Not So Ancient, Not So Forest: Europe's Open Secret

A scientific review challenges dense forest as the natural baseline for humid-temperate Europe, arguing that large herbivores helped sustain a shifting mix of woods and open ground. ^[3]

The researchers trace that pattern across the Neogene, from 23 million to 2.6 million years ago, and the Quaternary, from 2.6 million years ago to 1900. Their conclusion puts a practical question before conservationists: when a project plants trees, which version of the past is it trying to restore? ^[3]

Led by Szymon Czyżewski, the team brought together evidence from ancient pollen, plant remains, fossils and other records. They describe humid-temperate Europe as a changing patchwork of open land, lightly wooded areas and closed-canopy forest, with abundant wild herbivores likely helping to maintain the variety. ^[3]

The peer-reviewed article, published in Biological Conservation in April 2026, argues that uniform, dense woodland may differ from the conditions in which many European species evolved. The authors recommend giving habitat variety and the ecological role of large herbivores greater weight in restoration. ^[3]

Senior author Jens-Christian Svenning said current reforestation practices were "on the wrong track", arguing for a mixture of woodland and open habitat, including through the restoration of large herbivores. ^[1]

A landscape with room for large animals

The animals in this reconstruction include elephants, rhinoceroses, aurochs and bison, according to Aarhus University. Their feeding and disturbance of vegetation offered a way to keep some ground open alongside wooded patches. A forest landscape could contain large trees without having an unbroken roof of leaves. ^[1][4]

That mechanism makes the animals part of the explanation for the vegetation, rather than simply occupants of it. The authors argue that herbivores helped shape the conditions in which plants grew, keeping woodland and grassland together across long stretches of European ecological history. ^[3]

The review draws on several kinds of evidence because each preserves a different part of that history. Pollen and larger plant remains provide traces of vegetation. Fossils and measurements of animal teeth help reconstruct past animal communities and their ecology.

The team also considered isotope evidence, microscopic charcoal and ancient DNA preserved in the environment. ^[1][3]

A grain of pollen cannot, by itself, show whether trees stood in a dense block or among patches of grass. Bringing different records together is central to the authors' argument that the long-term pattern was a mosaic. Their conclusion rests on the convergence of those records. ^[3]

This is a synthesis of existing research, not a new excavation or field experiment. It covers humid-temperate Europe, not every European landscape, and offers neither a single continent-wide percentage of open ground nor a measured verdict on any present-day planting programme. ^[3]

The trees did not stand still

The reconstruction also describes major changes. After the collapse of large-animal populations in the Late Pleistocene, tree cover expanded as pressure from herbivores fell. From the Mesolithic onward, human burning and hunting increasingly shaped the vegetation. ^[3]

Domestic grazing animals later partly recreated the disturbance once supplied by wild herbivores, while agriculture brought other changes. In this account, a meadow maintained by livestock can preserve something of an older ecological process even though the animals and the people managing them are different. ^[3]

Czyżewski said traditional extensive grazing had disappeared from large parts of the landscape, with the most dramatic change occurring largely within the past hundred years. The loss of that grazing matters to the researchers' argument because it removes a force that had helped keep vegetation open. ^[1]

There is strong evidence for extensive forest at particular times, too. A separate study published in Scientific Reports in 2018 estimated that mid-latitude and northern Europe was about 73% forested around 6,000 years ago. It used more than 1,000 radiocarbon-dated site records to examine the last 11,000 years. ^[2]

That estimate belongs alongside the new review. The studies address different regions, periods and questions. A landscape can be heavily wooded overall and still contain open patches, while the review itself describes tree expansion after the decline of large herbivores. ^[2][3]

The useful historical baseline is therefore a landscape capable of changing. Treating one moment of high forest cover as the permanent natural condition would discard much of the history the review assembles.

What should restoration restore?

A separate study by Czyżewski and Svenning gives the debate a more immediate biological anchor. Their 2025 Nature Plants paper examined the habitat preferences of 917 native temperate-forest plant species across central and western Europe. Most favoured varied, semi-open canopy conditions rather than uniformly closed woodland. ^[5]

The term "forest plant" can therefore conceal an important preference: a species associated with woodland may still benefit from light reaching the ground. The plant study supplies a separate line of evidence for the value of varied canopy conditions, rather than proof of the entire ancient reconstruction. ^[5]

The restoration argument is persuasive on a specific point: tree cover alone is too narrow a measure of whether a project recreates habitat. If the aim is to support species associated with a mixture of shaded and open conditions, making every patch densely wooded is a poor definition of success.

That does not turn the review into a prescription to clear forests or release large animals everywhere. The authors' recommendation is to give greater weight to habitat mosaics and herbivores when deciding what restoration should achieve. ^[1][3]

For those designing planting schemes, the next practical question is what mixture of woodland and open ground will support the species they want to conserve. The researchers are asking them to make that decision before the trees go in. ^[1][3]