Title: Two distinct modes of North American spring temperature evolution shaped by tropical SST regimes and midlatitude circulation
Abstract:
Abnormal temperature fluctuations during spring can substantially impact vegetation growth, agricultural productivity, and ecosystem stability. However, the drivers of year-to-year variations in the evolution of North American spring temperature anomalies remain poorly understood. Here, we apply Season-reliant Empirical Orthogonal Function (S-EOF) analysis to identify the leading modes of spring temperature evolution over North America. We focus on the first two modes here, which differ from conventional responses to winter El Niño–Southern Oscillation (ENSO) and are instead linked to distinct evolutions of tropical sea surface temperature (SST). The first mode (S-EOF1) features rapidly decaying tropical Pacific SST anomalies from winter to spring and the emergence of tropical North Atlantic (TNA) SST anomalies, leading to a persistent north–south temperature dipole over North America. This mode is jointly influenced by the North Pacific Oscillation (NPO), largely independent of winter ENSO, and TNA-driven North Atlantic circulation anomalies. In contrast, the second mode (S-EOF2) is associated with more persistent tropical Pacific SST anomalies and the absence of TNA SST anomalies, resulting in evolving temperature patterns with temporal phase reversals across large parts of North America. This mode is primarily driven by the North Pacific (NP) pattern, which is further amplified by persistent tropical Pacific SST anomalies. Dominance analysis further quantifies the relative contributions of tropical SST and midlatitude circulation patterns. We also briefly assess the seasonal forecast skill for these two modes and the representation of their associated physical processes in seasonal forecast models. These findings highlight the critical roles of ENSO decay rate, TNA SST variability, and their coupling with midlatitude circulation in shaping distinct spring temperature evolution patterns, with important implications for improving seasonal prediction.
