Monday, December 31, 2012

Is Climate Change causing an exponential rate of Ice sheet Mass Loss, sea level rise?

Climate scientist James Hansen and his colleague Makiko Sato have released a new discussion paper with updated data on ice sheet mass loss from Greenland and Antarctica, with implications for possible multi-metre sea level rise this century. It makes for some interesting reading - there is a link to Hansen's website and the paper at the end.

The thesis that Hansen has put forward for several years is that Ice Sheet collapse is a non-linear process: that with the inclusion of amplifying climate feedbacks it is likely to follow an exponential rate of acceleration - a doubling rate. It might be a 10 year doubling time, or less. This will lead to extensive sea level rise, perhaps in the order of 5 metres this century.

But accurate data measurements of ice mass loss via laboriously estimating mass input and output has only been available since the early 1990s, and accurate satellite measurements (Gravimetry) via the GRACE satellites since 2000.

What these measurements show is that ice mass loss from both Greenland and Antarctica are accelerating, but the data for the time period is still too short to determine whether ice sheet mass loss will follow a somewhat linear path, or an exponential path doubling every 10 years or shorter time period.

Caption: Figure 2: Greenland (a) and Antarctic (b) mass change deduced from gravitational field measurements by Velicogna (2009) and best-fits with 5-year and 10-year mass loss doubling times. From NASA: Earth's Climate History: Implications for Tomorrow Hansen and Sato July 2011

Update 14 Jan 2013: Matt Owens from Fairfax Climate Watch had a similar idea in reviewing Hansen's discussion paper. Read his initial analysis: Sea level rise could crimp GDP; US direct losses could top 1/4 trillion per year during 2040-2050. On 11 January 2013 he updated his analysis with a new article: Surprising negative feedback could mean epic disaster. Both are worthwhile reading.

Current ice sheet models too Conservative

The current ice sheet models simulate paleo-climate sea level change. But dissolution of ice sheets in paleo-climate scenarios involved relatively weak climate forcings acting over many thousands of years. We know from looking at paleo-climate changes that multi-metre sea level rise is possible within a century timeframe even with weak forcings.

Hansen said at the Fall AGU in December 2011 that "The paleo-climate record reveals a more sensitive climate than thought, even as of a few years ago. Limiting human-caused warming to 2 degrees is not sufficient," and that the paleoclimate record points towards potential rapid climate change.

We are now hitting the climate with the equivalent of a sledgehammer over a relatively short period of 100-200 years, several orders of magnitude greater than occurred in the paleo-climate. Hansen says that a forthcoming paper will present evidence "that even paleo-climate data do not support the degree of lethargy and hysteresis that exists in such ice sheet models."

According to the most recent data Greenland mass loss has been accelerating faster and faster over the last decade. The current rate of mass loss with amplifying feedbacks could fit either a linear or an exponential curve at the moment. Hansen and Sato calculate the doubling response:

"A 10-year doubling time would lead to 1 meter sea level rise by 2067 and 5 meters by 2090. The dates are 2045 and 2057 for 5-year doubling time and 2055 and 2071 for a 7-year doubling time."

You just need to look at the 2012 Arctic report card showing global warming breaking records, the unprecedented surface melt experienced in July, or the study that found the Global Warming threshold for Greenland Ice Sheet collapse reduced to 1.6 degrees C - all from 2012 to be concerned.

A study published in November 2012 found that there is a rapid response connection between polar temperature changes, ice volume and sea level according to a new method of dating of sea level records and comparing them with temperature data and ice core records for the last 150,000 years. Read more about this at Skeptical Science: Past 150,000 Years of Sea Level History Suggests High Rates of Future Sea Level Rise.

Caption: Exponential Sea level rise projection based on Hansen and Sato (2012) "This chart represents the 5-year doubling time scenario; negative feedbacks kick in after 1 meter of sea level rise is reached in 2045, and by 2067 rates of sea level rise slow significantly - but still continue at a devastating rate. I have added an approximately inverse rate of slowing SLR (starting before the 5 meter point) to demonstrate one potential rate of negative feedback." Used Courtesy of Fairfax Climatewatch

The basics: Sea Level Rise

Sea level rise occurs from multiple factors:

  • thermal expansion of the ocean when temperatures rise - a well understood linear process
  • Melting of small land glaciers and small ice caps
  • Groundwater extraction less any increased on land storage capacity (dams)
  • Ice Sheet mass loss (Greenland and Antarctica primarily)

Scientific reticence and the IPCC

One of the problems with the IPCC process is that it is a multi-level policy consensus process that can filter out scientific work pushing the boundaries or whose results or modelled predictions are far outside the consensus range. This work will tend to be excluded as part of the review process.

The problem with this is that some of these outlying model predictions may prove more accurate than the consensus. We therefore miss factoring in these outlying predictions when doing our risk assessment and policy formulations for risk management. Many of the climate feedback mechanisms, tipping points, fall into this category and aren't adequately considered in IPCC reports.

So we dither on making small national policy decisions on a very conservative view of the science. For good policy decisions we need to know the outlying risks to take them into account, which we are clearly not doing for climate change.

The question of Sea level rise this century epitomises the problems in the IPCC process to date. A recent study published in November 2012 found that Sea Level rising 60% faster than IPCC projections.

Five years ago the IPCC report (2007) projected sea level rise by 2100 of about 29 cm (midrange 20-43 cm, full range 18-59 cm), but excluded any contributions from Ice sheet loss as they decided ice sheet physics was not developed sufficiently to provide accurate estimates.

In 2007 the distinguished climate scientist from the Potsdam Institute Stefan Rahmstorf estimated that even using a linear model relationship between temperature and sea level resulted in about 1 metre of sea level rise by the end of the century. Other studies since then have shown 0.75-1.9 m for the full range of IPCC climate scenarios, or 0.9-1.3 m by 2100 for a middle IPCC scenario (A1B). So by 2009 the 1 metre sea level rise had become the consensus. The outlying prediction of 5 metres of sea level rise is Hansen's, made in a scientific paper in 2007.

Hansen's argument is that the consensus predictions are primarily based on linear models and extrapolation, but that ice sheet disintegration is highly non-linear due to the nature of ice sheet dynamics and the presence of climate feedback mechanisms.

This youtube video from NASA uploaded on 8 February 2012 shows the GRACE ice mass loss from Greenland and Antarctica in a 21 second animation.

Climate Feedbacks kicking in

We are already seeing with record reduction in summer sea ice a non-linear climate feedback at work. What was predicted to take until mid-century to occur according to most modelling, now appears certain to occur this decade and perhaps as soon as 2016. Reduction in sea ice reduces the albedo (reflectivity) allowing more warming to occur.

That is just one feedback mechanism. There are more.

There are signs of permafrost thaw and perhaps the start of destabilisation of methane hydrates in ocean sediments which are also important climate feedback mechanisms enhancing global warming in the Arctic. Methane is a very strong greenhouse gas 21 to 25 times as strong as carbon dioxide over a 100 year period. But for a shorter period under 20 years it can be 100 times the strength of CO2. A really powerful feedback mechanism in the short-term.

Warmer ocean waters from the North Atlantic are now entering the Arctic Sea from Fram Strait adding to the warming in the Arctic sea and decline of sea ice.

In the Antarctic southern ocean warm currents are undermining the ice shelves that hold back the massive glaciers from discharging. There are also potential methane reserves in the Antarctic which may at some stage be triggered adding more climate feedback.

Caption: Increase in mass loss by the West Antarctic ice sheet. The mass loss has been steadily increasing since the 1970s as a result of accelerations in glacier flow; snowfall has not changed significantly in Antarctica over the past 50 years. Source: Rignot E. et al. 2008b. Rignot, E. 2008. available for download.

We know the West Antarctic Ice sheet is unstable due to it's grounding in a deep depression below sea level. The Antarctic Peninsula and much of West Antarctica are now warming at rates 3 times the global average, to match the warming in the Arctic. Pine Island and Thwaites Glaciers have shown a continuous rate of acceleration of mass loss over the last decade. Here is what Hansen says on this process in his new paper:

"West Antarctica's >Pine Island Glacier (PIG) illustrates nonlinear processes already coming into play. The floating ice shelf at PIG's terminus has been thinning in the past two decades as the ocean around Antarctica warms (Shepherd et al., 2004; Jenkins et al., 2010). Thus the grounding line of the glacier has moved inland by 30 km into deeper water, allowing potentially unstable ice sheet retreat. PIG's rate of mass loss has accelerated almost continuously for the past decade (Wingham et al., 2009) and may account for about half of the mass loss of the West Antarctic ice sheet, which is of the order of 100 km3 per year (Sasgen et al., 2010)."

Iceberg Cooling effect may temporarily slow warming, increase storms

The amplifying feedbacks are not all positive. Hansen and Sato in their simulations show that at about 1 metre of sea level rise a strong negative feedback will kick in.

With substantial cooling and freshening of the North Atlantic and Southern Ocean as the giant ice bergs melt, sea level rise may slow. The bad news is that increased latitudinal temperature difference, due to these temporary regional cooling effects, will drive the formation of more powerful cyclonic storms: superstorms. Much like Hurricane Sandy, only warmer air will carry more moisture, and sea levels will be much higher: look out for the storm surge and flooding that results.

Fig. 9. Surface air temperature change in 2065 (above) and 2080 (below) relative to 18501900 in simulations with GISS climate model using IPCC A1B scenario. Maps on left include ice melt, which is put half into the North Atlantic and half into the Southern Ocean, with ice melt doubling every ten years. (From Hansen and Sato 2012)

Hansen and Sato state:

"We conclude that available data for the ice sheet mass change are consistent with our expectation of a non-linear response, but the data record is too short and uncertain to allow quantitative assessment. A 10-year doubling time, or even shorter, is consistent with the gravity field data, but because of the brevity of the record even a linear mass loss cannot be ruled out. Assessments will rapidly become more meaningful in the future, if high-precision gravity measurements are continued."

Policy Response Inadequate

Our policy response on climate change on a worldwide level so far has been less than half-hearted. From Copenhagen climate talks in 2009 to the recent Doha climate negotiations, action on reducing emissions is far behind what the science is telling us we need to take on a global scale. We are in a critical decade if we are to limit global warming to the widely agreed level of 2 degrees Celsius.

If drastic reductions in emissions are not made this decade - by both the first world and the developing world, we are heading for a climate meltdown of 6 degrees C or more.

We need to realise that at this stage Sea Level Rise is unstoppable: emissions reduction needed to enable climate adaptation.

Just on sea level rise some countries are only starting to adapt coastline infrastructure in expectation of 1 metre of sea level rise. But what if Hansen is correct and Ice Sheet mass loss does proceed on an exponential basis. Any multi-metre sea level rise will be extremely costly and disastrous, let alone 4 or 5 metres towards the end of the century.

Although we may get some negative climate feedback happening at the 1 metre sea rise level, the huge cyclonic storms that will result from the temperature differentials will devastate the north American east coast. We saw a taste of that with Hurricane Sandy.

With multi-metre sea level rise there will be substantial incursion and innundation of fertile farming land. Underground aquifers we rely on for drinking water will become contaminated with sea water and unusable for drinking or farming.

Many low lying regions around the world will be decimated causing migrations of people. Small Island States in the Pacific, Indian Ocean may be the first to go under, but Caribbean nations will be at risk. Large parts of Bangladesh will become unliveable. So will much of Florida and California.

Matt Owens from Fairfax, Virginia has done a review of Hansen's latest paper and applied the direct impacts to US GDP: that direct losses could top 1/4 trillion per year during 2040-2050 with millions of houses threatened with innundation as well as the destruction of commercial infrastructure.

At 1.1 metres of sea level rise in Australia, thousands of residential and commercial buildings will be at risk.

Australia thinks it has a refugee crisis with 5000 boat people currently, but this will be dwarfed by the migrations to come in the following decades fleeing drought, famine and sea level innundation.

Biodiversity will also suffer. Mangroves, seagrass meadows and coastal wetlands, all valuable carbon sinks, will be innundated more rapidly than they can advance, and often their advancement will be blocked by human infrastructure. As humans move their farming and infrastructure away from the coast, ecosystems and biodiversity further inland will suffer in a secondary impact.

Up to now we have only been looking at this century, but ice sheet collapse will continue raising sea levels many metres in forthcoming centuries. The result is likely to be more than 20 metres of sea level rise for future generations. This article, based on linear assumptions, looks at sea level rise for the next 500 years.

It is time to face the challenges of climate change. Watch Professor Kevin Anderson explaining in the Cabot Institute Annual Lecture 2012 at the University of Bristol in the UK the challenges we face with combating climate change. The future is looking pretty bleak, but unless we do a realistic assessment we are not going to be successful in resolving the challenge of climate change. The stakes are high with human civilisation and human survival in the balance.


  • Dr James E Hansen website - Dec. 26, 2012: Update of Greenland Ice Sheet Mass Loss (PDF). Discussion of how the loss rate is changing.
  • Lead image from Hansen and Sato 2012 - Fig. 1. Annual mass change of Greenland ice sheet based on the input-output method, an analysis of gravity measurements, and a best-estimate composite (Shepherd et al., 2012).