Environmental changes in biodiversity hotspot ecosystems of
South Ecuador: RESPonse and feedback effECTs (FOR2730)

Abstract:

While tropical deforestation proceeds at high pace, conservation biologists remain divided over the extent to which anthropogenic habitats will be able to offset the loss of biodiversity from primary forests. Case studies addressing disturbance effects on epiphyte assemblages have yielded contrasting patterns, and underlying mechanisms remain unclear. At present, one of the questions most widely debated is the importance of growth conditions vs. dispersal limitations for the persistence of epiphyte populations following disturbance. The present dissertation aims at contributing to our understanding of the processes that determine epiphyte diversity in anthropogenic landscapes, focussing primarily on assemblages of vascular epiphytes on isolated trees. Isolated trees offer an excellent model system since they constitute the smallest possible forest fragment, exposed to multiple edge effects and dispersal limitations. Field work was done from 2003?2006 at two lower montane sites in the Ecuadorean Andes: (1) moist forest (2200 mm rain/yr; 1800?2200 m elevation) and adjacent matrix habitat around Estación Científica San Francisco (ECSF) on the eastern Andean slope (Zamora-Chinchipe Province), and (2) perarid Interandean forest (530 mm rain/yr; 2300m elevation) and adjacent pastures at Bosque Protector Jerusalén (Pichincha Province). At the moist forest site (ECSF), I studied the diversity of vascular epiphytes on remnant trees ca. 10?30 years after their isolation in pastures (chapter 2). The objective of this study was to document the importance of remnant trees for the survival of vascular epiphyte communities following forest clearance. Trees were divided into five zones following Johansson (1974) and climbed with the single rope technique. Abundance and diversity of epiphytes were significantly lower on remnant trees than on forest trees. Several families rich in xerophilous taxa (Bromeliaceae, Orchidaceae, Piperaceae, Polypodiaceae) were relatively well represented on remnant trees in terms of species richness and abundance, whereas essentially hygrophilous families such as Dryopteridaceae, Ericaceae, Grammitidaceae and Hymenophyllaceae were poorly represented or absent. Impoverishment was greatest on the stem base and in the outer crown, and least in the inner crown of the host trees. These results suggest that the observed impoverishment on isolated trees is closely related to microclimatic changes. Experimental work done at the same site gave some insight into the mechanisms behind the pronounced impoverishment of epiphyte assemblages. In one field trial (chapter 6) I studied abundance, diversity and floristic composition of epiphyte seedling establishment on isolated and adjacent forest trees. All vascular epiphytes were removed from plots on the trunk bases of a common tree species (Piptocoma discolor). Newly-established epiphyte seedlings were recorded after two years, and their survival after another year. Seedling density and the number of taxa (families and genera, respectively) per plot were significantly reduced on isolated trees relative to forest trees, as was the rarefied total number of epiphyte taxa. Seedling assemblages on trunks of forest trees were dominated by hygrophilous understorey ferns whereas assemblages on isolated trees were dominated by xerophilous canopy taxa. Colonisation probability was significantly higher for plots closer to forest but not for plots with greater canopy or bryophyte cover. Seedling mortality on isolated trees was significantly higher for hygrophilous than for xerophilous taxa. These results show that altered recruitment can explain the long-term impoverishment of post-juvenile epiphyte assemblages on isolated remnant trees. Altered establishment is attributed to a combination of dispersal constraints and the changed microclimate that was documented by measurements of temperature and humidity. Although isolated trees in anthropogenic landscapes are key structures for the maintenance of forest biodiversity in many aspects, my results show that their value for the conservation of epiphyte diversity can be limited. Abiotic seedling requirements may increasingly constitute a bottleneck for the persistence of vascular epiphytes in the face of ongoing habitat alteration and atmospheric warming. In a second field trial at the same site (chapter 7) I studied the response of assemblages of well-established vascular epiphytes to severe forest disturbance. Individual plants were marked on isolated remnant trees (up to 5 m above ground) in a fresh clearing and in undisturbed forest (controls), and their growth and survival was followed during three consecutive years. Wind-throw and branch breakage caused the loss of 24% of plants from isolated trees but only of < 1% in the forest. Epiphyte mortality on the remaining, intact phorophytes was substantially higher on remnant trees than on forest trees, averaging 72% vs. 11% after 3 years. Mortality on isolated trees was greatest after the first year (52%) and among ferns and dicots, and lowest after the second year (20%) and among monocots (aroids, bromeliads, orchids). Mortality differed significantly between taxa but was elevated significantly for most taxa after the first year relative to forest levels. Plants that survived on isolated trees commonly showed a marked decrease in maximum leaf length. The annual increment in leaf number varied more widely, both between and within epiphyte families. This study provides first experimental, field-based evidence that increased physical exposure affects the performance of well-established vascular epiphytes. The results suggest that growth conditions may often be a more influential driver of vascular epiphyte diversity in disturbed habitats than dispersal. It is argued that elevated desiccation stress is only one of several mechanisms by which increased physical exposure may limit epiphyte performance in disturbed habitats. The two trials showed that the marked impoverishment of vascular epiphyte assemblages at the study site results from both lowered recruitment and survival of established plants, and that mid-term effects of changes in growth conditions can be drastic. At the dry forest site (Bosque Protector Jerusalén) I studied the development of epiphyte diversity and abundance across a disturbance gradient (chapters 3 and 4). Epiphytic bryophytes and vascular plants were sampled on 100 trees of the dominant tree species, Acacia macracantha, in five habitats: closed-canopy mixed and pure acacia forest (old secondary), forest edge, young semi-closed secondary woodland, and isolated trees in grasslands. Host trees were divided into four zones (cf. Johansson 1974). Species density and surface cover of vascular epiphytes were determined for each zone, bryophytes were only sampled in the inner crown. Total species richness in forest edge and on isolated trees was significantly lower than in closed forest types. Species density of vascular epiphytes did not differ significantly between habitat types. Species density of bryophytes, in contrast, was significantly lower in forest edge and on isolated trees than in closed forest. Surprisingly, edge habitat showed greater impoverishment than semi-closed woodland and similar floristic affinity to isolated trees and to closed forest types. Assemblages were significantly nested; habitat types with major disturbance held only subsets of the closed forest assemblages, indicating a gradual reduction in niche availability. This conclusion is lent further support by patterns of epiphyte abundance. Bromeliads were xerophilous and early-successional, being most abundant in the middle crown, whereas ferns were skiophilous and late-successional, being essentially restricted to the inner crown. While the former were significantly more abundant on isolated trees than in closed forest, the latter were restricted to closed forest and semi-closed woodland. Bryophyte cover was significantly lower on isolated trees and in forest edge than in closed forest. Total species density and covers of epiphytes on isolated trees were not related to the distance to forest, but total species density and bryophyte cover were closely correlated with crown closure as a measure of canopy integrity. These results suggest that microclimatic changes are key determinants of the observed impoverishment of epiphyte assemblages following disturbance, and that epiphytic cryptogams are sensitive indicators of microclimate and human disturbance in montane dry forests. Patterns of abundance and diversity further suggest that the ?similar gradient hypothesis? (McCune 1993) applies to tropical dry forest assemblages of epiphytes. The substantial impoverishment of edge habitat underlines the need for fragmentation studies on epiphytes elsewhere in the Tropics. Case studies on vascular epiphyte assemblages have failed to find a common, general response to human disturbance. In chapter 5, a hypothesis is proposed and explored that aims to explain and unite these conflicting patterns in the light of a non-linear effect of local climate: while the diverse assemblages of moist, slightly seasonal forests rely on the integrity of forest canopies during droughts, this dependency may be less pronounced in aseasonally wet or distinctly dry forests. This hypothesis is supported by the results of a literature survey.