Understanding climate
for the benefit of society

New publication

196 results

Earth has been a snowball. In a new study, Heiko Goelzer and colleagues have used an Earth system model to study the transitions between a glaciated and a non-glaciated Earth, around 700 million years ago.

During the Last Glacial Maxium the atmosphere contained less than half as much CO2 as today. New study finds that both the biological carbon cycling and the ocean circulation were different. 

Mighty floods have carved out deep canyons on Earth. New research suggests this required less power than previously believed. Collecting such data, however, may be demanding. 

Injecting particles into the atmosphere would reduce the temperature increase. But for the world’s ecosystems there is no alternative to mitigation efforts.

Sea level variations in Northern Europe are influenced by winds high above the Atlantic, new study shows. In the western part of the North Sea, the wind direction is more important than the wind speed.

Future climate scenarios are beyond comparison with recent times. When evaluating models, researcher Petra Langebroek goes 50 million years back.

A comparison of Greenland ice sheet models shows that while the surface mass balance is most realistic in the most complex models, simpler and faster models compare fairly well.

Cyclones in the Gulf Stream and Kuroshio regions

Cyclones in the Gulf Stream and Kuroshio regions Anonymous (not verified) Wed, 11/18/2020 - 14:38 Cyclones in the Gulf Stream and Kuroshio regions Cyclones in the Atlantic and Pacific Oceans are affected by the Gulf Stream and the Kuroshio. The temperature of the water below is only one of the factors controlling their development.

Written by Leonidas Tsopouridis, researcher at the Bjerknes Centre for Climate Reserach and the Geophysical Institute at the University of Bergen.

We compare the characteristics of cyclones following different tracks relative to the SST front in the Gulf Stream and the Kuroshio region. We highlight the importance of the land-sea contrast on increasing low-level baroclinicity and thus cyclone intensification in the Gulf Stream region, while we underline the role of the Pacific upper-level jet on intensifying cyclones and increasing large-scale precipitation in the Kuroshio region.

The Northwest Atlantic and the Northwest Pacific feature strong SST gradients providing favourable conditions for wintertime cyclone intensification in the midlatitudes. To investigate the cyclone characteristics and the mechanisms leading to cyclone intensification in the two regions we track individual cyclones and categorize them depending on their propagation relative to the SST fronts, which were identified using an objective frontal detection scheme.

For the cyclone propagation, we considered five categories, yet concentrated on cyclones staying either on the cold (C1) or warm (C2) side of the SST front, and those crossing the SST front from the warm to the cold side (C3).

a) Schematic of the cyclone classification based on the cyclone position relative to the SST front, b) an example of a C3 cyclone in the Gulf Stream region.

To compare the characteristics of cyclones following different tracks relative to the SST front, we performed a cyclone-relative composite analysis around the time of maximum intensification.

In our study for the Gulf Stream region, we identified that the land-sea contrast clearly influences cyclone intensification, particularly for cyclones propagating to the north of the SST front (C1), while the increased low-level baroclinicity observed for C3 cyclones is partially attributable to the Gulf Stream SST front. Both C1 and C3 cyclones are also associated with strong upper-level forcing.

We found convective precipitation to be closely related to the SSTs under the cyclones, while we hypothesized that low-level baroclinicity, via increased isentropic upglide, might be an additional factor determining precipitation intensity, in addition to moisture availability and cyclone intensity. However, given the special geographic characteristics of the western North Atlantic a generalisation of our results for other western boundary currents would not be straightforward.

Therefore, in our second study we focused on the Kuroshio region, which is located further away from the continent. Not surprisingly, we found land-sea contrast to play a less prominent role for the low-level baroclinicity in the Kuroshio compared to the Gulf Stream region. Contrary to our results for the Northwest Atlantic, we found cyclones remaining always on the warm side of the Kuroshio SST front to deepen more rapidly.

The propagation of cyclones over higher SSTs has been proved to enhance cyclone intensification, but the SSTs on the warm side of the Kuroshio SST front are lower compared to the respective in the Gulf Stream region, where C2 cyclones intensified the least. Moreover, low-level baroclinicity is weaker around the Kuroshio than around the Gulf Stream.

Therefore, we considered the upper-level forcing. We found the higher intensification of both C2 and C3 cyclones in the Kuroshio region to be consistent with their location close to the left exit of an intense upper-tropospheric jet stream, which also contributed to the higher observed large-scale precipitation.

Overall, our feature-based analyses highlighted the characteristics (e.g., moisture, surface heat fluxes, precipitation, wind speed) of cyclones following different pathways in the Gulf Stream and Kuroshio regions and identified the different mechanisms leading to cyclone intensification in the two regions.

Despite the fact that higher baroclinicity observed for cyclones in C3, was partially attributable to the SST fronts, which provide a conducive environment for cyclone growth, our results do not suggest a direct impact of the SST front on the intensification of individual cyclones propagating over the two ocean basins.


Tsopouridis L, Spensberger C, Spengler T., (2020) Characteristics of cyclones following different pathways in the Gulf Stream region, Quarterly Journal of the Royal Meteorological Society, 1-16, https://doi.org/10.1002/qj.3924

Tsopouridis L, Spensberger C, Spengler T., (2020) Cyclone Intensification in the Kuroshio Region and its relation to the Sea Surface Temperature Front and Upper-Level Forcing, Quarterly Journal of the Royal Meteorological Society, 1-16, https://doi.org/10.1002/qj.3929

Effects of global warming visible for tens of thousands of years

Effects of global warming visible for tens of thousands of years Anonymous (not verified) Mon, 11/16/2020 - 13:50 Effects of global warming visible for tens of thousands of years In 10,000 years the Greeland ice sheet will be gone, but what about Antarctica? Heiko Goelzer presents research for the coming millennia.

Written by Heiko Goelzer, researcher at the Bjerknes Centre and NORCE

Most projections of global sea level rise run until the end of this century. However, the melting of entire ice sheets is a very slow process and once the ice sheets are out of balance, it may take hundreds to thousands of years before a new equilibrium is reached.

In our new study, published in the journal Earth System Dynamics, a group of researchers from Brussels and Bergen has made projections for sea level rise over the next 10,000 years, using coupled models of the Greenland and Antarctic ice sheets, the atmosphere and the ocean.

Jonas Van Breedam who led the study states: “Emissions of greenhouse gases over the next 30 to 200 years could result in a global average sea level rise of 9 to 37 metres over the next 100 centuries.” 

We made projections for sea level rise over the next 10,000 years for a series of climate scenarios ranging from a CO2 peak over the next 30 years to the emission of most available CO2 reserves over the next 200 years. The highest scenario also takes into account an increased greenhouse effect when the permafrost on land and ocean floor melts and releases methane in large quantities. 

Depending on these greenhouse gas emissions, sea levels could rise by between 9.2 m and 37.4 m in 10,000 years. The Greenland ice sheet disappears completely in all scenarios, while the Antarctic ice sheet remains largely intact in a low-emission scenario and could lose up to half of its mass in a high-emission scenario. It is only after 10,000 years that ice sheets come into balance with the surrounding climate and the change in sea-level falls back to a maximum of 5 cm per 100 years.

In the high-emission scenario, sea level rises by more than 1 m per century on average during the first 2,000 years. For the low-emission scenario, the rate of sea-level rise is 10 times lower for the same period, at an average of 11 cm per century. The rate of sea-level rise is thus highly dependent on the total amount of CO2 emitted. 

Our study shows that the effect of current warming on ice sheets and sea level will remain visible for thousands of years, mainly due to the long response times of the Greenland and Antarctic ice sheets adjusting to a change in climate conditions.

Sea level rise
Contribution of the Antarctic ice sheet, the Greenland ice sheet, the expansion of ocean water and mountain glaciers to global sea level for a low-emission scenario (MMCP2.6) and a very high-emission scenario (MMCP-feedback). 


Van Breedam, J., Goelzer, H., and Huybrechts, P.: Semi-equilibrated global sea-level change projections for the next 10 000 years, Earth Syst. Dynam., 11, 953–976, https://doi.org/10.5194/esd-11-953-2020, 2020.

Cyclone variability in the Barents Sea

Cyclone variability in the Barents Sea Anonymous (not verified) Tue, 11/03/2020 - 19:53 Cyclone variability in the Barents Sea The jet stream is a highway for cyclones, while the sea ice edge has been thought to be a fuel station. Erica Madonnas new study shows that the fuel for cyclones is not simply linked to the ice edge.

Written by Erica Madonna, researcher at the Bjerknes Centre and the Geophysical Institute at the University of Bergen.

Extratropical cyclones (low-pressure systems) transport warm and moist air from the North Atlantic poleward. Some cyclones enter the Arctic and have strong impacts on the sea ice, including surface warming, sea ice drift and break-up, but they do not all take the same route and have the same effects.

The current study aims to better understand what controls the number of cyclones reaching the Barents Sea, as well as to characterize their climate impact upon reaching the Arctic circle.

There are two contrasting points of view explaining what controls the path of cyclones at high latitudes. On the one hand, some studies suggest that the atmospheric circulation, such as the orientation of the jet stream and the location of quasi-stationary high-pressure systems (referred to as “blocking”) control the cyclones’ path. In this perspective, one can imagine the jet stream as the highways where the cyclones travel, while the high-pressure systems constitute an obstacle on their way poleward.

On the other hand, other studies suggest that the location of the sea ice edge determines the region with a lot of “fuel” available for the growth of cyclones (termed as the Eady Growth Rate, EGR) and influences their propagation. In this perspective, the retreat of sea ice during the last decade in the Barents Sea coincides with a poleward shift of the “fuel filling stations” and thus fewer cyclones travel into the Barents Sea.

Eady growth rate and wind
Figure 1: Differences in EGR (left, in 1/day) and wind at 300 hPa (right, in m/s) between the high (HC) and low (LC) cyclone composites. The red (blue) line shows the ice edge (0.15 sea ice area fraction) for the HC (LC) composite.

In this study, we want to understand if both perspectives can describe the observed cyclones variability in the Barents Sea. For this purpose, we select winter months with the highest (HC) and lowest (LC) number of cyclone tracks in the Barents Sea and produce composites of the EGR, the location of the sea ice edge and the large-scale flow (Figure 1, right panel).

The EGR is enhanced when many cyclones enter the Barents Sea (red shading). However, the enhanced EGR is not linked to the location of the sea ice edge, which is comparable in both composites. Rather, the changes in the EGR reflect changes in the upper level-flow (i.e. the jet stream).

In fact, during months with many cyclones, the upper-level jet is tilted and directed towards the Barents Sea, while during months with few cyclones the jet is zonal and displaced southwards. When the jet is displaced south, a high-pressure system (i.e. a blocking) often sits over the Barents Sea.

Thus, we conclude that the atmospheric circulation rather than the sea ice location controls the path of cyclones at high latitudes.

We further explore the climate impact of cyclones reaching the Barents Sea separating cyclone tracks based on their origin (Figure 2). Cyclones born in the North Atlantic south of 60°N (1) have a stronger warming effect than cyclones born in the Nordic Seas (2) or in the Barents Sea (3). Cyclones entering the Barents Sea from the North Atlantic also have a tilted jet stream, that favours the advection of warm air from the south.

We also show that the winter-to-winter variability in the number of Arctic cyclones in all categories is large and no robust trends are observed over the last forty years.

Cyclone tracks
Figure 2:  (Top row) Cyclone tracks based on cyclones’ origin (1 North Atlantic, 2 Nordic Seas, 3 Barents Sea) and (bottom row) composites of daily 2-meter temperature anomalies (shading, in °C) and wind speed at 500 hPa (contours, in 5 m/s intervals) for the first time step when the cyclone is in the Barents Sea (black box).


Madonna, E., Hes, G., Li, C., Michel, C., & Siew, P. Y. F. Control of Barents Sea wintertime cyclone variability by large‐scale atmospheric flow. Geophysical Research Letters, e2020GL090322.

The sea can nurture or calm storms

The sea can nurture or calm storms Anonymous (not verified) Mon, 11/02/2020 - 11:58 The sea can nurture or calm storms The ocean is a questionable babysitter. Kristine Flacké Haualand takes us to the cradle of midlatitude cyclones.

Written by Kristine Flacké Haualand, PhD candidate at the Bjerknes Centre and the Geophysical Institute at the University of Bergen. The article was first published on Scisnack.  

Between the warm tropics and the cold polar regions exists a broad belt of strong latitudinal temperature contrasts: the midlatitudes. The weather here is dominated by cyclones, associated with wet and windy weather, and anticyclones, associated with calm and sunny weather. Before these weather systems hit land and affect our daily lives and picnic plans, they have often formed and traveled over open ocean, from which they’ve picked up a lot of heat and moisture. Let’s start at the cradle of midlatitude cyclones and see how the ocean influences their intensity before they potentially grow to mature storms and make landfall.

Many baby storms are born along the Gulf Stream, where warm, tropical water flows poleward. With the warm ocean below, the cold sector of the storm gets heated by the ocean such that the horizontal temperature contrast across the storm core weakens. This is sad news for the baby storm because temperature contrasts provide an important energy source for her growth. But someone is happy! Further down the path of the storm sit people like us enjoying the benefits of calmer weather.

While the ocean provides heat that calms down the weather, the ocean also provides moisture to the storms. With more moisture comes more clouds that intensify the storms. The role of the ocean is therefore two-fold, with the heat from the ocean weakening storm development, and the moisture from the ocean intensifying it.

For typical environmental conditions, especially in a warmer climate, the effect of moisture dominates, such that the overall effect of the ocean is storm intensification. Maybe just enough to postpone those picnic plans?

Read about how clouds and rain influence lows here


Haualand, K. F., and T. Spengler, 2020: Direct and Indirect Effects of Surface Fluxes on Moist Baroclinic Development in an Idealized Framework. J. Atmos. Sci., 77, 3211–3225, https://doi.org/10.1175/JAS-D-19-0328.1.