you're reading...

Climate Change

Global Consequences of Melting the Pine Island Glacier


Green, J.A.M., and Schmittner, A., (2015), Climatic Consequences of a Pine Island Glacier Collapse, Journal of Climate 28, 9221-9234, DOI: 10.1175/JCLI-D-15-0110.1



The input of freshwater to the ocean resulting from melting or collapsing ice sheets has been well studied. It is therefore well understood that large pulses of freshwater (also known as “hosing”) can inhibit the formation of deepwater masses in the ocean. In the North Atlantic Ocean, inhibited deepwater formation results in the slowing or shut down of the Atlantic Meridional Overturning Circulation (AMOC), a system of currents that modulate the Earth’s climate (Figure 1).

Figure 1. Atlantic Meridional Overturning Circulation is part of the global ocean conveyor belt. This system of currents is responsible for the formation of water masses and the transport of heat throughout the global ocean.

Figure 1. Atlantic Meridional Overturning Circulation is part of the global ocean conveyor belt. This system of currents is responsible for the formation of water masses and the transport of heat throughout the global ocean. Photo: Modified from NASA Earth Observatory, retrieved from WikiCommons, file in the public domain.


On the topic of melting ice sheets, retreat of the Greenland Ice Sheet has been primarily in the spotlight, with less attention paid to melting in Antarctica. However, recent studies have shown accelerated retreat of portions of the West Antarctic Ice Sheet (WAIS), particularly the Pine Island Glacier (PIG) (Figure 2). The PIG serves as a major ice stream (meltwater outlet) for the WAIS, with nearly 10% of the WAIS being drained through the PIG to the Amundsen Sea. The topographic configuration makes the PIG extremely vulnerable to irreversible retreat due to melting. The base of the PIG sits below sea level, sloping down and away from the ocean. This creates a scenario such that once retreat of the PIG is in motion, there is nothing preventing its continued retreat. Evidence already exists for accelerated retreat of the PIG.

To better understand the global impacts of the retreating PIG, this study used three experiments relying on climate simulations. Meltwater pulses (hosing) from the PIG were modeled in three experiments to assess global changes in climate such as changes in sea surface temperature and AMOC.

Figure 2. Pine Island Glacier is located in the West Antarctic Ice Sheet. Melting of the Pine Island Glacier leads to hosing or meltwater pulses to the Amundsen Sea.

Figure 2. Pine Island Glacier is located in the West Antarctic Ice Sheet. Melting of the Pine Island Glacier leads to hosing or meltwater pulses to the Amundsen Sea.  Photo: Modified from NASA, retrieved from WikiCommons, file in the public domain.


Experiment 1: Sensitivity Simulations

Experiment 1 tests the effects of meltwater pulses on global climate by holding other climate variables constant. Pre-industrial climate conditions, atmospheric CO2 concentrations of 280 parts per million (ppm), were used to initialize this experiment and a 100-year meltwater pulse (hosing) was simulated. Three hosing rates were considered: 0.003 Sv, 0.006 Sv and 0.1 Sv. A Sverdrup is a measure of volume transport, or the rate at which a volume of water is transported (1 Sv = 1,000,000 m3 per second). The 0.003 Sv hosing scenario represents the present rate of hosing, whereas 0.1 SV represents a far higher amount of hosing due to the entire collapse of the PIG (the extreme future example). Experiment 1 continues to run for additional 400 years after the 100-year hosing to evaluate the impacts on global climate.


Experiment 2: Rapidly Increasing CO2 Simulations

Experiment 2 tests the effects of hosing of the PIG in a rapidly warming world. In this experiment, CO2 exponentially increases over 70 years, until it reaches a level twice that of the pre-industrial levels and then is held constant. The total duration of experiment was 500 years. Similar to Experiment 1, this experiment also used a 100-year hosing at 0.003 Sv and 0.1 Sv.


Experiment 3: “Realistic” Simulations

Experiment 3 represents a realistic climatic forcing scenario using an intermediate CO2 projection from the latest IPCC reports, where atmospheric greenhouse gas concentrations stabilize by the year 2150. This experiment begins in the calendar year 1800. At calendar year 2000, hosing begins at 0.003 Sv and ramps up to 0.1 Sv by the end of the 100-year hosing period. This hosing scenario is equivalent to the PIG losing half its volume over the 100-year period. An additional simulation was run where hosing increased from 0.003 Sv to 0.1 Sv over a longer period: 200 years.



Experiment 1: Sensitivity Simulations 

Despite small rates of hosing at the 0.003 and 0.006 Sv simulations, the atmosphere and ocean responses were fairly large. In all simulations, there was initial cooling in the Southern Ocean, the magnitude of cooling positively correlated to the rate of hosing. The North Atlantic experiences weak warming, resulting from the observable phenomenon called the bipolar seesaw. For conceptual purposes, imagine a planetary-scale seesaw aligned from the north pole to the south pole. When the southern polar region cools (pushes down on the seesaw), an opposite effect is felt in northern polar regions – warming (the seesaw rises)! This seesaw-like behavior occurs due to changes in global ocean circulation that affect heat transport from southern polar regions all the way to northern polar regions.

Approximately 100 years after the hosing ends, the seesaw tips the opposite direction, causing cooling in the N. Atlantic and warming in the southern polar region. By the end of the 500-year simulation the seesaw has tipped once again, with you guessed it, warming back in the N. Atlantic and cooling in the southern polar region.


Experiment 2: Rapidly Increasing CO2 Simulations

The rapid increases in atmospheric CO2 dominate these simulations resulting in warming of the surface ocean by 2 to 5 °C and a much lower global climate response to hosing when compared with Experiment 1. In the case of the 0.003 Sv hosing, the large increases in atmospheric CO2 completely overwhelm any impacts of the low rate of hosing. In the 0.1 Sv simulation, the hosing does impact AMOC, and there is noticeable cooling in the North Atlantic.


Experiment 3: “Realistic” Simulations

As is expected, the realistic simulations have results falling somewhere between Experiments 1 and 2. This is due to a more conservative increase in CO2, which results in a globally averaged surface ocean warming and a temporary decline in AMOC. The bipolar seesaw, which was very readily observed in Experiment 1, operates on top of rising CO2 in these simulations.



The bipolar seesaw, where the North Atlantic Ocean warms and the Southern Ocean cools due to northward heat transport, has been well known through observational data over the 20th century. This study provides evidence that the bipolar seesaw continues to operate under present hosing scenarios and will continue to operate under future scenarios. A consistent result in all simulated experiments is the warming of Greenland. This result suggests enhanced melting of Greenland, an ice sheet that is already experiencing accelerated losses. The authors also note that the experiments only consider hosing from the PIG, and there very well could be additional hosing from other portions of the West and East Antarctic ice sheets, which would result in even more extreme changes in global climate.

Brian Caccioppoli
I am a recent graduate (Dec. 2015) from the University of Rhode Island Graduate School of Oceanography, with a M.S. in Oceanography. My research interests include the use of geophysical mapping techniques in continental shelf, nearshore and coastal environments, paleoceanography, sea-level reconstructions and climate change.


No comments yet.

Talk to us!

oceanbites photostream

Subscribe to oceanbites

@oceanbites on Twitter