Treeline advances along the Urals mountain range - driven by improved winter conditions?

High-altitude treelines are temperature-limited vegetation boundaries, but little quantitative evidence exists about the impact of climate change on treelines in untouched areas of Russia. Here, we estimated how forest-tundra ecotones have changed during the last century along the Ural mountains. In the South, North, Sub-Polar, and Polar Urals, we compared 450 historical and recent photographs and determined the ages of 11,100 trees along 16 altitudinal gradients. In these four regions, boundaries of open and closed forests (crown covers above 20% and 40%) expanded upwards by 4 to 8 m in altitude per decade. Results strongly suggest that snow was an important driver for these forest advances: (i) Winter precipitation has increased substantially throughout the Urals (~7 mm decade(-1) ), which corresponds to almost a doubling in the Polar Urals, while summer temperatures have only changed slightly (~0.05°C decade(-1) ). (ii) There was a positive correlation between canopy cover, snow height and soil temperatures, suggesting that an increasing canopy cover promotes snow accumulation and, hence, a more favorable microclimate. (iii) Tree age analysis showed that forest expansion mainly began around the year 1900 on concave wind-sheltered slopes with thick snow covers, while it started in the 1950s and 1970s on slopes with shallower snow covers. (iv) During the 20th century, dominant growth forms of trees have changed from multistemmed trees, resulting from harsh winter conditions, to single-stemmed trees. While 87%, 31%, and 93% of stems appearing before 1950 were from multistemmed trees in the South, North and Polar Urals, more than 95% of the younger trees had a single stem. Currently, there is a high density of seedlings and saplings in the forest-tundra ecotone, indicating that forest expansion is ongoing and that alpine tundra vegetation will disappear from most mountains of the South and North Urals where treeline is already close to the highest peaks.