This semester has been very busy. I have been working simultaneously on two papers, as well as written and submitted my doctoral thesis. Two days before I submitted my PhD-thesis I received some very good news. My paper on the Filchner overflow was accepted! I was very pleased to include the acceptance status in my thesis.
You can read the full version of the paper if you click here … or read the summary below:
During a large part of my PhD, I have been studying processes associated with the production and pathways of cold Ice Shelf Water (ISW) in the Weddell Sea (see map in Figure 1). ISW is formed under the Filchner-Ronne ice shelf and is flowing northward along the Filchner Trough. The ISW overflows the Filchner sill, mixes with warmer water masses and form Antarctic Bottom Water.
(You can read more about ISW here )
At the Filchner Sill, several year-long records of current velocity exist between 1977 and 2017. The records show large fluctuations in the Filchner overflow velocity. However, no previous studies have been able to figure out which mechanisms contribute to the strong current fluctuations. Most of the current records contain about one year of data, and are therefore too short to capture long-term variations that may be related to climate change or long-term variability. We focused instead on monthly time scales, and found a link between the variability of the Filchner overflow and the wind forcing. Strong wind along the continental slope leads to higher Filchner overflow velocity (Figure 2).
So how can the along-slope wind upstream of the Filchner Trough influence the Filchner overflow?
We think that the slope current, which is flowing westward along the continental slope, may hold the key to answering this question. In a previous model study (Daae et.al, 2017 ), we found that parts of the slope current takes a detour, and circulates over the Filchner trough mouth region during strong wind-forcing (indicated by the thin red arrow in Figure 1). This circulation may interact with the Filchner overflow and lead to enhanced overflow. Although the existing data set is insufficient to prove that this is what happens, we present measurements at different locations which are consistent with this hypothesis.
Elin was part of a team that deployed several moorings across the Filchner sill and the continental slope in 2017. We hope that the data from these moorings, will contribute to increase our understanding of the Filchner overflow variability and out hypothesis of interaction between the slope current and the Filchner overflow.
In February last year we recovered a mooring at the Filchner Ice Shelf front (See map below) that we since long had consider lost. The large German ice-breaker Polarstern had failed to reach it twice due to sea ice, and it had now been in the water for more than four years. When we reached the location with (the much smaller) JCR last year, the mooring was only a few hundred meters from the advancing ice shelf front, and the captain was somewhat hesitant to go there – but he did, and the acoustic release on the mooring SA responded and released as promptly as if it had been deployed the day before! Most of the instruments had run out of battery and thus stopped recording – but one of them were still running, providing a four year long data record!
The mooring had several temperature and salinity sensors, and the records from them showed that there is a pulse of very cold (-2.3C!) ice shelf water (see explanation below) leaving the cavity during late summer and autumn each year. The water has been cooled down so much through interaction with the ice shelf base at depth, that there are ice crystals forming within it as it rises and leaves the cavity (I’ll write about what the ice crystals did to our instruments in a later post). The salinity of the cold water was relatively high – telling us that the water most likely entered the ice shelf cavity in the Ronne Depression, west of Berkner Island (see map).
In an earlier paper**, we had shown (using a numerical model) that ice shelf water flowing northward along the Berkner island would turn east when it reaches the ice shelf front (because conservation of potential vorticity hinders water to flow across the ice shelf front where the water depth suddenly changes by hundreds of meters) and exit the cavity in the east. But now the data showed that water was exiting the cavity in the west anyway?! What about the potential vorticity?? Our data also show that when cold water is flowing out of the cavity in the west during late summer, there is layer of less dense (and warmer) water present above it. In the paper we suggest that the presence of the upper, lighter layer breaks the potential vorticity constraint. The layer of less dense water reaches down roughly as deep as the ice shelf itself – and you can imagine that to the outflow it acts as a continuation of the ice shelf.
We now know that water leaves the ice shelf cavity also in the west – but where does it go then? Is there a flow of dense ice shelf water also along the western part of the Filchner trough?
Ice shelf water: We define water that has a temperature below the surface freezing point (which is about -1,9C for sea water) as “ice shelf water”. The water leaving the cavity was as cold as -2.3C (See figure 2 above)! How can it be so cold? It is a combination of two physical facts: 1) The freezing point decreases as pressure increases and 2) water in contact with ice will have a temperature equal to the freezing point. In an ice shelf cavity we have ice in contact with water at large depth ( i.e. at large pressure) and the water will then be cooled down (the heat given off by the water is used to melt ice) to the local freezing point – and voila, you’ve got ice shelf water!
* I say my, but it’s a team effort: many thanks to J.B. Sallée who co-authored the paper and to all the people involved in deploying and recovering the moorings!
**Darelius, E., Makinson, K., Daae, K., Fer, I., Holland, P. R., & Nicholls, K. W. (2014). Circulation and hydrography in the Filchner Depression. Journal of Geophyscial Research, 119, 1–18. http://doi.org/10.1002/2014JC010225
Writing a scientific article is a long process – you collect the data, you calibrate them, process them and you analyze them. You plot them, think about them, discuss them, think about them again until hopefully, at some point, the data give you results that you can understand and – publish. So you write the paper – in between meetings and teaching you somehow manage to squeeze your outstanding results and neatly prepared figures into the template provided by the journal. Then you submit – and forget about it all until you hear back from the editor three months later: the REVIEWS are back… sometimes it’s like this:
i.e. you quickly find out that your results were not that outstanding and your figures not that neat… the reviewers have filled page after page with “Did you consider…”, “why didn’t you calculate…” how does this compare to..”, “can you really ignore the effect of….” and “you ought to refer to the paper by mr so and so”…so you start over, you do all the extra analyses that reviewer three asked for, you make new figures, you clarify and expand and include a citation of mr so and so (the reviewer?). You read and write the text over and over and at some point you realize that you’ve done all that they ask for… and that version 6.2 of the paper is indeed much better than version 1.0. So you write a very polite letter to the editor, where you respond to each and every comment from the reviewers and explain what you’ve changed – and then you resubmit. And you wait. Again. For three months.
… but then sometimes, you get three short lines from the editor stating that you paper is accepted! It will be published!!! YES!!!
I received one of these e-mails the other day – and once the paper get online in a couple of days I’ll let you know what it is all about!
Different types of experiments, and why we use such a weirdly-shaped “Antarctica” and are happy with it.
When we want to show people images of our model experiments in a tank, people often imagine that they will be shown cute little miniature landscapes, looking much like the ones you see for really fancy model train setups. And then they are hugely disappointed when they see pictures like the one below and we tell them that yes! that’s our Antarctica that Nadine is climbing on, while Elin is sitting in the Southern Ocean.
The kind of experiment everybody hopes to see could, according to Faller (1981), be classified as a simulation: representing the natural world in miniature, including every detail. Data from those experiments — since they would in theory be realistic representations of the real world — could be used to fill in missing data from the real world in regions that are hard to get real data from, like for example the Southern Ocean. However, since those experiments are designed to represent the complexity of the real world, interpretation of the experiments is as complex as it is to interpret data from the real world: There are so many processes involved that it is hard to isolate effects of individual processes.
The kind of experiments we are doing would be classified as abstractions. Faller describes this kind of experiment as similar to abstract art: Only the main features, or better: the artist’s interpretation of the main features, are reproduced and everything else is omitted. That makes the art difficult to understand for anyone who isn’t well versed in abstract art, but for the experts it is obvious what the point is.
In case of our experiments that means that we have all the relevant features, or better: our interpretation of what we believe to be relevant features, of Antarctica present in the tank: the parts of topography that we think have an influence on how the current should behave, i.e. a V-shaped canyon, a source that supplies water of the correct properties into the ambient “ocean” water, an ice shelf. And when that ice shelf is tilted, we feel like our experiments are already becoming pretty realistic!
These abstractions are the kinds of experiments in which you can, because they are relatively simple, develop new theories when new features of the circulation emerge that you then have to rationalize and include in your theories after the fact.
We have actually also done another type of experiment, a verification. I wrote about it in this post: we tilted the ice shelf because this is a case for which we actually knew from theory how our current should behave, in contrast to all the previous experiments where we didn’t actually know what to expect, and we were happy when we observed exactly what we expected based on theoretical considerations. So in this case the experiment wasn’t about discovering something new, but rather making sure that our understanding of theory and what goes on in the tank actually match.
Faller describes a last type of experiment: the extension. That is the kind of experiment that you could perform after a successful verification experiment: Pushing the boundaries of the theory. Does it still hold if the current introduced in the tank is very fast or very slow? If the water is very deep? If the slope of the ice shelf is very large or small? Basically, every parameter could now be changed until we know for which cases the theory holds, and for which it does not.
So why am I writing all of this today? Faller’s (1981) article, before he goes on to describe the framework to think about geophysical fluid dynamics experiments that I mentioned above and which I find quite helpful to consider, starts with the sentence “No one believes a theory, except the theorist. Everyone believes an experiment — except the experimenter.” On this blog, our goal is to bring the two together and not make anyone believe either of them, but to show how both can work together to mutual benefit.
Faller, A. J. (1981). The origin and development of laboratory models and analogues of the ocean circulation. Evolution of Physical Oceanography, 462-479.
After our excursion to ‘real’ Antarctica, we are back in the idealised world. Hopefully you all have followed the blog and have seen how our continental shelf has been constructed and the source for our current has been tested. The same thing has been done by one of our colleagues, Kjersti, on the computer to then run it with a regional ocean model. In Figure 1 you can see the similarity of the set-up that she created to the one in our tank. The advantages of here experiments are that she can add a wind-forcing of any strength at the surface and can also change the surface temperature and salinity fields to mimic the seasonally varying effect of sea ice. Furthermore, she located a dense source at the end of the trough representing the dense water outflow.
In her recent publication (Daae et al., 2017) she uses this model to study the sensitivity of the warm deep water entering a continental shelf with a coastal trough to the magnitude of wind stress, the shelf salinity and the upper-layer hydrography. She finds that stronger along-slope winds create a stronger slope current which is also shifted toward shallower isobaths, causing a stronger interaction of the flow with the trough. At low wind speeds the core of the current is located below the depth of the sill and is not affected. The southward transport of warm deep water increases for a denser outflow and higher salinities on the shelf. This effect is stronger for weak winds compared to strong winds, potentially because a strong barotropic flow passing the mouth of the trough will create less baroclinic instabilities.
They find that the warm deep water mainly accesses the shelf when no work against the buoyancy force has to be done. This is the case when dense water on the shelf connects the density surfaces between the shelf and off-shelf water masses. Furthermore, more warm water is found on the shelf in summer, when a fresh surface layer is present due to the sea ice melting. It induces a shallow eddy overturning cell that acts to flatten the isopycals, hence providing easier access to the shelf.
You see that you can sort of play god with these models, but you actually have to be very careful to choose all your parameters correctly and in a way that they representable for the processes in nature. Handled with care, models provide another important tool for understanding the climate system and individual processes.
Daae, K. B., Hattermann, T., Darelius, E., & Fer, I. (2017). On the effect of topography and wind on warm water inflow – An idealized study of the southern Weddell Sea continental shelf system. Journal of Geophysical Research : Oceans, 122, 2017-2033. http://doi.org/10.1002/2016JC012541
We have already written about the article of Elin, where she shows that for the first time pulses of warm water have been measured in the vicinity of the ice front. This means that under certain conditions the warm water, can travel several hundred kilometers south along the eastern side of the Filchner depression, i.e. our trough. Of course everyone wants to know whether the trough provides a permanent pathway to the south for the warm water, which would be a big threat to the Filchner Ronne Ice Shelf.
We will come back to this problem in a little while but first have to explain something that we call ‘Antarctic Slope Front’. I think most of you are familiar with weather fronts, where for example cold and warm air meet. The same things exist in the ocean, when warm and cold or light and dense water masses encounter each other. This exactly is the cast almost all the way around Antarctica where fresh and cold water is found on the continental shelf and warm water (CDW/WDW) flows along the continental slope. The resulting front is the Antarctic Slope Front (ASF) as show in Figure 1. The depth of this front determines whether the deeper warm water can access the continental shelf or not. During our experiments we will also introduce a density difference at some point by have saline water in the tank and a fresh source.
Several mechanisms can influence the ASF such as the wind and the upper-layer hydrography, both having a strong seasonal cycle. Therefore, it is no surprise that the ASF depth also varies seasonally and it has been shown by various authors, that it is shallower in summer, favoring on-shelf flow of warm water and deeper in winter, reducing the access for warm deep water.
So, now we know that there might not be warm water available at the shelf break to flow toward the ice front at any time. Another important factor is whether the circulation on the continental shelf would transport the warm water toward the ice front all year around. Hence, we took the German ice breaker RV Polarstern and went to the Filchner region to put some more instruments in the water and I analyse the data in my recent publication (Ryan et al., 2017). The map in Figure 2 should sort of look familiar to you now as it similar to the one in a previous article. You can see the coast to the right, then the continental shelf (Eastern Shelf) that opens up and the trough cross-cutting the shelf toward the Filchner Ice Shelf. We put moorings, i.e. a long upright floating lines with instruments at different depths, where the three red stars are on the map and left them there for two years. It is the eastern flank of the trough, where the warm water (dashed gray arrow) was observed to flow south adjacent to the northward flowing cold water, called Ice Shelf Water, emerging from underneath the ice shelf (blue arrow). You can see in Figure 3 how this looks like in a temperature section and where we took our measurements. You might wonder, why we did not put instruments shallower than 300m. If we did, we would risk these instruments to be ripped off by ice bergs, and there are plenty of them around. We found that there is only a certain period in summer-autumn, where we detect southward flowing warm water at our moorings. In winter, the water column becomes very cold and uniform with temperatures close to the surface freezing point and there is no southward flow anymore. So for now it seems like there is no permanent pathway for the warm water toward the Filchner Ice Front. However, in a warming climate the conditions on the continental shelf in winter could change, with warmer atmospheric temperatures and reduced sea ice production. The latter, could also reduce the production of ISW which is currently filling the whole trough and is sort of ‘blocking’ the warm water from entering the centre of the trough.
Of course we do not have winds, ice etc. in our tank experiments but there is still so much more to be understood on how the warm water can be transported on the shelf and how for example the ASF changes in the vicinity of a coastal trough. Most measurements or time series are too short or too scattered in order to really understand fundamental processes and mechanisms, this is where a big rotating tank can help us!
Ryan, S., Hattermann, T., Darelius, E., & Schröder, M. (2017). Seasonal Cycle of Hydrography on the Eastern Shelf of the Filchner Trough, Weddell Sea, Antarctica. Journal Geophysical Research – Oceans. http://doi.org/10.1002/2017JC012916
The shelf break and the ice shelf front—Two topographic barriers
So far, we’ve explained you that outside the Antarctic continent, warm and saline water (Circumpolar Deep Water, CWD) is located beneath fresh and cold water (Surface Water). We’ve also explained you that if this warm and saline water gets onto the continental shelf and mixes with high salinity shelf water, it melts the ice shelf from below. You think that sound easy? Well, it’s not! The ice shelves are actually pretty well protected from this warm subsurface water…
In Figure 5 from our post on Tuesday you can see that especially the largest ice shelves—Filchner-Ronne and Ross Ice Shelves—are protected from warm water through the continental shelves that keep the warm water at a distance of several hundreds of kilometers. Oceanic currents tend to flow along the bathymetry (slopes), not across it. The continental slope—the steep slope connecting the deep Southern Ocean to the continental shelf— thus acts like a wall and limits the flow of warm water onto the shelf. In the Amundsen and Bellinghausen Sea, however, the warm water already reaches on the continental shelf, and it reaches all the way to the ice shelf front. The ice shelf front reaches many hundreds of meter down into the water, and it forms a second wall that the water has to cross in order to reach the cavities beneath the ice shelves. The pathway of the warm water across these two walls, or topographic barriers as we like to call them, is still poorly understood and therefore the main focus of our project. How does the warm water that is located outside the continental shelf and in a depth of hundreds of meters flow onto the continental shelf and beneath the ice shelf?
The rotation of the Earth causes ocean currents flow parallel to topographic slopes, i.e. to roughly follow lines of constant depth. The currents around Antarctica therefore follow the continental slope, and water from the slope doesn’t easily make it onto the continental shelf. Similarly, at an ice shelf front currents mainly flow parallel to the front instead of entering the ice shelf cavity. The shelf break and the ice shelf front form a topographic barrier
Warm water has been measured on the continental shelf and beneath ice shelves. In fact, the water can cross topographic barriers after all!
Certain processes help the warm Circumpolar Deep Water to cross the barriers:
Troughs crosscutting the continental shelf and the shelf break reduce the barrier effect and enable on-shelf transport.
Eddies formed within the currents are able to move across the “barriers” bringing (warm) water with them.
During our time at the Coriolis platform, we will investigate these points by varying the trough geometries, current thickness and density. If you are interested, please follow our blog throughout the experiments and learn about the control of topography on ocean currents and ice shelf melt!
The Antarctic Ice Sheet is melting because it is losing more mass through increased air and ocean temperatures than it gains mass by snow fall. Melting of ice is most efficient through contact with water, because water has a higher heat conductivity compared to air; simply said this means that water removes heat easier from the ice than air. In case of the Antarctic Ice Sheet it is also of importance that air temperatures are mainly far below the freezing point, whereas the ocean is always warmer than or close to the freezing point. The strongest melting in Antarctica therefore occurs beneath ice shelves. You want to know more about the melting process and what happens beneath an ice shelf?
The ice pump beneath ice shelves
Ice shelves are melting from beneath as ocean water reaches into the cavity. To explain the processes, we come back to the sketch of an ice shelf that we already showed you before.
In front of ice shelves the ocean freezes to sea ice, which is a process that produces very saline and dense water (high salinity shelf water). Because of its high density, it sinks down and mixes with circumpolar deep water (CDW) that spills over the continental shelf edge. The water mixture is then relatively dense and warm. It flows along the bottom of the continental shelf that is—in many cases—deepening inland and therefore reaches far beneath the ice shelf. When the water reaches to the ice shelf base it is far below sea level and at a higher pressure. Water there can still be liquid at temperatures down to below -2⁰C! And the ice starts melting as soon as the temperatures reach above this pressure melting point! If the ocean water is warmer than this temperature, it melts the ice shelf from below. The melt water is fresh with a low density and brings water out of the ice shelf cavity in form of a buoyant melt plume. All together, this forms an ocean circulation beneath the ice shelf, called ice pump.
Where does the warm water come from?
The rates at which all Antarctic ice shelves melt from beneath are estimated to be about 1325 Gt/yr (gigatons per year; or 3.7 mm sea level equivalent per year). Yes, this is a lot! We are wondering where all the energy comes from and how the warm water reaches onto the continental shelf…
The dynamics governing the flow of warm water towards and underneath the ice shelves are non-trivial and still poorly understood. However, we know that the heat reservoir threatening the ice shelves is located off the continental shelf, in the deep Southern Ocean, where relatively warm water resides below a shallow, cold and fresh surface layer. Well, you may wonder how cold water can float on top of warm water!? If you have ever been swimming in a lake or in the ocean, you may have realized that it usually gets colder the further down you get.
Water layers in the oceans are stratified due to their different densities, with the densest water masses lying at the bottom of the sea floor and the lightest water masses floating at the surface. Density is given by the salt content and the temperature: high salinity and low temperatures lead to the densest water masses. In the tropics, the temperature is more important and it usually gets colder with depth. Closer to the poles, however, salinity gets more important. In the Southern Ocean, the warmer water can stay beneath the cold and fresh surface layer, because of its high salt concentration. The warm subsurface water it transported with the Antarctic slope current, which flows westwards along the Antarctic slope almost around the whole Antarctic continent.
West Antarctic threatened by a warm ocean
In West Antarctic, especially in the Amundsen and the Bellinghausen Sea, the water found on the continental shelf is relatively warm compared to East Antarctic (Figure 5), resulting in high melt at the base of the ice shelves. Some of the ice shelves therefore thin tens of meters per decade! This thinning also expands upstream to the proximal grounded glaciers which then causes a sea level rise. The figure shows that already today one drainage area in West Antarctica alone lost almost 100 Gt/yr to the ocean, with a trend towards increasing numbers due to basal malt. These basal melt rates are challenging to estimate, which makes the contribution of West Antarctic Ice Sheet melt to sea level rise to the largest source of uncertainty in the fifth assessment report of the Intergovernmental Panel on Climate Change (IPCC).
West Antarctic – A marine ice sheet
Besides the ocean temperatures, there is another substantial difference between West and East Antarctica that leads to the high mass loss in the former one: The bedrock beneath the West Antarctic Ice Sheet lies far below sea level (up to more than 1 km) in most parts (the blue areas in Figure 6). Thus, most of the ice sheet’s base is located far below sea level, what we call a marine ice sheet. Marine ice sheets are particularly instable for several reasons:
Their bed slopes inland, which generally is an instable configuration. As submarine melt moves the grounding line—the transition zone between grounded and floating ice—further inland, it reaches into deeper areas, that make the glacier floating again and also causes a larger flux through the grounding line. Once this process is triggered, the glacier retreats continuously until the bed geometry changes again.
The pressure melting point—the temperature at which ice freezes or melts—is reduced when the pressure increases. When the glacier base reaches into deeper waters where the pressure is higher, the melting point therefore decreases and the surrounding water is warm relative to the melting point. It may be warm enough to melt the ice from below.
The uncertainty regarding the West Antarctic Ice Sheet melt to sea level rise therefore results from the exposure of the ice sheet to relatively warm water. The ice sheet is also close to a tipping point at which the ice sheet may irreversibly lose large amounts of ice. To better estimate future sea level rise, we need a good estimate on the ocean heat flux to the ice shelves and good knowledge on the response of the ice sheet.
Are you exited about our experiments at the Coriolis platform? We can’t await to finally getting started with the experiments! To shorten the waiting time, we would like to introduce you into the topic of our research throughout the next three days, so that it will be easier for your to follow up on the happening in Grenoble. Today, we will introduce to the Antarctic Ice Sheet and its fringing ice shelves…
The Antarctic Ice Sheet and sea level rise
The Antarctic Ice Sheet contains 26.5 Mio km³ of ice. Such a big number is difficult to grasp, but it is 70% of the Earth’s freshwater and equals—if the ice is completely melted—a sea level rise by 58 m. Luckily melting of an ice sheet is a slow process which is compensated by the accumulation of snow as shown in the sketch below. However, this compensation got out of balance during the last two decades and we know from satellite observations that the Antarctic Ice Sheet is losing more mass to the ocean than it gains by snow fall. In average, it lost about 310 +- 74 km³ (or 0,78 mm of sea level rise) per year during 2003-2012; especially in West Antarctic this mass loss has accelerated so that the area has already lost 70 % more ice compared to a decade ago (Paolo 2015).
Ice shelves are important!
Most of Antarctica’s periphery (75%) is buttressed by ice shelves—the floating extension of an ice sheet. Ice shelves occur, where the ice is thin and the bed rock far enough below sea level that the ice gets lifted off the underlying bedrock. As soon as the ice is detached from the bed, ocean water can penetrate beneath it and start melting as shown in the sketch below.
The total area of all Antarctica’s ice shelves is almost as large as the whole Greenland Ice Sheet and cover about 11% of Antarctica’s area. In other words, 11 % of the Antarctic Ice Sheet is floating! Figure 3 shows the Antarctic Ice Sheet where all ice shelves are shown in color. The colors correspond to their thickness change: red indicates strong thinning, blue indicates slight thickening. The circles associated with the different ice shelves show the percentage of thickness change.
Now you may think that there is a lot of blue color, so thickening of the ice shelves. It’s not so bad after all?! Only looking at the eastern part you are totally right! Many ice shelves have actually been thickening since 1994, which means they were growing! However, the largest mass change is happening in West Antarctic, where the colors are reddest and the circles largest. The small time line in the lower left corner compares the volume change of East Antarctica (blue) with West Antarctica (red). The western part has been losing mass since 1994, whereas the eastern part actually gained mass in the beginning until also the eastern part started losing mass in 2003. The black line is the total mass loss and you see that the volume went down continuously since 2003—so, West Antarctica wins in the end!
When talking about thinning of ice shelves, we always have to keep in mind that ice is almost as heavy as water; 9/10th of an ice shelf is therefore floating below the sea surface. This means that they are already part of the ocean and do not contribute much to sea level rise when they melt. But still, their role is to buttress the ice that is drained through ice streams—slow rivers of ice—from the interior towards the ocean. Thinning will therefore accelerate the ice flow further upstream in the grounded part and increase the ice sheet’s contribution to sea-level rise by acceleration and increased mass loss.