We know that the Antarctic ice shelves are melting, however the pace of this loss and its cause are still uncertain. A new study from the Scripps Institute in San Diego has provided better resolution in both time and space to the first question and indicates that the melting of floating ice shelves has accelerated faster than previously thought over the last decade. This has important consequences for predicting the rate of sea level rises over the next century and provides more precise data which can be later be analyzed to try to determine the causes for such fast melting.
Melting of floating sea ice alone will not cause sea level changes, as described here. Their mass is already displacing water and contributing to current sea levels, so their phase change to water will take up the same volume. However, their loss can have a dramatic impact on the melting of the connected ice sheet over the continental shelf (known as grounded ice). Floating ice connects to ground ice at the grounding line point (see left part of figure above) – when all the floating ice melts away, this can cause an instability at the grounding line that will result in a fast flow of ground ice into the water. This WILL cause sea level rises, as the ice over the ground was not previously displacing sea water already. You can think of floating sea ice as a kind of guard or buffer preventing the ground ice from flowing into the water.
So understanding how fast floating ice is melting will provide important information about when ground ice will be affected and possibly lead to large jumps in sea levels. Although previous studies have examined melting rates, they have used linear fitting over large spatial regions or only used small, several-year timescales, which limits the ability to find spatially different rates as well as decadal oscillations.
The current study addresses this problem by combining three sets of satellite data providing information about sea ice heights, from which thickness data can be extrapolated. Known as satellite altimetry, this is currently the only robust method of analyzing sea ice changes over large areas and long time periods. In contrast to previous work, the researchers were able to resolve sea ice changes into small cells (30 km sides) and over small increments in time (3 months) over the past 18 years. The hope is that this type of precision will provide more detailed data about where the greatest loss is occurring.
The good news: the researchers did collect better resolved data using this technique. The bad news: this better resolved data indicates that the Western Antarctic ice shelf is melting faster than previously known. Here’s a general visual description of the results:The red and blue dots indicate regions with rates of net loss and gain, respectively, over the past 18 years (the size of the dot indicates the rate magnitude). Clearly, most regions across the Western shelf have been melting, compared to small increases in thickness for floating shelves along the Eastern side.
This, however, only tells part of the story, because it is averaged across the past 18 years. The primary advantage to this study was not only the spatial but also temporal resolution. Analyzing time series of thickness changes provides some more interesting data, as shown below for each named shelf (compare with its location on the figure above):Each gray point corresponds to a 3-month time-averaged thickness change. Looking at these thickness changes over time allows us to see trends of where the ice shelves are heading. A polynomial fit in blue also highlights any oscillations in the data.
So what do we see? The first five locations – Ross, Amundsen, Bellingshausen, Larsen, and Filchner Ronne – are all along the Western shelf and all show significant melting trends across the past 18 years. The pattern of loss for the Ross shelf is a little different, showing oscillations over the first 10 years before plummeting over the last eight or so. The Ross shelf is located on the southern part of the Western shelf (see graphic above), so this clearly indicates some different forcing variables affecting that region compared to the far West shelf containing the other three that all show similar behavior.
A similarly troubling but different pattern appears for the Eastern shelves. All of them actually show increases in thickness from 1994-2003, balancing out the loss from the Western shelf and leading to an overall steady thickness averaged across all of Antarctica (bottom time-series from 1994-2003). However, over the past decade or so, even Eastern shelves have begun to see slight losses, and this combined with accelerated Western shelf losses has led to a recent decrease in total Antractic sheet thickness (bottom time-series from 2003-2012). Only time will tell if these Eastern losses are part of an oscillation over long time-scales, or if a climate forcing is changing their trajectory.
Just to give some numbers to these trends, from 2003-2012, the Western shelf loss rate has increased by 70%, fro 144 – 242 cubic kilometers per year. Across all of Antarctica, little loss was seen from 1994-2003 (25 km3/year), but has been declining rapidly from 2003-2012 (310 km3/year).
The Antarctic ice shelves have appeared to enter a new pattern of behavior – a natural question now would be to wonder about the cause of this accelerated melting. The authors do not address this question, as it is a complicated one, with the obvious suspects being anthropogenic emissions, El Nino, or other Pacific and Atlantic decadal oscillations. Only more research can answer this definitively.
Paolo, F., Fricker, H., & Padman, L. (2015). Volume loss from Antarctic ice shelves is accelerating Science DOI: 10.1126/science.aaa0940