For several decades wild ungulate densities are increasing, e.g. in North America and Europe. These high densities suppress woodland recruitment, with major socioeconomic and ecological consequences. Due to this increase in density, ungulates are expanding into more productive habitats. While, most studies have focused on woodland recruitment under high density in unproductive habitats, little is known about this in productive habitats. Productivity may have two contrasting effects on woodland recruitment 1) increase browsing tolerance and 2) increase ungulate densities. Here we experimentally test if tree recruitment is possible under high ungulate densities using red deer and feral horses, and if so, under which conditions. The experiment is performed in The Netherlands using a productive nature reserve where ungulate densities are bottom-up regulated. We transplanted over 4000 saplings across ten plots in two different vegetation types. Five plots were located in a tall unpalatable sward (winter grazed area) and five in a short palatable lawn (summer grazed area). Each plot contained saplings of two pioneer, two defended and two forest species. Within each plot saplings of each species were transplanted in and outside an exclosure.
Results/Conclusions
After one year, none of 1736 saplings outside the exclosure survived. Inside the exclosure, 1160 out of 2616 saplings (44%) survived. Furthermore we find that inside the exclosure all species have a greater chance of survival in the short lawn (on average 1.8 times higher) than in the tall sward. This indicates the importance of light competition in productive habitats if ungulates are removed. Consequently, in productive habitats after a potential ungulate population crash, the resultant woodland recruitment will most likely occur in the short palatable lawn (summer grazed area). This study shows that woodland recruitment is impossible in productive habitats under bottom-up regulated ungulate density, despite differences in sapling traits and vegetation types. Therefore, relatively lower densities are needed for the system to shift from open grassland to a (partial) woodland.