Linkages have been established between the Atlantic Multidecadal Oscillation (AMO) and surface air temperature variations, low-level jet streams, and precipitation trends in both northeastern Brazil and southeastern South America. Previous studies have discerned distinct wet-season (March-May) precipitation responses in northeastern Brazil, with cold (warm) AMO phases triggering increased (decreased) precipitation. Findings from various records indicate that the AMO's variability extends for thousands of years. A recent reconstruction suggests a significant AMO role in the shift from the Medieval Climate Anomaly (MCA) to the Little Ice Age (LIA) and reveals the LIA as the longest period with a persistent cold anomaly in the North Atlantic over the past similar to 3 millennia. Stable oxygen isotope records from South America show typical AMO periodicities (similar to 65 years), however, despite increased paleo precipitation data, the AMO's role in South American precipitation during the LIA and MCA remains unclear. In this study, the influence of AMO phases on atmospheric dynamics, precipitation patterns, and stable oxygen isotope composition (delta 18O) of precipitation over South America is assessed using the water isotope-enabled version of the Community Earth System Model version 1.2 (iCESM1.2). This research sheds light on the connection between AMO-induced precipitation anomalies and isotopic signals observed in paleoclimate records and emphasizes the significance of isotope-enabled climate models in unraveling the mechanisms behind past variations. The analysis involves comparing delta 18O simulations with published reconstructions from South America. By utilizing climate models that incorporate isotopes, we can delve deeper into understanding the impact of the AMO on precipitation patterns and isotopic ratios. Contrary to expectations, the simulated delta O-18(p) signal differ from speleothem records over the western Amazon and Andes during the LIA. That is, the simulated significant total precipitation amounts change over the western Amazon and Andes are not reflected in delta O-18(p) depletion.

The impact of hydrological and geological disasters has resulted in significant social, economic, and human losses, which added climate change impacts, and such events have become more frequent and intense. Therefore, our objective is to analyze the extreme rainfall (trends) in the Metropolitan Region of the Paraiba do Sul Valley and North Coast of Sao Paulo (RMVPLN). This analysis will support the most affected areas by landslides identification, which mainly impact roads and their population. In addition, evaluate the atmosphere conditions that supported these extreme rainfall events. To achieve our objectives, we have surveyed historical landslide data reported by the Brazilian government and information related by press and media. The precipitation evaluation used CHIRPS v.2 data and ETCCDI indices and the vertically integrated moisture flow and wind speed were calculated by ERA5 reanalysis. Our results show that the frequency and intensity of rainfall indicators such as seasonal PRCPTOT, R20mm, R30mm, and SDII have increased, particularly in the coastal and mountainous regions of São Paulo. This is due to positive anomalies of moisture transport and an increase of ocean winds influenced by the intense South Atlantic Subtropical Anticyclone (SASA). The region with the highest susceptibility to landslides triggered by extreme rainfall is the one that combines deforested areas, high slope topography, and excessive anthropic intervention. The presence of mountainous regions increases the risk of landslides, which can damage local infrastructure and expose the vulnerability of populations in these risk areas.

Climate science has long explored whether higher resolution regional climate models (RCMs) provide improved simulation of regional climates over global climate models (GCMs). The advent of convective-permitting RCMs (CPRCMs), where sufficiently fine-scale grids allow explicitly resolving rather than parametrising convection, has created a clear distinction between RCM and GCM formulations. This study investigates the simulation of tropical-extratropical (TE) cloud bands in a suite of pan-South America convective-permitting Met Office Unified Model (UM) and Weather Research and Forecasting (WRF) climate simulations. All simulations produce annual cycles in TE cloud band frequency within 10-30% of observed climatology. However, too few cloud band days are simulated during the early summer (Nov-Dec) and too many during the core summer (Jan-Feb). Compared with their parent forcing, CPRCMs simulate more dry days but systematically higher daily rainfall rates, keeping the total rain biases low. During cloud band systems, the CPRCMs correctly reproduced the observed changes in tropical rain rates and their importance to climatology. Circulation analysis suggests that simulated lower subtropical rain rates during cloud bands systems, in contrast to the higher rates in the tropics, are associated with weaker northwesterly moisture flux from the Amazon towards southeast South America, more evident in the CPRCMs. Taken together, the results suggest that CPRCMs tend to be more effective at producing heavy daily rainfall rates than parametrised simulations for a given level of near-surface moist energy. The extent to which this improves or degrades biases present in the parent simulations is strongly region-dependent.

The increase in greenhouse gasses (GHG) anthropogenic emissions and deforestation over the last decades have led to many chemical and physical changes in the climate system, affecting the atmosphere's energy and water balance. A process that could be affected is the Amazonian moisture transport in the South American continent (including La Plata basin), which is crucial to the southeast Brazilian water regime. The focus of our research is on evaluating how local (i.e. Amazon deforestation) and global forcings (increase of atmospheric GHG concentration) may modify this moisture transport under climate change scenarios. We used two coupled landatmosphere models forced by CMIP6 sea surface temperatures to simulate these processes for two scenarios: i) increase in carbon dioxide (CO2) - RCP8.5 atmospheric levels (00DEF), and ii) total Amazon deforestation simultaneous with atmospheric CO2 levels increased (100DEF). These scenarios were compared with a control simulation, set with a constant CO2 of 388 ppm and present-day Amazon Forest cover. The 30 -year Specific Warming Level 2 (SWL2) index evaluated from the simulations is set to be reached 2 years earlier due to Amazon deforestation. A reduction in precipitation was observed in the Amazon basin (-3.1 mm & sdot;day- 1) as well as in La Plata Basin (-0.5 mm & sdot;day-1) due to reductions in the Amazon evapotranspiration (-0.9 mm & sdot;day-1) through a stomatal conductance decrease (00DEF) and land cover change (100DEF). In addition, the income moisture transport decreased (22 %) in the northern La Plata basin in both scenarios and model experiments. Our results indicated a worse scenario than previously found in the region. Both Amazon and La Plata hydrological regimes are connected (moisture and energy transport), indicating that a large-scale Amazon deforestation will have additional climate, economic and social implications for South America.

Global air temperature increase has caused changes in the global climate such as droughts, floods and severe events. Land cover shifts and the increase in greenhouse gas emissions can intensify these impacts. Therefore, the application of Specific Warming Level 2 (SWL2) technique has been considered to evaluate the global warming issue in various climate scales, which may induce different responses in the climate system. The objective of this study is to evaluate the impacts of a doubling of current atmospheric CO2 concentrations as well as the influence of deforestation of the Amazon Forest on the climate of South America. In this study, the CPTEC-BAM1.2 global model was used to simulate these processes. The first scenario of simulations considered doubled atmospheric CO2 concentration (2XCO(2) = 776 ppm) and the second scenario is the total conversion of the Amazon Forest to pasture (DEF). In the third scenario both CO2 and deforestation are considered simultaneously (2XCO(2) + DEF). A control simulation considers the natural Amazon Forest shape and a steady CO2 concentration of 388 ppm. Results suggest that the temperature increased in all scenarios, the highest increment in the 2XCO(2) + DEF experiment (5.4 & DEG;C in 2XCO(2) + DEF, 5.1 & DEG;C in DEF and 3.2 & DEG;C in 2XCO(2)). The model simulated drastic negative rainfall anomalies in each of three scenarios (2XCO(2) + DEF: -2.1 mm day(-1), 2XCO(2): -1.3 mm day(-1) and DEF: -0.9 mm day(-1)). Besides less precipitation and evapotranspiration, shifts in moisture transport were identified in each of the three scenarios. The effect of doubled atmospheric CO2 produces results similar to deforestation for hydrological water budget changes, and total pasture conversion may intensify such changes.