Melting Ice - Part 5 - Permafrost

And now for the final episode of my Melting Ice mini-series. Unfortunately I have no plans to reveal whodunit, no-one is getting married, and there isn't going to be a surprise twist at the end, or is there? No, seriously there won't be. This post is about permafrost, and yes, you've guessed it, the effect of changing permafrost on sea level. So no cliff hangers then, probably we won't get commissioned for a second series.

I don't know anything about permafrost (yet) but apparently this is it (CapitalOTC). Fiona has a really great blogpost about what happens when permafrost explodes (and a picture that is perhaps a little more exciting than this one...)

Permafrost is permanently frozen ground either beneath land or sea - to be classified as permafrost it has to have been below zero degrees Celsius for two years or more, so we're not talking a randomly cold winter then (IPA). In the Northern Hemisphere permafrost is found further north than 35°N, except for high altitude areas (Zhang et al, 2003). Turns out there's alot more permafrost than I imagined - according to Zhang et al (2003) 23.9% of land in the Northern Hemisphere is permafrost, this doesn't even include the areas covered by ice sheets and glaciers which would increase the proportion of frozen ground to over a quarter. These values include permanently frozen ground, seasonally frozen ground, and intermittently frozen ground. Different sorts of permafrost contain highly variable amounts of ice and therefore water, for instance bedrock in the tundra can hold very little (IPCC).


However, there is increasing evidence that permafrost is reacting to recent global warming (Lawrence et al, 2012). The active layer, the part which thaws with natural variation in the summer, is getting thicker (IPA), the areas within permafrost areas where the ground is not frozen (taliks) are increasing (IPCC), and the temperature of the permafrost itself is rising.  For instance Vonder-Muhll, (2001) measured the temperature of alpine permafrost between 1987 and 2000 using two boreholes 20km apart. He found a general warming trend with permafrost temperature rising by over a degree for the whole period. You can also see this warming trend in Alaskan permafrost since the 80s:


Here you can see that the temperature of permafrost in Alaska has been rising, temperature measurements were taken at depths of 20 metres so there would be no seasonal temperature effect (UNEP)

So what's all this got to do with sea level?

As permafrost thaws the ice frozen within it becomes water. While some of the water remains in the soil once it melts, some will also flow out into the oceans. If all the permafrost melted it could cause a sea level rise of 3 - 10 cm (NSIDC). If this sounds like a bit a vague value, that's because modelling permafrost decline is very difficult. Even Earth System Models (ESMs) which include more details of the climate system (such as biochemical processes and vegetation changes) than General Circulation Models cannot capture the complicated feedbacks that occur as permafrost starts to thaw (Lawrence et al, 2012). Many of these feedback processes are not even properly understood yet, for instance it is difficult to quantify how permafrost thaw will effect local hydrological conditions.  NCAR are trying to improve understanding and representation of of permafrost in models such as the Community Land Model.


The more I look into the causes of changing sea level, the more I realise that rising sea level is just a reaction to something much bigger. I'm reading (and writing) all this stuff about how melting ice will effect global mean sea level, but I'm ignoring the local changes in habitats brought on by melting ice. I think that it can be really easy to imagine that all these really cold areas are barren and inhospitable, but there is just so much life out there, from algae in the ice right up to penguins and polar bears. Right now it feels that by only looking at sea level I'm missing out a huge part of the story.