Thursday 19 November 2020

Ancient Europe's landscape: grassland savannah or forest?

There is an on-going controversy on what Europe’s original landscape, untouched by human influence, was like: either heavily forested, a grassland savannah or a mix of open forests, park-like landscapes and grassland. 
The traditional view is that of Europe being a heavily forested continent. This view has been challenged in recent decades. It has been proposed that herbivores prevent open landscapes from becoming forested by damaging forest growth with their feeding, and are even able to turn forests into open landscapes this way. As a result, Europe’s original landscape would not have been one big forest but a mix of open landscapes, park-like landscapes and forests [1]. Some even claim Europe was a large grassland savannah. 
Indeed no large mammal in Europe is dependent on forests while they need at least some open landscape in their habitat [2,3]. Africa is used as an analogue, where large herbivores are claimed to create the open landscape we are familiar with. Especially elephants, which uproot trees are suspected to create open landscapes [3]. 
There is conflicting data, however. Studies suggest that the uprooting of trees by elephants does not diminish the forest but instead speeds up the forest rejuvenation [4]. And it is overlooked that not only the savannah but also the deep rainforests are home to large herbivores, such as the forest elephant (Loxodonta cyclotis), the forest buffalo (Syncerus nanus) and the okapi (Okapia johnstoni) [4]. There are also several species of large herbivores living in the rain forests of South-East Asia (banteng, gaur, Asian elephant). If large herbivores create open landscapes, elephants in particular, the forest elephant would be a contradiction in itself: a forest-adapted elephant (f.e. the smaller size) would not exist if elephants created open landscapes with their feeding and uprooting of trees. 
Furthermore, and more importantly, palynologic data suggests that Europe was densely forested until very recent millennia when man started agriculture in Europe [4]. Also the insect fauna shows that forest-dwelling species were very common in Europe until recently [4]. It is argued, however, that also an intensely grazed grassland area has the same palynologic signature as a closed forest (see for example the works of Frans Vera). But this apparently is only the case at a very high herbivore density [4], and it cannot explain why the data from insects suggest high forestation. 
Another argument against ancient Europe being a grassland savannah and for a strongly forested continent is the distribution of the European wild horse in the Holocene. The horse is heavily dependent on a grassy diet and is an open land animal. It got very rare in Europe after the last glacial [5]. It virtually disappeared from Central Europe [5]. When agriculture began, the equine remains increased again, possibly due to an increase in open landscapes [5]. These distribution patterns contradict the hypothesis that Europe would have been a grassland savannah due to grazing and instead supports the hypothesis of Europe without human influence being a mostly forested continent. 
 
Europe’s original landscape continues to be a subject of debate. The data that is available to me suggests that the idea of ancient Europe being a grassland savannah with large herds of horses is not the likeliest scenario. 
 
Literature 
 
[1] Bunzel-Drüke et al.: Der Einfluss von Großherbivoren auf die Naturlandschaft  Mitteleuropas. 2001. 
[2] Beutler: Die Großtierfauna Europas und ihr Einfluss auf Vegetation und Landschaft. 1996.
[3] Bunzel-Drüke et al.: Überlegungen zu Wald, Mensch und Megafauna. 1994.  
[4] van Vuure: Retracing the aurochs – history, morphology and ecology of an extinct wild ox. 2005. 
[5] Sommer et al.: Holocene survival of the wild horse in Europe – a matter of open landscape? 2010.
 

8 comments:

  1. It is worth noting that, while they do occupy forests, banteng, gaur, and other bovines do require the presence of open patches to graze in, as none of them are obligate browsers. Their presence would thus imply that at least some open patches exist. Certainly we know from bison that the carrying capacity of these herbivores is much higher in open habitats.

    ReplyDelete
  2. https://www.pnas.org/content/pnas/113/4/847.full.pdf

    ReplyDelete
  3. I think there is one important difference between european dicidious forests and tropical rain forests: the vegetation periode is much shorter in european forests.Especially in winter large herbivores can have great effects on the vegetation by removing bark.
    I don't think europe was a savannah like grassland but maybe there were open forests with much more ground vegetation than in closed forests.
    One problem is: humans never were absent in europe since the last ice age. So there may have been a heavy hunting pressure on large herbivores like the aurochs or the wisent so that these animals never reached densities high enough to clear up the forests.

    ReplyDelete
    Replies
    1. I would certainly agree with that last part.

      Delete
    2. Well said, do not underestimate the hunting pressure on large longlived grazers (etc) even by a small number of humanoids. Also the remains of prehistoric trees found almost always are branched rather low; these trees therefore must have grown in rather open situations.

      Delete
  4. forest elephants actually do create open places called bai's theese open places are a keystone biotope for many large forest mammals who get their salt, water and food from these places as well as gathering hhere. look them up they are common in the congo basin

    ReplyDelete
  5. europe was a various landscape with grasslands, forests and semi open landscapes of all types imaginable. from large plains to thick untouched forests. they were all visible. this is the most likely hypothesis seeing the field work done as well as the niches of europes wild life

    ReplyDelete
    Replies
    1. The palynologic and entomologic data doesn't seem to support this.

      Delete