No abstract

Cultivation of cash crops, such as cardamom (Elettaria cardamomum) in the forest understorey is a common practice in many tropical forests. Over time, cultivation may change forest structure and species composition, leading to gradual degradation of biodiversity and ecosystem services. Effective conservation of these forests requires an enhanced understanding of the demographic processes such as soil seed bank that may greatly influence future forest composition. We examined how the soil seed bank structure and composition responds to cardamom cultivation in a high conservation value Sri Lankan montane rain forest. Soil samples from natural forest with abandoned cardamom plantations (CP) and adjacent natural forest (NF) patches without cardamom were collected in dry and wet seasons. Soil samples were spread out in trays in a shade house and germination was recorded weekly for 19 weeks. The density of seeds in the soil seed bank was much higher in CP than NF. While grasses and forbs contributed the highest number and percentage of seeds in soils of both forest types, their densities in the soil seed bank were 9 and 2 times greater in the CP than the NF, respectively. Seeds of the nonnative herbs Ageratina riparia and E. cardamomum were 4 and 20 times greater in the soil of CP, respectively. Seeds of light demanding tree species such as

Question What strategies are most appropriate for restoring tree seedling recruitment whilst avoiding the spread of invasive plant species in high conservation value tropical forests disturbed by planting a shade‐demanding crop? Location Knuckles Forest Reserve, Sri Lanka (7°21′–7°24′ N, 80°45′–80°48′ E). Methods An experiment was conducted to test the effects of clipping or removal of established cardamom plants on recruitment of native tree seedlings and spread of non‐native plants in a tropical montane forest with abandoned cardamom stands in the understorey. The number and composition of tree seedling emergents, the cover of herbaceous plants and the recovery of cardamom were assessed for 3 yr. Results Tree seedling recruitment was higher in plots from which above‐ground cardamom biomass had been removed through slashing (mean ± SE per 5 m2; 28.9 ± 2.70) and those where cardamom plants had been removed completely through uprooting (32.2 ± 3.17), or when dead cardamom leaves and stems were removed with small‐scale extraction of pods (22.5 ± 2.16), than in unmanipulated control plots (16.6 ± 1.13) over 15 mo. The species composition of tree seedling emergents did not differ in response to removal of cardamom. However, the cover of herbaceous plants, including the non‐native invasive Ageratina riparia, increased in response to removal of cardamom. Recovery of cardamom was higher when the plants had been slashed than when entire plants were uprooted and removed. Conclusion Slashing or uprooting cardamom plants is a potential strategy for restoring tree seedling recruitment in forests with abandoned cardamom stands in the understorey, but these interventions would need to be repeated annually over many years to be successful, and they risk promoting expansion of the cover of herbaceous plants, including non‐native species. Hence, this approach would be labour‐intensive and costly. An alternative approach to promoting tree seedling emergence and establishment is to clear dead cardamom leaves and stems, and to encourage small‐scale extraction of pods from the residual cardamom plants. Harvesting pods reduces the likelihood that the crop will be sustained in situ through natural regeneration, and supplies an income to local communities, which would enhance the social acceptability of the intervention.