Wild ungulates as forest engineers

Earth’s surface is changing rapidly due to anthropogenic disturbance and thus threatening the conservation of entire biomes: from deserts to tropical rainforests and from oceans to mountains. Within this spectrum, forests are one of the biomes which has suffered the most due to the exploitation of natural resources and the expansion of the urban environment into natural landscapes. Specially the case of temperate forests which offer favourable conditions for a variety of economic activities, such as: food and timber production but as well for a blooming human life. The high concentration of humans in temperate forests has led to high indices of land conversion and fragmentation, threatening the conservation of the unique biodiversity harboured in these forests. For example, the loss of apex predators has led to the overabundance of wild ungulates in most of the temperate forests, increasing the top-down control that these animals have over vegetation. The top-down control of wild ungulates can influence several forest attributes such as forest diversity, structure and functioning. However, these effects are highly dependent on ungulate density, successional forest stage and the ungulate assemblage composition. Yet, there is a limited amount of field and experimental studies that elaborate on these important factors, mainly because there is a lack of long-term exclosures and a lack of studies that have evaluated the effect of a complete ungulate density gradient on forests in the same region. This thesis aims to evaluate the effects of wild ungulates on temperate forests at different scales. That is, how do wild ungulates affect forests at different 1) spatial scales (from areas with low to areas with high ungulate abundance)?, 2) temporal scales (from short-term to long-term effects)?, and 3) hierarchical scales (from trophic levels of primary producers to detritivores)? To tackle these questions, robust methods were designed in this thesis, such as a chronosequence of forest succession in presence and absence of ungulates, camera-traps paired to vegetation plots across a landscape, a global literature review on the effects of ungulate density on vegetation and litter bags placed inside and outside ungulate exclosures. Next, a description is provided for each of the chapters on their research and findings. To evaluate how ungulates affect different temperate forest attributes, a global quantitative literature review was conducted in chapter two. This meta-analysis provided a better understanding on how ungulate density interacts with forests at a global scale and identifies critical thresholds in ungulate density and tipping points in forests when the effect of ungulate density switches from neutral to negative in forests. From a total of 164 studies, ungulate density averaged 23.6 km-2 across studies. Ungulate density had negative effects on forest regeneration, structure and functioning in 70% of the evaluated cases. The dose-response curves had a sigmoidal shape. Critical tipping points, where ungulates started to have a negative effect on forest regeneration, were found at an ungulate metabolic weight density of 115 kg km-2 for forest regeneration, 141 kg km-2 for forest structure, and 251 kg km-2 for forest functioning, which is roughly equivalent to 10, 13 and 23 roe deer km-2. Forest regeneration was most sensitive to immediate browsing and trampling impacts of small seedlings, while forest functioning was least sensitive because of time lags. Yet, these effects may build-up over time. Chapter three on the short-term evaluated the shape of the dose-response for ten sites across at the Veluwe, the Netherlands, using approx. 210 camera points paired with vegetation plots. Five of the eleven forest variables measured were related to deer utilization level. With increasing red deer utilization, there was a decrease in litter depth. With increasing fallow deer utilization, there was a decrease in sapling richness. With increasing roe deer utilization, there was a decrease in sapling richness and diversity and shrub cover and an increase of sapling stem density. The dose-response curve between deer utilization and different forest attributes followed a curvilinear response, large changes at low followed by small changes at high utilization level, yet the exact shape of the curve can vary according to (a)biotic factors from each study location. Considering that the slopes of the responses were quite slight, it is possible to conclude that the influence that deer have on temperate forest structure and diversity is limited. Yet, these relationships may be different in a long-term study. To provide a better understanding of how forest succession proceeds in a situation of chronic browsing and trampling, chapter four evaluated the long-term effect of ungulates on temperate forests. A chronosequence approach was used, in which 17 paired fenced and unfenced plots were compared, ranging in age from 1 to 33 years since their establishment at the Veluwe. In fenced plots, where ungulates were excluded, there was a reduced understory vegetation cover and an increased canopy cover, tree species richness, tree Shannon diversity and litter layer compared to unfenced plots. In fenced compared to unfenced plots, woody vegetation developed with palatable broadleaved species such as Betula pendula, Betula pubescens, Prunus serotina, and Quercus robur. These results suggest that current ungulate densities in this system have pronounced long-term effects on forest structure, composition and litter depth, implying that ungulates can slow down natural succession of temperate forests, from light demanding to shade tolerant tree species. Yet, it remains unclear whether these long-term effects on vegetation trigger cascading effects on lower trophic levels. In chapter five, the potential cascading effects promoted by ungulates on temperate forests were evaluated. Using a network of twelve paired fenced and unfenced plots established across the Veluwe, soil quality, litter depth, forest regeneration, soil macro-invertebrates and decomposition rates were evaluated. Plots with ungulates had significantly higher soil compaction, but lower litter depth, tree diversity and density, rodent activity, invertebrate biomass and litter decomposition rate compared to plots with ungulates. Furthermore, ungulates presence: decreased sapling density which indirectly decreased rodent activity; decreased litter depth which indirectly reduced macro-invertebrate diversity and mineralization rates of magnesium and nitrogen in pine and oak litter, and increased soil compaction which decreased invertebrate diversity and mineralization rates of magnesium and nitrogen in pine and oak litter. If the system continues in the same direction, in the future, it will no longer sustain a diverse plant community due to nutrient limitation in soil. In chapter six, the main outcomes of each of the chapters were discussed and contextualized in a broader perspective of temperate forest ecology and plant-animal interactions. Although the literature review (Ch. 2) and the long-term effect chapter (Ch. 4) reported a strong top-down relationship between ungulates and different forest attributes, the short-term effect chapter (Ch. 3) and the cascading effect chapter (Ch. 5) only reported slight relationships. These results illustrate that current ungulate densities have a gradual shaping effect on forest composition, structure and functioning of these Dutch temperate forests; although, ungulates have a smaller effect on forest functioning than (a)biotic factors like soil pH, light availability and forest type (Ch. 3 and 4). Coming back to the main research question, ungulates are important disturbance agents that create, maintain or arrest forest dynamics. According to this thesis, ungulates mostly maintain forests properties, although they also arrest forest succession and in few cases, create opportunities for the establishment of new species. Spatially, forest dose-response to ungulate abundance is curvilinear, with large changes at low and small changes at high abundance. Temporally, ungulates have steeper relationships with forest vegetation at long-term scale compared to short-term because effects on vegetation accumulate in time. Hierarchically, ungulates cascade-down their effects to other trophic levels and ecosystem processes. Ungulate relationships with forests are difficult to understand due to the system’s complexity; yet, when their interactions occur at the extreme of the scales, their effects on forests can be irreversible.