What we can learn about ecosystem collapse from Iceland

A large proportion of Iceland is either unvegetated completely or only very sparsely so. This is generally true of highland regions, but also some lowland regions where sand plains may be found.  These are called deserts: not in the sense of a dryland, but in the sense of a deserted place where the paucity of vegetation makes it exceptionally difficult to live. 

This could be the Atacama Desert or Mars. It's a highland desert of Iceland.

A view of Sandvatn, on the way to Langjökull glacier. Notice that little island of green by the lake...more on these islands below.

It is also curious that despite a mild climate for the latitude, the land does not support trees generally. Often, it seems that the dominant vegetation are mosses in rocky areas like lava flows, or grasses where there has been some soil formation. Maybe this isn't so strange in the highlands, where you might have expected more low-growing plants. But, this holds true even in areas closer to the coast which are well within the environmental tolerances of trees.

Your first instinct may be to look at the glaciers and ice sheets, the most extensive in Europe. You might think that after the last ice age, the land never really supported plants because of the recent glaciation. Or you might pin it on the several active volcanoes. And yet old writings, including the sagas, report a vegetated place. Another key to the past is in place names, for example many place names refer to a forest or woodland, yet if you go there there is no such thing present. There's also some hard physical evidence of more extensive vegetation. For example you can go to some nearly barren areas and find charcoal pits. The charcoal would have been made to fuel iron forges (settlers were iron age people), but where are the trees? Surely people didn't drag wood into some barren area just to make charcoal. A sensible person would produce the charcoal where the wood is, then transport the much lighter charcoal.

Then, most informative of all are the rofabörðs. Rofabörðs are actual remnants of vegetated land, usually in small patches with abrupt cliff-like boundaries, that border unvegetated land. I find it poetic that they are sometimes shaped like puzzle pieces. They are that final piece that allows you to solve the puzzle of Iceland's past. You only have to see one to guess the amazing truth: 1. Iceland had vegetation, 2. Now it's mostly gone, 3. The soil, sometimes alot of it, is now gone too. When you start piecing this together with some of the other lines of evidence,  you can further infer that: 4. people were a part of this change. 

A rofabörð at the bottom of a slope at Einifell.

The other side of the hill. Another rofabörð ends abruptly in a cliff-like edge of greater than 1 m.

You can see quite a bit digging into a rofabörð. First, they are deep! Second, they preserve a stratigraphic record of key events like volcanic eruptions. The long term presence of vegetation seems to associated with substantial soil development, and furthermore vegetated patches accumulate rather than lose soil over time, alot of it.

Oli Arnalds opening a soil pit into a nearby rofabörð. Note that the top is vegetated by heath including grasses and low growing shrubs.

We exposed over 1.5m of soil, and still did not reach a marker called the "settlement tephra layer", which is dated to the year 871. This patch must have gained all of that soil in the last 900 years or so. You can see a black tephra layer here, probably one of the more recent (a few centuries ago) Hekla eruptions. 

The scope of the ecosystem collapse is huge. Iceland's vegetation loss led to catastrophic erosion in about one third of the country. The ability of these severely eroded areas to support vegetation ever again is seriously compromised, and it doesn't happen very often without a serious shove from people (future post alert!)

Ok, why did this happen to Iceland? A perfect storm of multiple factors, here's my summary attempt:

1. From the beginning,  Iceland has some unique vulnerabilities. Volcanic ejecta (tephra: ash, cinders, etc.) can, has, and does sometimes bury low-growing vegetation, knocking back the productivity and coverage or snuffing it out completely. This has happened many times to greater or lesser degrees anywhere in the country. The record is captured in soil remnants with datable tephra layers that you can clearly see. Another more localized factor is that many meandering rivers emanating from the ice sheets can render large plains bare (e.g. Skeidarsandur), especially when there is an ice flood. Finally, the soils are andosols, derived from volcanic ash. Such soils may present some nutrient availability challenges, may poorly aggregate without vegetation and organic inputs, and may contain some unusually light mineral particles that can be easily transported by wind or water.

2. Over this fabric of vulnerability, settlers cleared the forests and woodlands to build and heat homes, conduct metal working, and make farmland. The Icelandic birch would have been the dominant species of the lowlands, growing in usually open canopies in small shrub to tree form. this is the only native timber. Such woodlands covered about one third of the country. The highland was likely vegetated with heath with low growing, spreading dwarf shrubs. This heath was not grazed by any native mammalian herbivores, there are none. So the introduction of sheep or goat grazing in the highlands was a new ecological pressure for this system and would further impact recruitment and porductivity of shrubs.

3. Without much tall vegetation anymore (very few trees, fewer shrubs), the former woodlands had lost their resistance to volcanic ash deposition. Some of these areas became deserts after ash fall smothered the rest of the vegetation.

4. Without vegetation, soil erosion is much more prevalent. All of this leads to permanent widespread soil loss, sometimes measurable in meters, and expansion of desert regions. Some sandy deserts also tend to encroach into adjacent areas, compromising vegetation more.

5. A last insult was the Little Ice Age, which was an unconducive period for plant growth, especially with all of these other pressures also occurring.

The prevailing story here is that multiple factors lead to the ecosystem collapse, but new anthropogenic pressures like wood harvest and grazing speeded, expanded and exacerbated natural disturbance regimes. This is difficult to reverse now, but a few ways have been developed by the oldest Soil Conservation Service in the world.