Differential effect of drought regimes on the seedling performance of six floodplain grassland species.

The performance of seedlings is crucial for the survival and persistence of plant populations. Although drought frequently occurs in floodplains and can cause seedling mortality, studies on the effects of drought on seedlings of floodplain grasslands are scarce. We tested the hypotheses that drought reduces aboveground biomass, total biomass, plant height, number of leaves, leaf area and specific leaf area (SLA), and increases root biomass and root-mass fraction (RMF) and that seedlings from species of wet floodplain grasslands are more affected by drought than species of dry grasslands. In a greenhouse study, we exposed seedlings of three confamilial pairs of species (Pimpinella saxifraga, Selinum carvifolia, Veronica teucrium, Veronica maritima, Sanguisorba minor, Sanguisorba officinalis) to increasing drought treatments. Within each plant family, one species is characteristic of wet and one of dry floodplain grasslands, confamilial in order to avoid phylogenetic bias of the results. In accordance with our hypotheses, drought conditions reduced aboveground biomass, total biomass, plant height, number of leaves and leaf area. Contrary to our hypotheses, drought conditions increased SLA and decreased root biomass and RMF of seedlings. Beyond the effects of the families, the results were species-specific (V. maritima being the most sensitive species) and habitat-specific. Species indicative of wet floodplain grasslands appear to be more sensitive to drought than species indicative of dry grasslands. Because of species- and habitat-specific responses to reduced water availability, future drought periods due to climate change may severely affect some species from dry and wet habitats, while others may be unaffected.

Distinct germination response of endangered and common arable weeds to reduced water potential.

Arable weeds are one of the most endangered species groups in Europe. Modern agriculture and intensive land-use management are the main causes of their dramatic decline. However, besides the changes in land use, climate change may further challenge the adaptability of arable weeds. Therefore, we investigated the response pattern of arable weeds to different water potential and temperature regimes during the phase of germination. We expected that endangered arable weeds would be more sensitive to differences in water availability and temperature than common arable weeds. To this end, we set up a climate chamber experiment where we exposed seeds of five familial pairs of common and endangered arable weed species to different temperatures (5/15, 10/20 °C) and water potentials (0.0 to -1.2 MPa). The results revealed a significant relationship between the reaction of arable weed species to water availability and their Red List status. The effects of reduced water availability on total germination, mean germination time and synchrony were significantly stronger in endangered than in common arable weeds. Therefore, global climate change may present a further threat to the survival of endangered arable weed species.

Negative and positive interactions among plants: effects of competitors and litter on seedling emergence and growth of forest and grassland species.

Living plant neighbours, but also their dead aboveground remains (i.e. litter), may individually exert negative or positive effects on plant recruitment. Although living plants and litter co-occur in most ecosystems, few studies have addressed their combined effects, and conclusions are ambivalent. Therefore, we examined the response in terms of seedling emergence and growth of herbaceous grassland and forest species to different litter types and amounts and the presence of competitors. We conducted a pot experiment testing the effects of litter type (grass, oak), litter amount (low, medium, high) and interspecific competition (presence or absence of four Festuca arundinacea individuals) on seedling emergence and biomass of four congeneric pairs of hemicryptophytes from two habitat types (woodland, grassland). Interactions between litter and competition were weak. Litter presence increased competitor biomass. It also had positive effects on seedling emergence at low litter amounts and negative effects at high litter amounts, while competition had no effect on seedling emergence. Seedling biomass was negatively affected by the presence of competitors, and this effect was stronger in combination with high amounts of litter. Litter affected seedling emergence while competition determined the biomass of the emerged individuals, both affecting early stages of seedling recruitment. High litter accumulation also reduced seedling biomass, but this effect seemed to be additive to competitor effects. This suggests that live and dead plant mass can affect species recruitment in natural systems, but the mechanisms by which they operate and their timing differ.

Litter effects on seedling establishment interact with seed position and earthworm activity.

Seedling establishment is influenced by litter cover and by seed predators, but little is known about interactions between these two factors. We tested their effects on emergence of five typical grassland species in a microcosm experiment. We manipulated the amounts of grass litter, seed sowing position and earthworm activity to determine whether: (i) the protective effect of litter against seed predation depends on cover amount and seed sowing position, i.e., on top or beneath litter; (ii) seed transport by earthworms changes the effect of seed sowing position on seedling emergence; and (iii) seeds transported into deeper soil layers by earthworms are still germinable. Litter cover and presence of earthworms lowered seedling emergence. The impact of seed position increased with seed size. Emergence of large-seeded species was reduced when sown on the surface. Additionally, we found an important seed position × earthworm interaction related to seed size. Emergence of large-seeded species sown on top of the litter was up to three times higher when earthworms were present than without earthworms. Earthworms also significantly altered the depth distribution of seeds in the soil and across treatments: on average 6% of seeds germinated after burial. In contrast to the seed position effect, we found no size effect on mobility and germinability of seeds after burial in the soil. Nevertheless, the fate of different-sized seeds may differ. While burial will remove large seeds from the regeneration pool, it may enhance seed bank build up in small-seeded species. Consequently, changes in the amount of litter cover and the invertebrate community play a significant role in plant community composition.

Genetic structure among and within peripheral and central populations of three endangered floodplain violets.

Understanding the partitioning of genetic variance in peripheral and central populations may shed more light on the effects of genetic drift and gene flow on population genetic structure and, thereby, improve attempts to conserve genetic diversity. We analysed genetic structure of peripheral and central populations of three insect-pollinated violets (Viola elatior, Viola pumila, Viola stagnina) to evaluate to what extent these patterns can be explained by gene flow and genetic drift. Amplified fragment length polymorphism was used to analyse 930 individuals of 50 populations. Consistent with theoretical predictions, peripheral populations were smaller and more isolated, differentiation was stronger, and genetic diversity and gene flow lower in peripheral populations of V. pumila and V. stagnina. In V. elatior, probably historic fragmentation effects linked to its specific habitat type were superimposed on the plant geographic (peripheral-central) patterns, resulting in lower relative importance of gene flow in central populations. Genetic variation between regions (3-6%), among (30-37%) and within populations (60-64%) was significant. Peripheral populations lacked markers that were rare and localized in central populations. Loss of widespread markers in peripheral V. stagnina populations indicated genetic erosion. Autocorrelation within populations was statistically significant up to a distance of 10-20 m. Higher average genetic similarity in peripheral populations than in central ones indicated higher local gene flow, probably owing to management practices. Peripheral populations contributed significantly to genetic variation and contained unique markers, which made them valuable for the conservation of genetic diversity.