The ecology of seedlings in Central Amazonian forest fragments / Julieta Benitez-Malvido.

Tese (Ph.D.) - University of Cambridge, 1995

a) The effects of tropical rain forest fragmentation and succession following fragmentation, on the seedling community were evaluated according to different successional stages, size of the isolate and position of the seedlings within the forest fragments. The biotic and abiotic factors that might be determining seedling abundance, such as light, temperature, fine litter mass and herbivory were also measured. The seedling parameters measured to evaluate effects of forest disturbance were, density, for the naturally occurring seedlings and survival, growth and leaf number for transplanted seedlings of the Sapotaceae family. Germination tests ware carried out on seeds of the selected Sapotaceae species. b) The habitats studied were: i) continuous forest (control); ii) 100-ha fragment; iii) 10-ha fragment; iv) 1-ha fragment; v) secondary forest and vi) pasture. For 10 and 100-ha fragments seedling density and performance were measured in the centre, the edge and the corner of the fragments, giving a total of 10 different treatments. c) Overall, the mean density of seedlings (=5 cm but =100 cm tall) per m² declined from continuous forest (14.4±SD 3.1); to 100-ha fragments (12.9±SD 2.4); to 10-ha fragments (11.8±SD 3.0); to 1-ha fragments (9.6±SD 2.8); to secondary forest (3.7±SD 2.0). Different positions within 100 fragments differ in the abundance of seedlings. Corners of 100-ha fragments had lower densities of seedlings than plots in the centres and edges; seedling density increased with distance from the forest edge within 100 and 10-ha fragments. d) For transplanted seedlings the mean percentage survival, for all habitats, declined from Chrysoplyllum pomiferum (87%), to Pouteria caimito (69%), to Micropholis venulosa (44%) over 12 months. Seedlings of C. pomiferum and M. venulosa, had higher growth rates in the secondary forest whereas pasture appeared to be a more suitable habitat for seedlings of P. caimito. The range of habitats in which C. pomiferum seedlings established was wider than in the other species. There was no clear edge related pattern on growth for two of the species, whereas C. pomiferum showed a decrease in growth and number of leaves cioser to the edge of the fragments. e) Abiotic and biotic factors differed among different habitats. Mean maximum temperatures wore higher in smaller fragments, secondary forest and pasture; photosynthetically active radiation was much higher in secondary forest and pasture; but there was no apparent diifference between continuous forest and fragments; nor at different positions, or at different distances from the forest edge within 100 and 10-ha fragments; herbivory levels decreased with fragment size for Pouteria caimito and Micropholis venulosa, and were much lower in the pasture for the three species. f) Fine litter mass (FLM) differed among habitats, pasture had lower FLM than any other habitat. FLM was not different at different positions within fragments, nor at different distances from forest edges older than eight years but there was an increase of FLM closer to the edges of a two-year old isolated fragment. Seasonal differences on FLM ware found for continuous forest, 100-ha fragment and pasture. The responses of Sapotaceae seedlings to differences in litter quantity were species- specific. The study also showed that herbivory patterns may be altered by the distribution of litter. Seedlings of M. venulosa were heavily defoliated on plots where litter was removed whereas herbivory was much lower in seedlings growing with litter cover. g) Seeds from all species germinate =12 weeks after sowing. Germination tests showed that all species have very high rates of germination and that seeds of P. caimito can be defined as recalcitrant seeds. h) Forest fragmentation appeared to be more important in affecting seedling abundance than the size of the fragment. Tropical tree species that co-occur as adults have different abilities to cope with habitat alterations at the seedling stage. The size of the fragments did not affect species equally.