Wrapping up (1)

Unfortunately, it's almost time for this blog to come to an end, so for the last couple of posts I'll try and sum up what I've learned over the course of the past couple of months. At the beginning I posted a few questions that I wanted to answer, so I think these are a good place to start, so I'll use these as a structure for the next couple of posts.

How do we know the sea ice retreat in the past few years is unusual?

The blog started off by looking at this question, firstly by using satellite data. Since satellite data began in 1979, sea ice extent and thickness have been decreasing at an alarming rate. Sea-ice extent during the summer of 2011 (sept) was almost on par with the previous record low during 2007, even without the unusual weather patterns thought responsible for the 2007 low. This strongly suggests that even though sea ice doesn't decrease every winter, overall conditions are worsening for continued survival of permanent sea ice.

A number of proxy records show a sea-ice history going further back than any contemporary records. As part of the blog, I looked at two relatively new methods of constructing these histories; the IP25 biomarker and Quartz records. These, as well as other methods, including documentary and ice-diatom records, show that extent of sea ice has been strongly variable, but also strongly correlated with climate. Sea ice extent is therefore extremely sensitive to temperature changes, meaning if we were experiencing global warming today, we could expect a significant loss of sea ice. Is that what we've been seeing?

Well, the sea-ice is disappearing at a faster rate than any model used by the IPCC has forecast, even the pessimistic 'business as usual' scenarios shown on the graph below. The graph doesn't show the past 5 years, but if it did the trend would be continuing, with 2007 being the minimum.

Sea ice extent output values from 13 IPCC Global climate models with an ensemble mean in thick black.
Actual observations are shown as a thick red line. Direct from Stroeve et al 2007.

Of course, this means a few things could be happening. Firstly, there could be some unknown natural variability that

Arctic Lakes and Zoological Indicators

As we saw in the last post, it is highly likely that climate is indirectly altering Diatom communities in high Arctic lakes through changing levels and patterns of ice cover. Quinlan et al. (2005) used the same cores as the (previously reviewed) Douglas et al. (1994) study and counted the fossil chironomid (non-biting midge) larvae remains down the cores to investigate whether invertebrate communities have also 'shifted', and if they have, how likely it is that the changes are linked to the results from the previous diatom investigations. It's important to note, however, that these results are only taken from three archives (ie lakes and ponds) from Ellesmere island so further study is required to confirm that these findings are widespread.

Chironomidae are a family of flies which are aquatic organisms until their adult stages (see above) and

Arctic Lakes and Diatom Change - A Brief Summary

As promised, I'm going to cover the Smol et al (2005) article this time, as well as some associated references which will form part of the discussion. This post will focus on the Diatom changes seen in a wide range of Arctic lakes and their inferred causation. It's generally accepted that the Arctic regions will experience the effects of climate change disproportionately to other regions. This, coupled with the sensitivity of Arctic lakes, (potentially) makes this form of study an early indicator of the speed and severity of climate change in the future on a global scale. (Oops, I may have partly answered the final question). Oh well, on with the post...

A subarctic lake from Northern Quebec

Arctic Lake ecology is nice summarised by Smol and Douglas (2007) and shows that

Where’s the evidence?

A key question to ask before discussing potential impacts of human activity upon Arctic ecosystems is:

How do we know that the recent warming trend in the Arctic is unusual when compared to long-term natural cycles?

As instrumental records are limited both in spatial and temporal extent (earliest records go back only a couple of centuries) they are not adequate to answer this question. The answer comes in the form of palaeo-records; naturally stored records of past environmental conditions. Before I plunge into reviewing scientific papers, I thought it would be a good idea to review some key concepts which we can refer back to later in the blog. As we'll see, the answer to 'where's the evidence?' is: everywhere!

I’ve heard about ice cores and seen the hockey stick diagram, is this what you’re talking about?

In part, yes! Ice cores contain a number of proxies (indicators) of past conditions around the Arctic, and contribute to understanding of global interactions. Most key ice-core records from the Arctic are from Greenland including GISP/Grip and NGRIP. A common proxy used from ice-cores is the δ¹⁸O:δ¹⁶O ratio, a comparison between two stable forms of oxygen isotope found bonded with hydrogen in the water molecules (H₂O, I’m trying not to insult your intelligence!).

As δ¹⁸O is two protons heavier than δ¹⁶O, it requires more energy to evaporate and condensates more readily. Therefore, by measuring this ratio in ice cores, it is possible to reconstruct past temperatures as well as a history of moisture transport. Cold temperatures generally are indicated by a lower concentration of the heavier isotope δ¹⁸O. Oxygen isotope records are also found in other proxies including in tree-rings and in the chemical composition of marine organism shells. Other proxies included in ice cores are deuterium ratios (heavy hydrogen), greenhouse gas concentrations (care is required due to a certain amount of gas exchange) as well as many others.

So, that’s it?

Fortunately (or I suppose unfortunately depending on your viewpoint) not, there’s a couple more key records used to reconstruct Arctic environments which will crop up a lot in this blog, proxies found in lake sediments and marine records as well as tree rings.

How can lake sediments show past environment?

Common to most palaeo-records is a build-up of material over time, as happens with lake sediments. All manner of organisms and materials build up with this sediment, and these themselves are the proxies. A common proxy is diatom frustules, siliceous shells of microscopic algae of which there are thousands of species. These all respond to different conditions favourably, so by recording the species community composition, it is possible to reconstruct the past conditions qualitatively (descriptively) or quantitatively (through the use of transfer functions). Diatoms can be identified relatively easily under a light microscope, if you know what you’re doing that is (as a side point, if you don’t, as I’ve experienced, the number of species can be quite overwhelming).

Other proxies include chironomidae (non-biting midge species), fossil pollen and marine ostracod records (bivalve crustaceans). Also, a recently developed but useful record is SCP (spheroidal carbonaceous particles) which are produced from burning. They indicate atmospheric contamination when found in lake sediments.

From these records and knowledge of the conditions individual species prefer today, it is possible to reconstruct many aspects of the environment including temperatures, pH, Total Phosporous (TP) and ice cover as well as many others.

What about tree rings?

Everyone knows you can count tree rings to find the age of a tree, as they build up in layers over time. Due to this quality of a ring layer growing during one year, tree-ring proxies have the useful feature of being annually resolved; having an environmental record for each individual year. The longest complete records extend past 10,000 years before 1950 (10Kyr BP) include fossilised tree remains, giving an almost complete record of the Holocene (they also allow the calibration of radiocarbon dates to calendar years as described here). Analysis of tree rings is called ‘dendrochronology’.

Fossilised tree records contain much information, including the dating of extreme events as well as climate records. Trees destroyed due to an earthquake, for example, can show the exact date of that event if patterns are widespread. Climatic information is revealed by the widths of tree rings themselves; narrow rings indicate drought or cold years. If these patterns are seen across a wide area, a regional climatic pattern may be inferred to a very high resolution.

Thanks! You’ve made everything absolutely clear and I have no further questions.

Great! (If, as is more likely, you have any issues with this post, please comment).

It's important to make the point at this stage that reconstructions from palaeo-records are not perfect and do not always agree. Natural systems are complicated and are not always fully understood, and the patterns observable today may not have always worked in the same way throughout all glacial-interglacial cycles. Therefore, assumptions are stretched for very old reconstructions. Multi-proxy studies are desirable and when possible will be preferentially referred to in this blog.

The next post will consider some key papers which have used these proxies, focussing on this article.

______

For a much better and thorough summary of all of these proxies see the book ‘Global Change in the Holocene’ by Mackay et al eds. (2005) published by Hodder Arnold 528pp.