Search

Aim: The amount of forest edges is increasing globally due to forest fragmentation and land-use changes. However, edge effects on the soil seed bank of temperate forests are still poorly understood. Here, we assessed edge effects at contrasting spatial scales across Europe and quantified the extent to which edges can preserve the seeds of forest specialist plants. Location: Temperate European deciduous forests along a 2,300-km latitudinal gradient. Time period: 2018–2021. Major taxa studied: Vascular plants. Methods: Through a greenhouse germination experiment, we studied how edge effects alter the density, diversity, composition and functionality of forest soil seed banks in 90 plots along different latitudes, elevations and forest management types. We also assessed which environmental conditions drive the seed bank responses at the forest edge versus interior and looked at the relationship between the seed bank and the herb layer species richness. Results: Overall, 10,108 seedlings of 250 species emerged from the soil seed bank. Seed density and species richness of generalists (species not only associated with forests) were higher at edges compared to interiors, with a negative influence of C : N ratio and litter quality. Conversely, forest specialist species richness did not decline from the interior to the edge. Also, edges were compositionally, but not functionally, different from interiors. The correlation between the seed bank and the herb layer species richness was positive and affected by microclimate. Main conclusions: Our results underpin how edge effects shape species diversity and composition of soil seed banks in ancient forests, especially increasing the proportion of generalist species and thus potentially favouring a shift in community composition. However, the presence of many forest specialists suggests that soil seed banks still play a key role in understorey species persistence and could support the resilience of our fragmented forests.

A change in land use from agriculture to forest generally increases soil acidity. However, it remains unclear to what extent plant traits can enhance or mitigate soil acidification caused by atmospheric deposition. Soil acidification is detrimental for the survival of many species. An in-depth understanding of tree species-specific effects on soil acidification is therefore crucial, particularly in view of the predicted global increases in acidifying nitrogen (N) deposition. Here, we report soil acidification rates in a chronosequence of broadleaved deciduous forests planted on former arable land in Belgium. This region receives one of the highest loads of potentially acidifying atmospheric deposition in Europe, which allowed us to study a ‘worst case scenario’. We show that less than four decades of forest development caused significant soil acidification. Atmospheric deposition undoubtedly and unequivocally drives postagricultural forests towards more acidic conditions, but the rate of soil acidification is also determined by the tree species-specific leaf litter quality and litter decomposition rates. We propose that the intrinsic differences in leaf litter quality among tree species create fundamentally different nutrient cycles within the ecosystem, both directly through the chemical composition of the litter and indirectly through its effects on the size and composition of earthworm communities. Poor leaf litter quality contributes to the absence of a burrowing earthworm community, which retards leaf litter decomposition and, consequently, results in forest-floor build-up and soil acidification. Also nutrient uptake and N2 fixation are causing soil acidification, but were found to be less important. Our results highlight the fact that tree species-specific traits significantly influence the magnitude of human pollution-induced soil acidification.

Fertilisation of agricultural land causes an accumulation of nutrients in the top soil layer, among which phosphorus (P) is particularly persistent. Changing land use from farmland to forest affects soil properties, but changes in P pools have rarely been studied despite their importance to forest ecosystem development. Here, we describe the redistributions of the P pools in a four-decadal chronosequence of post-agricultural common oak (Quercus robur L.) forests in Belgium and Denmark. The aim was to assess whether forest age causes a repartitioning of P throughout the various soil P pools (labile P, slowly cycling P and occluded P); in particular, we addressed the timerelated alterations in the inorganic versus organic P fractions. In less than 40 years of oak forest development, significant redistributions have occurred between different P fractions. While both the labile and the slowly cycling inorganic P fractions significantly decreased with forest age, the organic fractions significantly increased. The labile P pool (inorganic ? organic), which is considered to be the pool of P most likely to contribute to plant-available P, significantly decreased with forest age (from[20 to\10% of total P), except in the 0–5 cm of topsoil, where labile P remained persistently high. The shift from inorganic to organic P and the shifts between the different inorganic P fractions are driven by biological processes and also by physicochemical changes related to forest development. It is concluded that the organic labile P fraction, which is readily mineralisable, should be taken into account when studying the bioavailable P pool in forest ecosystems.

A change in land use from agriculture to forest generally increases soil acidity. However, it remains unclear to what extent plant traits can enhance or mitigate soil acidification caused by atmospheric deposition. Soil acidification is detrimental for the survival of many species. An in-depth understanding of tree species-specific effects on soil acidification is therefore crucial, particularly in view of the predicted global increases in acidifying nitrogen (N) deposition. Here, we report soil acidification rates in a chronosequence of broadleaved deciduous forests planted on former arable land in Belgium. This region receives one of the highest loads of potentially acidifying atmospheric deposition in Europe, which allowed us to study a ‘worst case scenario’. We show that less than four decades of forest development caused significant soil acidification. Atmospheric deposition undoubtedly and unequivocally drives postagricultural forests towards more acidic conditions, but the rate of soil acidification is also determined by the tree species-specific leaf litter quality and litter decomposition rates. We propose that the intrinsic differences in leaf litter quality among tree species create fundamentally different nutrient cycles within the ecosystem, both directly through the chemical composition of the litter and indirectly through its effects on the size and composition of earthworm communities. Poor leaf litter quality contributes to the absence of a burrowing earthworm community, which retards leaf litter decomposition and, consequently, results in forest-floor build-up and soil acidification. Also nutrient uptake and N2 fixation are causing soil acidification, but were found to be less important. Our results highlight the fact that tree species-specific traits significantly influence the magnitude of human pollution-induced soil acidification.

Fertilisation of agricultural land causes an accumulation of nutrients in the top soil layer, among which phosphorus (P) is particularly persistent. Changing land use from farmland to forest affects soil properties, but changes in P pools have rarely been studied despite their importance to forest ecosystem development. Here, we describe the redistributions of the P pools in a four-decadal chronosequence of post-agricultural common oak (Quercus robur L.) forests in Belgium and Denmark. The aim was to assess whether forest age causes a repartitioning of P throughout the various soil P pools (labile P, slowly cycling P and occluded P); in particular, we addressed the timerelated alterations in the inorganic versus organic P fractions. In less than 40 years of oak forest development, significant redistributions have occurred between different P fractions. While both the labile and the slowly cycling inorganic P fractions significantly decreased with forest age, the organic fractions significantly increased. The labile P pool (inorganic ? organic), which is considered to be the pool of P most likely to contribute to plant-available P, significantly decreased with forest age (from[20 to\10% of total P), except in the 0–5 cm of topsoil, where labile P remained persistently high. The shift from inorganic to organic P and the shifts between the different inorganic P fractions are driven by biological processes and also by physicochemical changes related to forest development. It is concluded that the organic labile P fraction, which is readily mineralisable, should be taken into account when studying the bioavailable P pool in forest ecosystems.