Mechanisms and predictability of North Atlantic Decadal Climate variability
Final Report Abstract
The North Atlantic is now recognized as a key region for decadal prediction. Currently it is the only region where ocean circulation leads to skillful predictions a decade in advance. However, there remain many challenges to realizing the full predictability of climate in this region. In this regard, the project contributed key results to our knowledge of ocean-atmosphere interaction, the mechanisms for decadal variability, and the impact of model error. Methods were also developed to extend the period over which predictability is assessed, so as to achieve better estimates. The predictability in the North Atlantic found in current systems is generally restricted to regions of the subpolar gyre, and does not extend significantly to the atmosphere or continental areas. Our research suggests that in reality the ocean in this region does influence the large-scale atmospheric circulation, opening the possibility of skillful climate prediction in the North Atlantic sector. Firstly, historical observations were used to reconstruct heat exchange between the North Atlantic Ocean and atmosphere. With these data we provided the first observational proof of the famous Bjerknes Conjecture, which states that on year-to-year timescales chaotic atmospheric variability drives most sea surface temperature anomalies, while on the timescales of a decade, ocean dynamics drive the sea surface temperature anomalies in the Gulf Stream and its extension. Thus, on these longer timescales the ocean may drive the atmosphere. Given the chaotic nature of mid-latitude atmosphere and limited data, it is difficult to confirm the ocean’s influences on the atmosphere. Thus, in a second step we employed numerical models to show that observed shifts in the winter atmospheric circulation in the North Atlantic sector were partly driven by decadal shifts in the underlying ocean temperature. Proper representation of the dynamical interaction between the stratosphere and troposphere was shown to be essential, and explaining to some degree the weaker results of other research groups. A poor understanding of the mechanisms for North Atlantic decadal variability is another factor that limits the prediction of climate in this region. We have analysed a range of climate models and performed additional sensitivity studies to better diagnose simulated decadal variability and the cause of model deficiencies. The role of the dominant pattern of winter atmospheric variability – North Atlantic Oscillation – in driving North Atlantic decadal variability was confirmed, and we showed that much of the observed decadal variability might result from chaotic atmospheric motions. The degree to which ocean-atmosphere interaction drives a coupled and potentially more predictable mode of variability is not clear. Our analysis of existing models suggests that two-way ocean-atmosphere interaction does not play an important role in enhancing decadal variability in the North Atlantic Sector. However, the climate models showed that variations of the Atlantic Meridional Overturning Circulation, which dominates poleward heat transport in the ocean, plays an important role in sea surface temperature variations of the subpolar North Atlantic. Major uncertainties in the models exist in the contributions of temperature and salinity to modulating deep oceanic convection in high-latitudes. In addition, uncertainties in the response of the ocean to external radiative forcing were related to highlatitude surface salinity budget. Overall, our results suggest that climate prediction in the North Atlantic sector will benefit from a reduction of mean model errors, which could improve the representation of both oceanic variability and how it interacts with the atmosphere, and by resolving stratospheretroposphere interaction. Assessment of the predictability of North Atlantic climate on decadal timescales is confounded by limited knowledge of past oceanic variations, and the period over which prediction experiments can be performed. Our work demonstrated that past variations of the Atlantic Meridional Overturning Circulation might be best constrained in the subpolar North Atlantic. Furthermore, we developed a novel technique to reconstruct past oceanic variations using only sea surface temperature data and a coupled model. Lastly, we explored statistical methods that could be used to both assess predictability of observed climate and to benchmark decadal prediction systems. In the long-term we expect these findings will help to provide a robust assessment of the predictability of North Atlantic climate. This information is critically important for the potential users of decadal climate predictions. The project attracted significant media attention: the group was featured on the Deutsche Welle Brilliant Minds series; the results on ocean and atmosphere communication were highlighted by EurekAlert and ScienceDaily; and the data assimilation work featured in the Norwegian 2°C Magazine.
Publications
- 2010: Prospects for decadal climate prediction. Wiley Interdisciplinary Reviews: Climate Change, 1, 627-635
Keenlyside, N.S. and J. Ba
- 2011: A perspective on decadal climate variability and predictability. Deep Sea Research Part II: Topical Studies in Oceanography, 58, 1880-1894
Latif, M. and N.S. Keenlyside
- 2013: A mechanism for Atlantic multidecadal variability in the Kiel Climate Model. Clim. Dyn., 41, 2133-2144
Ba, J., N.S. Keenlyside, W. Park, M. Latif, E. Hawkins, and H. Ding
(See online at https://doi.org/10.1007/s00382-012-1633-4) - 2013: North Atlantic Ocean control on surface heat flux on multidecadal timescales. Nature, 499, 464-467
Gulev, S.K., M. Latif, N.S. Keenlyside, W. Park, and K. P. Koltermann
(See online at https://doi.org/10.1038/nature12268) - 2013: Potential of equatorial Atlantic variability to enhance El Niño prediction. Geophys Res Lett, 40, 2278-2283
Keenlyside, N.S., H. Ding, and M. Latif
(See online at https://doi.org/10.1002/grl.50362) - 2013: Stochastically-forced multidecadal variability in the 2 North Atlantic: A model study, Clim. Dyn., 43, 271-288
Mecking, J.V., N.S. Keenlyside, and R.J. Greatbatch
(See online at https://doi.org/10.1007/s00382-013-1930-6) - 2013: Stratosphere key for wintertime atmospheric response to warm Atlantic decadal conditions. Clim. Dyn., 42, 649-663
Omrani, N.-E., N.S. Keenlyside, J. Bader, and E. Manzini
(See online at https://doi.org/10.1007/s00382-013-1860-3) - 2014: A multi-model comparison of Atlantic multidecadal variability. Clim. Dyn., 43, 2333-2348
Ba, J., N. Keenlyside, M. Latif, W. Park, H. Ding, K. Lohmann, J. Mignot, M. Menary, O. Otterå, B. Wouters, D. Salas y Melia, A. Oka, A. Bellucci, and E. Volodin
(See online at https://doi.org/10.1007/s00382-014-2056-1) - 2014: Simulated response to inter-annual SST variations in the Gulf Stream region. Clim. Dyn., 42, 715-731
Hand, R., N.S. Keenlyside, N.-E. Omrani, and M. Latif
(See online at https://doi.org/10.1007/s00382-013-1715-y) - 2015: North Atlantic multi-decadal variability - mechanisms and predictability. Climate Change: Multidecadal and Beyond, C.-P. Chang, M. Ghil, M. Latif, and M. Wallace, Eds., World Scientific Publishing Company, Singapur, ISBN 978-9814579926
Keenlyside, N. S., J. Ba, J. Mecking, N.-O. Omrani, M. Latif, R. Zhang, and R. Msadek
