The role of historical forcings in simulating the observed Atlantic multidecadal oscillation

Abstract

We analyze the Atlantic multidecadal oscillation (AMO) in the preindustrial (PI) and historical (HIST) simulations from the Coupled Model Intercomparison Project Phase 5 (CMIP5) to assess the drivers of the observed AMO from 1865 to 2005. We draw 141 year samples from the 41 CMIP5 model's PI runs and compare the correlation and variance between the observed AMO and the simulated PI and HIST AMO. The correlation coefficients in 38 forced (HIST) models are above the 90% confidence level and explain up to 56% of the observed variance. The probability that any of the unforced (PI) models do as well is less than 3% in 31 models. Multidecadal variability is larger in 39 CMIP5 HIST simulations and in all HIST members of the Community Earth System Model Large Ensemble than their corresponding PI. We conclude that there is an essential role for external forcing in driving the observed AMO.

Pronounced differences between observed and CMIP5‐simulated multidecadal climate variability in the twentieth century

“…On one hand, if our derived CMIP5‐based forced signals are realistic, these differences must arise from internal climate system dynamics presumably misrepresented in CMIP5 models, such as sea ice dynamics [ Wyatt and Curry , ], oceanic mesoscale eddies [ Siqueira and Kirtman , ], positive cloud and dust feedbacks [ Evan et al , ; Martin et al , ; Brown et al , ; Yuan et al , ], or SST‐forced NAO response [ Kushnir et al , ; Eade et al , ; Stockdale et al , ; Siegert et al , ]. On the other hand, however, it is possible that CMIP5 models underestimate multidecadal variations in the true response of the climate system to external forcing or misrepresent the forcing itself [ Booth et al , ; Murphy et al , ]; if this is true, the model‐data differences reflect the mismatch between the actual and CMIP5‐simulated forced signals, whereas the real world's internal climate variability may be consistent with that simulated by the models. In either case, we strongly believe that model development activities should strive to alleviate the present large discrepancies between the observed and simulated multidecadal climate variability, as these discrepancies hinder our fundamental understanding of the observed climate change.…”

Insights into Atlantic multidecadal variability using the Last Millennium Reanalysis framework

“…While there has been much hypothesized regarding these oscillations in the instrumental record and whether they are a result of internal climate variability or a result of external forcing (see, e.g., Booth et al, 2012;Zhang et al, 2013;Murphy et al, 2017;Bellucci et al, 2017, and others), we point out two important characteristics of the instrumental period that render it a poor era for studying multidecadal variability in the climate system. First, the instrumental record is very short; as a result, there is very little that can be said about multidecadal timescale variability over this time period that carries any statistical weight (see, e.g., Wunsch, 1999;Vincze and Janosi, 2011).…”

“…This argument, however, has been challenged by (Zhang et al, 2013), who suggest that the GCM used in (Booth et al, 2012) incorrectly modeled aerosol effects and show that improved representation of these effects reveals that 20th century surface temperature variations cannot be explained in full by changes in anthropogenic aerosol emissions. Nevertheless, more recent work by (Murphy et al, 2017) and (Bellucci et al, 2017) further challenges the premise that North Atlantic SST variability over the industrial period is unaffected by anthropogenic forcings.…”

Global Mean Surface Temperature Response to Large‐Scale Patterns of Variability in Observations and CMIP5

“…To test whether the process of removing the forced response also removes a component of AMV, the piControl experiments were analyzed, after accounting for model drift (Text S1; Sen Gupta et al, 2013). This result conflicts with Murphy et al (2017), who argue that historical forcings have enhanced AMV. This result conflicts with Murphy et al (2017), who argue that historical forcings have enhanced AMV.…”