Soil Seed Banks of Tree Species from Natural Forests, Restoration Sites, and Abandoned Areas in Chiang Mai, Thailand

Soil seed banks have been used for investigation of natural regeneration of forests. In this study, we compared seed density and species composition of soil seed banks of trees among natural forests, restored forests of different ages, and abandoned agricultural land. The soil seed banks were collected from a natural forest (NF), 12-year-old and 17-year-old restoration sites (RF12y and RF17y), and 17-year-old abandoned site (AA) at Ban Mae Sa Mai, Chiang Mai, Thailand. A seedling emergence technique was used to assess seed density and species of emerged seedlings was identified. We found 5-8 tree species at each site. Seed densities in the study areas ranged from 43 to 298 seeds/m2. The seed density of RF12y was significantly higher than that of both NF and AA but not significantly different than RF17y (p < 0.01). Although there was no significant relationship between the restoration ages and the seed densities of the soil seed banks, the species composition of standing vegetation was related to the seed bank species. Sorensen’s similarities between the species composition of the soil seed banks and the existing trees in each area were between 0 and 13.79%, suggesting seed dispersal of both within and across study sites. Eight out of fourteen species in the soil seed banks were dispersed into restoration sites without standing vegetation of those species. Seven of those were animal-dispersed species. The selected native trees, framework species, attracted small seed dispersers into the study areas, especially at the restoration sites. This finding suggests that active forest restoration improved natural regeneration in restoration sites as well as neighboring areas via seed dispersal.

Seed Rain, Soil Seed Bank, and Seedling Emergence Indicate Limited Potential for Self-Recovery in a Highly Disturbed, Tropical, Mixed Deciduous Forest

Human activity negatively affects the sustainability of forest ecosystems globally. Disturbed forests may or may not recover by themselves in a certain period of time. However, it is still unclear as to what parameters can be used to reasonably predict the potential for self-recovery of human-disturbed forests. Here, we combined seed rain, soil seed bank, and seed emergence experiments to evaluate the potential for self-recovery of a highly disturbed, tropical, mixed deciduous forest in northeastern Thailand. Our results show a limited potential for self-recovery of this forest due to low seedling input and storage and an extremely high mortality rate during the drought period. There were 15 tree species of seedlings present during the regeneration period in comparison with a total number of 56 tree species in current standing vegetation. During the dry season, only four tree seedling species survived, and the highest mortality rate reached 83.87%. We also found that the correspondence between the combined number of species and composition of plant communities obtained from seed rain, soil seed bank, and seedling emergence experiments and the standing vegetation was poor. We clearly show the temporal dynamics of the seed rain and seedling communities, which are driven by different plant reproductive phenology and dispersal mechanisms, and drought coupled with mortality. We conclude that this highly disturbed forest needs a management plan and could not recover by itself in a short period of time. We recommend the use of external seed and seedling supplies and the maintenance of soil water content (i.e., shading) during periods of drought in order to help increase seedling abundances and species richness, and to reduce the mortality rate.

Soil seed banks and restoration potential of tussock sedge meadows after farming in Changbai Mountain, China

Vast tracts of the wettest agricultural fields have been abandoned worldwide, hence knowledge of their restoration potential is becoming important. Soil seed banks can be important components of ecological restoration, particularly if the seeds of key structural dominants can survive periods of cultivation. In Changbai Mountain, China, we compared the seed banks and standing vegetation between natural sedge meadows and adjacent paddy fields under drained and flooded conditions. The tussock-forming sedge Carex schmidtii dominated plant communities in the natural sedge meadows. However, this key structural dominant species was not found in the paddy fields. Other important wetland species survived cultivation as seeds (e.g. Cyperus fuscus, Lythrum salicaria and Sagittaria trifolia). Species of various life history types require either drawdown (emergent; e.g. C. schmidtii) or flooding (e.g. Potamogeton pectinatus) for successful germination. This study indicates that the seed banks of farmed sedge meadows could contribute towards the establishment of new wetland vegetation assemblages. Active revegetation of sedge meadow, particularly the native tussock-forming Carex, and changes in environmental conditions are necessary to support the region’s biodiversity.

Impact of forest fire on soil seed bank composition in Himalayan Chir pine forest

Aims To explain how plant community copes with a recurring anthropogenic forest fire in Himalayan Chir pine forest, it is important to understand their postfire regeneration strategies. The primary aim of the study was to know: (i) how fire impact soil seed bank composition and (ii) how much soil seed bank composition differs with standing vegetation after the forest fire. Methods Soil samples were collected from burned and adjoining unburned sites in blocks using three layers down to 9 cm depth immediately after a forest fire and incubated in the net-house for seedling emergence. Same sites were revisited during late monsoon/early autumn season to know the species composition of standing vegetation recovered after a forest fire. Important Findings Soil contained viable seeds of &gt;70 species. The average seed bank density was 8417 and 14 217 seeds/m2 in the burned and unburned site, respectively. In both sites, it decreased with increasing soil depth. Overall fire had no significant impact on seed density; however, taking individual layers into consideration, fire had a significant impact on seed density only in the uppermost soil layer. The species richness of soil seed bank and standing vegetation was 73 and 100, respectively (with 35 shared species), resulting in a similarity of about 40%. In contrast, &gt;80% species in soil seed bank was found similar between burned and unburned sites. Further, there were no significant differences in species richness of standing vegetation in burned (87 spp.) and unburned (78 spp.) sites. Our results showed that fire had an insignificant impact on soil seed bank composition and restoration potential of a plant species from seeds. The understory herb and shrub plant community’s ability to form a fire-resistant viable soil seed bank and capable to recover in the postfire rainy season, explains how they reduce the risk of recurring fire damage in maintaining their population.

Germinable Soil Seed Bank in Biancana Badlands

Seed banks are important for understanding vegetation dynamics and habitat regeneration potential. Biancana badlands are vanishing landscapes where recurring and non-recurring management has been advocated to restore vegetation. Here, we investigated germinable seed bank structure and composition of a biancana badland in central Italy and evaluated the relationship between the standing vegetation and soil seed bank. We identified four land cover classes in five biancana badlands of Tuscany (central Italy) and collected data from 132 vegetation plots and 660 soil cores. We recorded 117 species in the standing vegetation. The seedlings that emerged from the soil samples, mostly annual species, numbered 183 and belonged to 31 taxa (392.5 seedlings/m−2 on average across the four land cover classes). Standing vegetation showed an aggregated spatial pattern with distinct communities while the seed bank showed a less aggregated spatial pattern. The similarity between the seed bank and standing vegetation was low. In contrast with the features generally found for disturbed and pioneer communities, but in line with seed bank characteristics of other badlands, the seed bank was particularly poor in species.

Seedling Emergence from Seed Banks in Ludwigia hexapetala-Invaded Wetlands: Implications for Restoration

Soil seed banks play a critical role in the maintenance of wetland plant communities and contribute to revegetation following disturbances. Analysis of the seed bank can therefore inform restoration planning and management. Emergence from seed banks may vary in response to hydrologic conditions and sediment disturbances. To assess the community-level impact of exotic Ludwigia hexapetala on soil seed banks, we compared differences in species composition of standing vegetation among invaded and non-invaded wetlands and the degree of similarity between vegetation and soil seed banks in northern California. To determine potential seed bank recruitment of L. hexapetala and associated plant species, we conducted a seedling emergence assay in response to inundation regime (drawdown vs. flooded) and sediment depth (surface vs. buried). Plant species richness, evenness, and Shannon’s H’ diversity were substantially lower in standing vegetation at L. hexapetala invaded sites as compared to non-invaded sites. Over 12 months, 69 plant taxa germinated from the seed banks, including L. hexapetala and several other exotic taxa. Seedling density varied among sites, being the highest (10,500 seedlings m−2) in surface sediments from non-invaded sites subjected to drawdown treatments. These results signal the need for invasive plant management strategies to deplete undesirable seed banks for restoration success.

Relationships between soil seed bank composition and standing vegetation along chronosequences in a tropical dry forest in north-eastern Brazil

To better understand the role of seed banks in ecological succession of dry forests, we compared similarities between vegetation and seed banks and assessed the relative contributions of seed dispersal and persistence in chronosequences in the Brazilian semi-arid region. To sample the standing vegetation and the seed bank, we collected data in three sites with three successional ages in each one (5 y, 25 y and 45 y). A total of 180 soil samples (three sites × three successional ages × 10 plots × two components) were collected. The composition of the seed bank was assessed by the seedling emergence method. Of 166 species identified in the standing vegetation, only 50 (30.1%) were also present in the seed bank, resulting in low similarity (Jaccard index = 0.02–0.21) and reflecting the rarity of woody species and the dominance of annuals (71% of richness). The relative importance of seed persistence and seed dispersal to seed banks composition were balanced in most cases (difference was not rejected in four out six comparisons). Those results suggest that seed banks in tropical dry forests are largely the result of high dispersal rates and the persistence of allochthonous annual species that contribute to decoupling seed bank and vegetation composition.

Seed banks of subtropical grassland patches within an urban matrix in South Africa: reflecting the past and foretelling the future

Urban grasslands often exist as fragmented patches with varying threat levels. These patches are generally inappropriately managed due to limited data on their responses to disturbance and seed bank regenerative potential. This study compared seed banks between non-naturally disturbed and undisturbed patches of South African subtropical grasslands in an urban environment. Standing vegetation was characterised via year-round quadrat and monthly transect sampling. The seed banks were sampled after the two main dispersal periods and assessed for species composition and density. The seed bank species (24) represented a subset (9%) of the standing vegetation, with forbs being the most speciose life form and grasses the most abundant. Disturbance had a moderate impact on seed bank density but significantly affected species composition and ordering. Clustering and ordination analyses separated seed banks based on disturbance history. However, seed banks in disturbed patches maintained the general character of the grassland type. Diagnostic, rare and threatened taxa were absent from seed banks, irrespective of disturbance history. Consequently, restoration will possibly require seed enrichment and species reintroduction rather than exploitation of natural seed banks. Identifying disturbance-sensitive species through seed bank surveys can inform grassland conservation and restoration strategies.