;Johns et al 2006;Martin et al 2011;Williams et al 2018;Andrews et al 2020), and the Brazilian Global Atmospheric Model version 1.2 (BAM-1.2), which was developed comparatively recently (Figueroa et al 2016;Coelho et al 2021). Furthermore, we wanted to understand how representation of land-atmosphere interactions in these two models compared with wider ensemble of CMIP6 models, to assess the generalizability of our findings.…”

Assessing the representation of South American monsoon features in Brazil and U.K. climate model simulations

“…This study aims to assess how well the CPTEC/INPE Brazilian Global Atmospheric Model version 1.2 (BAM-1.2; Coelho et al, 2021) and the atmospheric component of the UK Met Office Hadley Centre Global Environment Model version 3 (HadGEM3-GC3.1; Ridley et al, 2018;Williams et al, 2018) represent the main features of the South Amer-ican Monsoon system in Atmospheric Model Intercomparison Project (AMIP)-type climate simulations. The UK Met Office model simulations will hereafter be referred to as HadGEM3 for brevity.…”