Ice melting at Arctic Sea 4: effects, bad or good?

Continuing the previous series, this will be the last post about sea ice melting. From the early posts, both observations and simulations show that sea ice at Arctic sea has declined significantly with melting rate accelerated. Sea ice would be likely to decrease continuously within this century, and might reach nearly ice-free status. So, what are the effects?

 Climate 

Arctic sea ice affects global climate change through complex processes. Firstly, there is positive feedback between warming and ice melting (as showed in Figure 1, Dessler, 2011). Current global warming has caused and will cause more sea ice melting at Arctic sea. Decreasing ice cover while increasing sea surface leads to the decrease in albedo, as sea surface reflects less and absorbs more solar radiations than ice surface. More absorbed energy leads to extra warming, thus, melting more ice. Therefore, ice melting at Arctic sea will amplify the global warming in this century. 

Meanwhile, sea ice also insulates the below sea water from the atmosphere, preventing thermal exchange and gas exchange between the ocean and the atmosphere. Sea ice loss at Arctic sea changes the interaction between the ocean and the atmosphere. Melting water decreases the salinity of sea water as well, which changes the density of sea water affecting ocean circulation. This will affects the climate at global scale.

Figure 1. Ice-albedo feedback

 Ecosystem

Sea ice is the habitat to many species, from fungus to large mammals like polar bears. However, sea ice loss has threatened their survival, affecting the population of the species. The US Geological Survey studied how the Pacific walrus and polar bears response to the rapid decrease of sea ice at Arctic sea. They found that stampeding events (Oakley et al., 2012) could be to blame for the high mortality in young walruses, as walruses had to crowd living on the shores of Alaska and Russia with less sea ice available. They also found that longer swimming distance could be blame for the decrease in the survival of polar bears (Oakley et al., 2012), with less sea ice available as well. Loss of hunting habitats could also cause the decline in polar bear's population (IPCC WG2, 2014). An earlier study (Stirling and Parkinson, 2006) found that the decline could be substantially. 

Source: USGS

Sea ice loss has also challenged the local biodiversity. One problem is hybrid. A hybrid of a grizzly and a polar bear was hunted in the Arctic in 2006 (Kelly et al., 2010). The explanation could be that polar bears live in the same area as grizzlies because of the loss in their habitats, sea ice.


 Human activities

Though ice melting leads to many bad impacts, there is a good news for human. Nearly ice-free status can provide open Arctic ocean. Ships may be able to navigate through Arctic ocean within this century, saving lots of time. Melia et al. (2016) simulated routes through Arctic, and concluded that shipping from Europe to Asia through the Arctic could become more than 10 days faster, and could save 4 days from N. America to Asia.

Ice melting and the United Nations

Earlier this month (7-18th), COP 22, CMP 12 and CMA 1 to the UNFCCC were held in Bab Ighli, Marrakech, Morocco. Taking this opportunity, I really want to share the two major works relevant for ice melting done by the United Nations (UN) last year, since I had an internship at United Nations Environment Programme (UNEP) China office this summer.

Sustainable Development Goals (SDGs)

At the end of September, 2015, 17 Sustainable Development Goals (SDGs) were adopted during General Assembly of the United Nations, in which the Goal 13 was defined as “Goal 13. Take urgent action to combat climate change and its impacts*”. 5 targets were set up in order to achieve the goal. Mitigating ice melting was a key motivation to set up this goal, as sea ice extent at Arctic has decreased over the last several decades, as well as ice will continually melt within this century as a result of global warming.

 Paris Agreement

Paris Agreement aims to arrange how to combat climate change globally after 2020. The agreement was adopted on COP21 in Paris on 12 December 2015, and was opened to sign on 22 April 2016 in New York. 16 days ago (4th Nov. 2016), it went to effect. Up till writing this sentence, 122 parties have ratified this agreement, and more parties will ratify sooner. This agreement is a milestone in combating global warming globally. I think it has been the most efficient action so far against climate change, because it clarifies the obligatory goals all countries should work together to achieve. This time, countries are going to be responsible for global benefits, instead of only considering their owns.

 One obligatory goal is to control the global average temperature warming to well below 2 °C, and to pursue efforts to hold the increase in temperature to to 1.5 °C, with both compared with pre-industrial levels. Ice melting would be controlled, if this goal could be achieved by the end of this century.

Sea ice melting at Arctic Sea 3: Evaluation -- can we trust the future projections?

Whether we can trust the projections is based on how well the projections can match observations and represent what will likely happen in the future. Therefore, the models need to be evaluated, via comparing model simulations with observations. Model evaluation is a key component in my academic background (Environmental Modelling), and I may write some knowledge beyond those related to sea ice below. To begin with, I adapt two figures to demonstrate the quality of model simulations. 

Figure 1. Comparison between observed seasonal cycle and modelled seasonal cycle. (source: Stroeve et al., 2012). 

The first one (Figure 1) shows seasonal cycle of sea ice extent at Arctic sea from model simulations and observations 1979-2011 (Stroeve et al., 2012). The multi-model means, particularly the CMIP5’s (diamonds), quite match the observations (red line). All the diamonds are placed between the maximum and minimum observations of each month. IPCC (2013) stated that the error in multi-model mean is less than 10% of the observations. The quality of model simulations have been improved. 

Figure 2. Modelled and observed sea ice extent 1900-2012. Each colour line represents a singe simulation from an individual model. Black lines represent observations. Red line shows multi-model mean from CMIP5, and blue line shows that from CMIP3.  Red shade shows simulations range from CMIP5, and blue shade shows that from CMIP3. (source: IPCC,2013).

From Figure 2, it is obvious that the model simulations from CMIP5 is better than those from CMIP3’s, because the multi-model mean of September ice extent from CMIP5 better matches the value of observations as well as catches up the sharp decline in the last few decades. The improvements were achieved by improving parameters used in modelling sea ice and improvements in other environmental components affecting sea ice. For instance, according to Stroeve et al (2012), one contributor of improvement from CMIP3 to CMIP5 is the improvement in parameters of sea ice albedo. The improvement can also be contributed to the improvement in simulating atmosphere (Notz et al., 2013), as sea ice is formed by the interactions between the atmosphere and ocean.

With the comparison between model simulations and observations, can we now trust the projections? We still need to consider uncertainties. There are three major types of uncertainty, which are internal variability, model uncertainty and scenario uncertainty. Their relative proportions are various with different spatial and temporal resolution.Natural fluctuations, coming from when any radiative forcing is absent, causes internal variability. The internal variability is a key reason for Notz et al. (2013) to against comparing model simulations with the observations directly, as their study found that the internal variability can cause a range of trends in model realisation. Model uncertainty is the difference among the different models’ simulations when responding to the same radiative forcing. This is why models in CMIP5 provide different simulations (covering a wide range) in sea ice extent and its relationship with annual global surface warming (for details, see last post). The last one, scenario uncertainty, is the difference in predicting greenhouse gases emissions in the future, and this be used to explain why Arctic sea would reach nearly ice-free at different time under different RCPs. I found a nice figure to show the relationship among the uncertainties, but it is not about sea ice (Figure 3).

Figure 3. Relative relationship among the three types of uncertainty. (source: Hawkins and Sutton, 2009).

Sea ice melting at Arctic Sea 2: Future projections

Last time, we looked at the present decline in Arctic sea ice. The decline rate has become faster and faster, and this is a very serious issue in climate change. It is necessary to find out how sea ice would change in the future. This is dominantly relied on future projections, which are based on model simulations under different scenarios. This time, we are going to look at model simulations and future projections about Arctic sea ice. 

At the beginning, let’s clarify some definitions and knowledge that’ll  help you to understand what I am going to write below. If you've already known these, please skip this part. (Since I'm studying modelling now and my undergraduate dissertation was related to modelling and projections, I list some extra academic journal articles here which I think are worth reading if you are interested in.)

• What is a model? And what’s a climate model?

 In one sentence, a model is a representation of the reality, helping to predict or understand something. It is a series of rules and principles to convert inputs (the data we have now) to outputs (what we want). 

A climate model uses quantitative methods to investigate how the climate system responses to variety of forcing, to predict how the future climate would be with different time scales and to make projections of future climate towards 2100 and beyond.

• What are CMIP3 and CMIP5?

CMIP3 and CMIP5 are the two phases of the Coupled Model Intercomparison Project (CMIP) for evaluation in Assessment Report  by Intergovernmental Panel on Climate Change (IPCC). IPCC reports summarise most advanced outstanding studies and researches in the world related to climate change. 

I think you just need to know that CMIP contains a series of climate models that simulate past and future climate, and the future projections are based on multi-model mean and variations among model simulations. That’s enough to understand what I am going to write below. If you want to know more about it, there is an excellent journal article providing the scientific explanation of CMIP5, i.e. Taylor et al. (2012).

•  What are RCPs?

RCP is the Representative Concentration Pathway, and there are four types of it : RCP2.6, RCP4.5, RCP6.0 and RCP8.5. You can simply think that the numbers at the end represent how much the radiative forcing would increase by the end of this century in W/m². e.g. in RCP8.5, radiative forcing reaches 8.5  W/m² by 2100, have highest temperature warming among the four. If you are interested in this, you can look at van Vuuren et al. (2011) for further information.

Okay, now let's go back to today’s topic. CMIP5 models under RCPs provide most comprehensive and advanced projections in the world, and my evidence and examples below are derived from CMIP5 model simulations in IPCC AR5 Chapter 12. The projections can be divided into near-term and long-term time scales. The near-term covers time from now to the mid of this century. IPCC (2013) stated that the near-term projections are not specific and precise enough as the result of changes in external forcing, so I’m not going to look at the near-term projections here.

I’m going to focus on the long-term (towards the end of this century) projections instead. Firstly, we have to determine the basis of model simalations. Sea ice melting in the future is very likely caused dominantly by further rise in surface temperature (IPCC, 2013). Figure 1 shows the relationship between decline in sea ice extent and annual global surface warming in CMIP5. Within the figure, it is obvious that there is a functional relationship (nearly linear) between sea ice extent decline and surface temperature rise before reaching nearly ice-free status (the black horizontal line at 1 × 106 km2  ). Nearly ice-free is that sea ice extent is continuously less than 1 × 106 km2  for no less than 5 years. 

Figure 1. Relationship between annual mean global surface warming and September Arctic sea ice extent relative to the period 1986-2005 as simulated by CMIP5 models. (Source: IPCC, 2013)

Then, let’s look at the future projections. Based on CMIP5, the decline in sea ice shows in all four RCPs but with different rates. According to Figure 2, a general decline in sea ice extent is simulated under each RCP, both for winter (Feb.) and summer (Sep.). I’m more interested in the changes in September and will focus on this in the following sentences, because the condition in September may change to nearly ice-free within this century. I’ve added a green horizontal line in the September one to show the nearly ice-free status. All four RCPs show the possibility of the nearly ice-free before 2100. Under RCP8.5, some model simulations show that the Arctic sea may reach nearly ice-free before 2040. By the end of this century, nearly all models (around 90% according to IPCC) under RCP8.5 reach the nearly ice-free status, while that is about half under RCP2.6 (around 45% according to IPCC). This means that if climate change keeps happening in the future, sea ice in Arctic sea will keep melting and the sea will become nearly ice-free within few decades. 

Figure 2. Changes in ice extent for period 1950-2100 as simulated by CMIP5 models under different scenarios. (Source: IPCC, 2013).

Additional information about observations of Arctic sea ice

In the last post, I forgot to put information about the sources of observations.

With year 1979 as the turning point, the observation periods can be split into two periods: pre-satellite and satellite. 

• Pre-satellite period

The pre-satellite data were based on terrestrial proxies regional offshore and aerial observations, etc. (IPCC, 2013). These data were the direct in-situ measurements collected on short time scales, difficult to measure accurately due to changes in edge and enormous size.

• Satellite period

Since 1979, satellite data have been available via microwave scanning, i.e. passive microwave remote sensing. This allows frequent collection of data on long time series available to global scale. Two major databases are operated by National Aeronautics and Space Administration (NASA) and the National Snow and Ice Data Center (NSIDC). 

 Instruments are :

• The Scanning Multichannel Microwave Radiometer (SMMR)
•  The Special Sensor Microwave/Imager (SSM/I)
• The Advanced Microwave Scanning Radiometer - Earth Observing System (AMSR-E)
• The Advanced Microwave Scanning Radiometer 2 (AMSR2)

Sea ice melting at Arctic Sea 1: Current declines

  

The video posted last time clearly demonstrates melting of Arctic glaciers and Greenland ice sheet. Still at Arctic, this time I want to write something about what is happening at Arctic sea, i.e. sea ice melting. This will be a series, and I want to use this series to systematically introduce ice melting problem. 

At the beginning, just to provide some basic information as the background. 

• Normally, Arctic sea ice expands to the peripheral seas in winter, and stays within the Arctic Ocean basin and the Canadian Arctic Archipelago in summer (IPCC, 2013). 

• Change in sea ice is measures by sea ice extent, which is the area of sea with no less than 15% is ice. Currently, the extent covers the area between 6× 10⁶ km²  and 15× 10⁶ km² , reducing to its minimum in February and expanding to the maximum in September (IPCC, 2013). Figure 1 (Comiso et al., 2008) shows this seasonal change of ice extent, as well as a decline of ice extent. 

• Current situation

Arctic sea ice extent has declined in the last few decades, and lots of evidence indicate this change. Comiso et al. (2008), as showed in Figure 2, found that average decline rate between 1979 and 2007 should be -3.7% per decade (black line). Another article (Comiso and Nishio, 2008) indicated that up to 2006, the decline rate was about -3.4  ±  0.2% per decade.The two decline rates are consistent with the range in IPCC AR5 (2013), in which concluded that the of decline rate should be within range of -3.5% to -4.1% per decade during 1979-2012. The decline rate, unfortunately, become faster and faster. In Figure 2, the decline rate had accelerated from -2.2 % per decade (1979-1996, green line) to -10.7% per decade (1978-2007, blue line).

Figure 2. Monthly ice extent anomalies, Nov.1978-Sep.2007. (Source:Comiso et al., 2008)

Figure 3 (IPCC, 2013) shows the decline in ice extent from more aspects. According to it, multi-year ice coverage and ice thickness also decreased, while drift speed and annual melting time increased. Rising in drift speed infers that ice melting increased within the region, as the acceleration was more likely caused by the weaker and thinner in ice (Spreen et al., 2011). Increase in length of melt season means that more ice melt while less ice formed within a year, therefore, causing net increase in ice melting.

The melting problem is more serious to multi-year and perennial ice.  The perennial ice, the minimum ice remained in summer, declined with the rate of -9% per decade, or quicker with -12.2% per decade. There was a declining rate at about -15.1% per decade in the multi-year ice (forms and survives for at least two years).IPCC (2013) indicated a decreasing rate of the perennial ice at -9.4% to -13.6% per decade, while the multi-year ice melted even faster with rate within range of -11.0 to -16.0% per decade.  

Figure 3. Annual ice extent, multiyear ice coverage, ice thickness, sea ice drift speed and average length of melt season , and their linear trends with decadal scale. (Source: IPCC, 2013)