Enhanced 20th-century heat transfer to the Arctic simulated in the context of climate variations over the last millennium

Abstract

Abstract. Oceanic heat transport variations, carried by the northward-flowing Atlantic Water, strongly influence Arctic sea-ice distribution, ocean-atmosphere exchanges, and pan-Arctic temperatures. Palaeoceanographic reconstructions from marine sediments near Fram Strait have documented a dramatic increase in Atlantic Water temperatures over the 20th century, unprecedented in the last millennium. Here we present results from Earth system model simulations that reproduce and explain the reconstructed exceptional Atlantic Water warming in Fram Strait in the 20th century in the context of natural variability during the last millennium. The associated increase in ocean heat transfer to the Arctic can be traced back to changes in the ocean circulation in the subpolar North Atlantic. An interplay between a weakening overturning circulation and a strengthening subpolar gyre as a consequence of 20th-century global warming is identified as the driving mechanism for the pronounced warming along the Atlantic Water path toward the Arctic. Simulations covering the late Holocene provide a reference frame that allows us to conclude that the changes during the last century are unprecedented in the last 1150 years and that they cannot be explained by internal variability or natural forcing alone.

Pseudo-proxy evaluation of climate field reconstruction methods of North Atlantic climate based on an annually resolved marine proxy network

“…The ensemble simulations r1 and r2 are started from the same initial conditions, but they consider a slightly different parameter setting regarding the volcanic aerosols. The experiments r2 and r3 consider the same parameter setting, but they differ in the initial conditions of the oceanic state (Jungclaus et al, 2014).…”

The importance of external climate forcing for the variability and trends of coastal upwelling in past and future climate

“…Only the volcanic scheme in the past1000-r1 differs from the ones in the past1000-r2 and past1000-r3 experiments in the prescribed distribution of the volcanic aerosol size. This distribution is assumed to be lognormal, with a standard deviation of 1.2 µm in r1 and 1.8 µm in r2 and r3 (Jungclaus et al, 2014). The forcings prescribed in the past1000 (Masson-Delmotte et al, 2013) follow a standard protocol defined by the project CMIP5 and comprise orbital forcing (Berger, 1978), variability in solar irradiance (Vieira and Solanki, 2010), volcanic activity (Crowley and Unterman, 2013), greenhouse gas concentrations (Flückinger et al, 2002;Hansen and Sato, 2004;MacFarling Meure et al, 2006), and land-use changes (Pongratz et al, 2009).…”

The PMIP4 contribution to CMIP6 – Part 3: The last millennium, scientific objective, and experimental design for the PMIP4 <i>past1000</i> simulations

“…Integrated assessment of reconstructions and simulations has led to progress in model evaluation and process understanding (e.g. Lehner et al, 2013;Sicre et al, 2013;Jungclaus et al, 2014;Man et al, 2012;Man and Zhou, 2014). The increasing number of available simulations and reconstructions has also created a need for development of new statistical modelling approaches dedicated to model-data comparison analysis (e.g.…”