Search for: All records

Abstract Increases in atmospheric greenhouse gases will not only raise Earth’s temperature but may also change its variability and seasonal cycle. Here CMIP5 model data are analyzed to quantify these changes in surface air temperature (Tas) and investigate the underlying processes. The models capture well the mean Tas seasonal cycle and variability and their changes in reanalysis, which shows decreasing Tas seasonal amplitudes and variability over the Arctic and Southern Ocean from 1979 to 2017. Daily Tas variability and seasonal amplitude are projected to decrease in the twenty-first century at high latitudes (except for boreal summer when Tas variability increases) but increase at low latitudes. The day of the maximum or minimum Tas shows large delays over high-latitude oceans, while it changes little at low latitudes. These Tas changes at high latitudes are linked to the polar amplification of warming and sea ice loss, which cause larger warming in winter than summer due to extra heating from the ocean during the cold season. Reduced sea ice cover also decreases its ability to cause Tas variations, contributing to the decreased Tas variability at high latitudes. Over low–midlatitude oceans, larger increases in surface evaporation in winter than summer (due to strong winter winds, strengthened winter winds in the Southern Hemisphere, and increased winter surface humidity gradients over the Northern Hemisphere low latitudes), coupled with strong ocean mixing in winter, lead to smaller surface warming in winter than summer and thus increased seasonal amplitudes there. These changes result in narrower (wider) Tas distributions over the high (low) latitudes, which may have important implications for other related fields. 

Remote influences of ENSO are known to vary with different phases of the interdecadal Pacific oscillation (IPO). Here, observational and reanalysis data from 1920 to 2014 are analyzed to present a global synthesis of the IPO’s modulation on ENSO teleconnections, followed by a modeling investigation. Regressions of surface air temperature T, precipitation P, and atmospheric circulations upon IPO and ENSO indices reveal substantial differences between ENSO and IPO teleconnections to regional T and P in terms of spatial pattern, magnitude, and seasonality. The IPO’s modulation on ENSO teleconnections asymmetrically varies with both IPO and ENSO phases. For a given ENSO phase, IPO’s modulations are not symmetric between its two phases; for a given IPO SST anomaly, its influence depends on whether it is superimposed on El Niño, La Niña, or neutral ENSO. The IPO modulations are linked to the atmospheric response to tropical SST anomalies, manifested in the local Hadley circulation and the local Walker circulation at low latitudes and the Rossby wave train in the extratropics, including the Pacific–North American (PNA) pattern in the Northern Hemisphere. A set of numerical experiments using CAM5 forced with different combinations of the IPO- and ENSO-related SSTs further shows that the asymmetric modulation arises from the nonlinear Clausius–Clapeyron relation, so that the atmospheric circulation response to the same IPO-induced SST departure is larger during a warm rather than a cold ENSO phase, and the response to a warm IPO state is larger than that to a cold IPO state. The asymmetry depends primarily on the tropical Pacific mean state and tropical SST anomalies and secondarily on extratropical SST anomalies.