Synthesis and dissemination of ocean and atmosphere heat transport to the Arctic (D2.4)

Summary

Meridional Energy Transport (MET), both in the atmosphere (AMET) and ocean (OMET), has significant impact on the climate in the Arctic. In this study, the quantification of atmospheric meridional energy transport (AMET) and oceanic meridional energy transport (OMET) at subpolar latitudes have been performed using six state-of-the-art reanalyses datasets (ERA-Interim, MERRA2, JRA55, ORAS4, GLORYS2V3, and SODA3). Emphasis is placed on the key processes regulating AMET and OMET from midlatitudes to the Arctic. The differences between these data sets were investigated. A forced NEMO-ORCA hindcast, two high resolution fully coupled HadGEM3-GC3.1 simulations and observations in the Atlantic from Rapid Climate Change-Meridional Overturning Circulation and Heatflux Array (RAPID ARRAY) and Overturning in the Subpolar North Atlantic Program (OSNAP) are included in the comparison. Based on the intercomparison of reanalyses data, model outputs and the observation data, sources of uncertainty are identified. The impacts of orography on the atmospheric moisture and heat transport toward the pole were studied with the IPSL-CM6 model experiments. Compensation and feedback between oceanic and atmospheric heat, moisture or energy transport impacts on the Arctic variability were checked with the CESM1 Large Ensemble simulations and the MPI-ESM-LR grand ensemble simulations (MPI-GE), which also reflect the respective role of the natural climate variability and externally forced climate change. To support our comparison of AMET and provide more insight, we further investigate AMET with multiple atmospheric model simulations (EC-Earth, HadGEM, NorSEM, WACCM6, CMCC-CM, IPSL-CM, IAP-AGCM, MPIESM) from the coordinated experiments, in collaboration with Blue-Action WP3 “Linkages of Arctic climate changes to lower latitudes”.

The main results are: 

• The mean transport in all chosen atmospheric (ERA-Interim, MERRA2 and JRA55) and oceanic (ORAS4, GLORYS2V3 and SODA3) reanalyses data sets agree well, while the spatial distribution and temporal variations of AMET and OMET differ substantially among the reanalyses data sets.
• For the ocean, comparisons with observed heat transports at subtropical and subpolar Atlantic confirm that the OMET estimated from reanalyses is consistent with observations.
• The existence of sources and sinks in reanalyses data sets introduces large uncertainties in the computation of energy transport. Based on our results, it seems that AMET and OMET cannot be constrained by the available observations.
• The AMET and OMET estimated from reanalyses at large time scales should be used with great care, especially when studying variability and interactions between the Arctic and midlatitudes beyond interannual time scales.
• Strong compensation between poleward AMET and OMET was found within a fully coupled model simulation (CESM1). It also shows that OMET has significant impact on the Arctic sea ice variability and the Arctic Amplification (AA). With a grand ensemble simulation (MPI-GE), a crucial role of the significant changes in the North Atlantic sub-polar gyre strength for the increasing high latitude heat transport to the Arctic and freshening of the North Atlantic under global warming is identified.
• Generally, the chosen high resolution ocean model (NEMO ORCA) and fully coupled model (HadGEM3-GC3.1) show very good agreement with both the reanalyses and observations on OMET. There are biases between the calculated AMET from coordinated experiments and reanalyses but the variabilities are similar. With IPSL-CM simulation, strong impact of orography on the poleward AMET is identified.