Mechanisms and Predictability of the Global Climate Impacts of Atlantic Multidecadal Variability

Atlantic Multi-decadal Variability (AMV), also known as the Atlantic Multi-decadal Oscillation (AMO), is characterized by a sharp rise and fall of the North Atlantic basin-wide sea surface temperatures (SST) on multidecadal time scales. During the instrumental record, AMV is characterized by a warming in the 1920-30s, a cooling in the 1960-70s and a return of the warming in the mid-1990s. Widespread consequences of these rapid temperature swings are noted by many previous studies, such as the record warming of Greenland in the 1920-30s, the drying of Sahel in the 1960-70s, the increase in drought frequency or decrease in precipitation over North America during warm phase of AMV, and change in the frequency and intensity of Atlantic hurricanes on multi-decadal time scales. Predictability studies suggest that as an oceanic phenomenon (i.e., changes in circulation and ocean thermal structure) the AMV has some potential predictability. Given that, it is important to understand the mechanisms that link AMV worldwide climate impacts, by season and location, and to quantify the influence of this phenomenon relative to that of other mechanisms such as anthropogenic influences, ENSO, and the underlying background of chaotic, and presumably unpredictable, climate variability. The proposed study will determine the worldwide climate impacts of AMV, with a particular focus on North American impacts, and the possible prediction of these based on the premise that AMV is predictable at least for a time interval of several years. To achieve this goal, we will build on analyzing instrumental and high-resolution proxy data such as available tree ring reconstruction of temperature and precipitation to build a data based statistical background. We will also use outputs from the IPCC AR4 coupled ocean-atmosphere model simulations to evaluate coupled model’s AMV-climate connections. Given the short records available from observations and many of the IPCC AR4 integrations, one cannot robustly separate signal (i.e., AMV impact) from noise or other influences. Therefore, the observed and modeled estimate of AMV impact will be compared with outputs from multiple member ensembles of model integrations forced with the AMV SST and analyzed with various statistical methods designed to detect signal from noise. We will focus on North American impacts of the AMV in variables such as surface air temperature, precipitation, and evaporation. Specific attention will also be given to the impact of AMV on extreme events such as tropical cyclones and extratropical storms, as well as droughts and heat waves. The underlying idea is to map out the regional and seasonal impacts of AMV relative to other climate variations on seasonal-to-interannual-todecadal time scales in a way that provides climate forecasters and decision makers with useful information on the impact of this phenomenon.