Pronounced decadal-scale variations are superimposed on the long-term, anthropogenic warming of climate. Over the past century, the global temperature evolved from an early twentieth century warming to a “grand hiatus” period around 1950-1980, followed by an accelerated warming and another hiatus period starting around the late 1990s. Long-term warming and decadal-scale variability are also observed in regional temperature records, as e.g. the Arctic. The strong decline in Arctic sea ice has been suggested to modulate the frequency and intensity of extreme weather events in mid-latitudes1,2,3. The decadal-scale variability in the Tropics drives hemispheric or even global climate variations, as suggested during the recent global warming hiatus4,5. Yet a detailed understanding of these complex linkages at decadal time scales does not exist6.
The proposal’s main goal is to investigate the inter-linkages among seasonal-to-decadal-scale variability in Arctic, mid-latitudes and Tropics, their mechanisms and their implications for reliable climate predictions. We aim at gaining a deeper understanding of the mechanisms leading to seasonal to decadal-scale climate variability at regional scale, focusing on atmospheric and oceanic teleconnections between distant regions that potentially affect extreme climate conditions. We further aim to improve prediction skill on seasonal to decadal time scales and for specific targets such as the East Asian, Indian and African Monsoon systems, Eurasian climate extremes and hiatus like events.
Our proposed research is divided into three inter-related components.
1. We will investigate inter-regional and inter-basin linkages on sub-seasonal to interannual time scales, achieved mainly through atmospheric teleconnections. At present, the role of air-sea interactions among different latitude belts in the Atlantic and Pacific, and of stratosphere - troposphere coupling in these linkages has not been clarified. We will explore those linkages in existing observational and reanalysis data sets and through sets of coordinated multi-model experiments with state-of-the-art atmospheric general circulation models (AGCMs), where possible remote climate impacts of daily distributions of observed SST and sea ice (1980-2015) in certain key regions (e.g. Arctic, North Atlantic, North Pacific, Tropics) will be assessed. We will further perform experiments with suppressed stratospheric pathways. We will also examine the interaction between the Madden Julian Oscillation and high-latitude weather patterns.
2. We will examine Inter-regional linkages on decadal-scale time scales in existing CMIP5 historical simulations and include oceanic teleconnections through sets of coordinated multi-model partially coupled climate model and pacemaker7 experiments, covering the entire 20th century. Partial coupled model simulations, which are forced by observed wind stress or SST anomalies, maintain a more consistent thermodynamic coupling compared to SST-driven AGCM experiments. To investigate the resolution-dependence of the identified linkages, we will make use of high-resolution climate model (~25km resolution) simulations (1950-2050) to be performed within the Horizon 2020 PRIMAVERA project.
3. Based on the results of part 1 and 2, we will assess the predictability of societally relevant climate variations and extreme events. Examples are the recent period with fast Arctic sea ice loss, periods with enhanced/reduced African and Asian Monsoon activity or extreme Eurasian summer and winter temperatures. We will explore the limits of predictability related to the identified processes and linkages by analysing existing multi-model repositories of seasonal-to-decadal hindcast simulations (CMIP5/CMIP6, SPECS, MiKlip, S2S). Additional pacemaker type prediction experiments will be performed to further explore the potential to enhance predictability in certain cases such as sharp climatic shifts or transitions.