https://escholarship.org/uc/item/65m9271b
Authors: Xing, Chen
2025
Abstract
Human activities reshape the climate system by altering both the mean state and its variability. The Pacific—home to the Pacific Decadal Variability (PDV), and the El Niño–Southern Oscillation (ENSO)—mediates global hydroclimate and regional temperature–precipitation patterns, yet the forced component of these modes remains contested. This dissertation uses large ensembles, including single-forcing experiments, to cleanly separate externally forced responses from internal variability and to clarify mechanisms relevant for projection.First, I evaluate PDV diagnostics and evaluate PDV response to anthropogenic forcing. Comparing three common methods, I show that the widely used approach that removes only the global mean sea surface temperature (SST) aliases evolving mean state trends into PDV, falsely implying a strong forced response. When mean state change is represented by the ensemble mean—capturing its spatial structure and temporal evolution—there is no significant forced change in PDV modes.Second, I assess solar radiation management (SRM) influence on ENSO. In idealized experiments, subtropical Pacific marine cloud brightening (MCB) reduces ENSO variance by roughly two-thirds, whereas stratospheric aerosol injection has minimal effect. A mixed layer heat budget analysis links the ENSO suppression to a La Niña–like mean state: cloud induced shortwave reductions cool the southeastern Pacific, strengthen easterlies, steepen the zonal thermocline tilt, and weaken the Bjerknes feedback. Given ENSO’s central role in global hydroclimate and ecosystems, such suppression carries substantial risks.Third, I quantify changes in global marine heatwaves (MHWs) intensity. By late century, increases in MHW intensity are dominated by mean SST warming, with internal variability playing a minor role. Motivated by recent “Blob” events, I then examine the Northeastern Pacific SST trend. Single-forcing ensembles indicate aerosol forcing dominated the SST trend before ~1980, greenhouse gases thereafter, and that the recent (2009–2025) abrupt warming is primarily an expression of internal variability rather than a direct consequence of recent Chinese aerosol reductions. Together, this work implies that accurate representation of the mean state is critical for credible assessments of climate variability and for robust model–observation evaluation.
Source: eScholarship