The overall goal of this collaborative project was to investigate the tropical and mid-latitude (collectively referred to as lower latitudes) oceanic and atmospheric drivers of regional Arctic changes as well as the Arctic impacts on the Northern Hemisphere climate and weather variability, using a combination of available observations and state-of-the-art climate models, to improve our understanding of the fundamental dynamics, modeling capabilities, and capacity for prediction. The primary foci were on how the Arctic influences the lower latitude climate changes through atmospheric connections, and on evaluation of and enhancing capacity for seasonal-to-decadal predictions in the Arctic and over the Northern Hemisphere. The project involved substantial collaborations with the European-Union funded Blue-Action Project led by the Danish Meteorological Institute involving many other, primarily European, partners. We highlight a few of our efforts and findings below.

To address the first focus area, we performed several sets of ensemble simulations using a state-of-the-art atmospheric model forced by different sea-ice and sea surface temperature conditions. We also performed additional sets of ensemble simulations with the Community Earth System Model (CESM) with the sea-ice conditions set to preindustrial, present day, and future conditions, respectively. Our simulations were combined with parallel simulations conducted by our Blue-Action partners to obtain robust conclusions based on this multi-model large-ensemble simulations. We found that the variability of the sea-level pressure near the Arctic driven by sea-ice variability is as large as that driven by the greenhouse gas increase, while the influence of internal atmospheric variability is much larger than both. Also, we found that a decrease in the Arctic sea-ice leads to a negative phase of the North Atlantic Oscillation (NAO) in mid-to-late winter over the North Atlantic sector. However, the amplitude of the NAO response to the Arctic sea-ice changes was found to be underestimated by the climate models compared to the observational estimates.

For seasonal-to-decadal predictions, we examined two different aspects of the Arctic and North Atlantic predictability. First, we found a new source of predictability for the spring sea-ice variability in the Barents Sea, which is a sea-ice dipole in the Arctic-Pacific sector in the preceding autumn, thus allowing a skillful prediction with up to a 7-month lead time. Over multi-year time scales, we found that the sea surface temperature variability, namely the Atlantic multidecadal variability, can impact atmospheric blocking over the high-latitude North Atlantic in a way that warmer sea surface temperatures lead to more frequent blocking conditions over Greenland and Iceland. This relation was found to be responsible for a high prediction skill for high-latitude blocking with a 2-8 year lead time in the CESM prediction system.

In another study, we investigated why the CESM prediction system failed to predict a very prominent extreme cold event (referred to as the cold blob) during 2015 in the subpolar North Atlantic. Our analysis showed that strongly positive NAO conditions during winter and spring 2015 likely sustained the cold blob but were not simulated in our model. A reason for this prediction failure is that the observed NAO state associated with this cold blob seems to be rather rare / exceptional and the model is unable to capture its features. This case could be a useful test bed for future prediction system development.

We also showed that marine climate predictions can be used to generate decadal-scale forecasts of shifts in the habitat and distribution of marine fish species, as exemplified by Atlantic mackerel, bluefin tuna, and blue whiting. We found statistically significant forecast skill for individual years for lead times of 3-10 years. These findings indicate that climate predictions can be translated into information directly relevant to stakeholders, and we anticipate that this will be critical in foreseeing, adapting to and coping with the challenges of a changing and variable future climate, particularly in most ocean-dependent nations and communities.

Our findings will help improve current capabilities for and understanding of seasonal-to-decadal predictions in the Arctic and over the Northern Hemisphere. Specifically, our work can be used to establish performance benchmarks for current seasonal-to-decadal predictions with a focus on key Arctic – lower-latitude interactions, and hence will advance our understanding of the mechanisms contributing to, as well as those currently limiting, prediction skill for climate and weather extremes over North America and Europe. Our work benefits not only the climate science community but also the broader science and application areas. In particular, a deeper understanding of Arctic – lower-latitude linkages, and our capacity for predicting Arctic and Northern Hemisphere variability at seasonal-to-decadal lead times, are important to a wide variety of societally-relevant applications, such as weather forecasting, fisheries management, commercial shipping, commercial insurance, and naval operations. Our project trained and educated two postdoctoral scientists (one female), one graduate student, and one undergraduate student. Both of the postdoctoral scientists have now tenure track faculty positions.

Last Modified: 12/18/2022
Modified by: Gokhan Danabasoglu