Internal fragmentation in the rainforest: edge effects of highways, powerlines and watercourses on tropical rainforest understorey microclimate, vegetation structure and composition, physical disturbance and seedling regeneration

Pohlman, Catherine Louise (2006) Internal fragmentation in the rainforest: edge effects of highways, powerlines and watercourses on tropical rainforest understorey microclimate, vegetation structure and composition, physical disturbance and seedling regeneration. PhD thesis, James Cook University.

PDF (Thesis front)
Download (5MB)
PDF (Chapters 1-4)
Download (4MB)
PDF (Chapters 5-8)
Download (4MB)
PDF (References and Appendices)
Download (657kB)


Human activities have led to the loss and fragmentation of rainforest around the world’s tropics and sub-tropics, whilst internal fragmentation by linear clearings has subdivided many remaining areas of rainforest, exacerbating the effects of the initial disturbance. Linear clearings for human infrastructure, including roads and powerlines, are widespread within the rainforests of the world and the Wet Tropics World Heritage Area of north-eastern Queensland in particular. Edge effects of these linear clearings have the potential to substantially increase the disturbed area within existing tracts of rainforest and affect a wide range of rainforest biota.

I examined the edge effects of two artificial linear clearings on rainforest understorey microclimate, physical disturbance regimes, vegetation structure and composition and tree seedling survival and growth. Edge effects adjacent to powerline clearings and highways were compared with those near perennial creeks, the only type of widespread natural linear canopy opening available in the complex mesophyll vine forest of the Palmerston Region in the Wet Tropics World Heritage Area of northeastern Queensland (average annual precipitation 3000 – 4000 mm).

Understorey microclimate of rainforest adjacent to highways, powerline clearings and creeks was significantly altered from that of the forest interior. Light intensity was elevated near the edges of powerline clearings, highways and creeks with the effect being strongest for creek edges. Air temperature and vapour pressure deficit were elevated near powerline clearing edges in the dry season and highway edges in both wet and dry seasons but were not elevated near creek edges in either season. In contrast, soil moisture was reduced near creek edges but remained unaltered near powerline clearing and highway edges. Air speed was not significantly related to distance from the forest edge when assessed with instantaneous measurements although slight increases in air speed were detected near the edge of the forest when diurnal patterns were assessed using data loggers. The predominant air speed throughout the study sites, however, remained at or close to 0 m s-1. Canopy temperature was elevated near highway edges and lowered near powerline edges in the wet season but no edge gradients in canopy temperature were detected near creek edges in either season. These different edge gradients may be largely the result of differences in the fluxes of latent and sensible heat within each type of linear canopy opening with these data indicating that the nature of the linear canopy opening is at least as important as the width in determining the severity and extent of microclimatic edge effects.

Variation in the understorey light environment was assessed using the red:far red ratio, which has been shown in other studies to correlate very closely with percent canopy transmittance as well as other commonly used measures of canopy cover and light availability. Average red:far red values were elevated near creek edges and reduced near powerline and highway edges although the proportion of high light measurements (≥ 0.70) was elevated within 12 m of all edge types. These data indicate that anthropogenic edges are more ‘sealed’ (ie have a greater foliage density near the edge) than creek edges. This more open vegetation structure near creek edges was accompanied by a decline in soil moisture (due to changes in soil texture) suggesting that periodic flooding may be one of the causative factors maintaining lower foliage density and thus a comparatively more open edge structure.

Measures of physical disturbance provided mixed results. There were greater numbers of fallen trees near the edges of powerlines, highways and creeks but no particular tree size appeared to be disproportionately at risk of tree-fall mortality. As there was also a greater number of living trees near the forest edge, it is uncertain whether the mortality rate of canopy trees truly was elevated near the forest edge. In contrast, small-scale disturbance to seedlings through falling leaf litter and canopy debris was not consistently greater near the forest edge than in the forest interior. Although the rate of small-scale disturbance, as measured with artificial seedlings, was elevated near highway edges, there was no consistent reduction in the stature of taller (≥ 50 cm) seedlings near the edge as might be expected amongst seedlings suffering greater rates of physical trauma. Whilst the elevated rate of damage to artificial seedlings near highway edges might be a result of greater moisture stress, perhaps as a result of elevated wet season canopy temperatures and year-round increases in understorey vapour pressure deficit near highways, the lack of any consistent signal in seedling stature suggests that this effect may not significantly influence the seedling community.

Vegetation structure and composition were altered near the edge of the forest, relative to the forest interior. The abundances of vines, small trees and saplings were elevated near the forest edge. Pioneer species were also more abundant near the edge, particularly adjacent to creeks. Light-demanding herbs occurred more frequently and late-successional shrubs less frequently near the forest edge than in the interior. Grasses and weeds were found almost exclusively on the very edge of the forest but were also recorded in large canopy gaps within 12 m of highway edges and at low abundance within 12 m of creek edges. Changes in the proportions of early-, mid- and latesuccessional tree species between adults and juveniles suggested that understorey light conditions were no longer suitable for the recruitment of early-successional trees near anthropogenic edges. This suggests that anthropogenic edge structure has become more sealed since these edges were created. In contrast, understorey light availability near creek edges remained sufficient for the recruitment of light-demanding pioneer species.

Proximity to the edge did not significantly affect seedling survival or growth for eight species included in a herbivore-exclusion experiment. The seedlings of pioneer species survived only in areas with high light availability (ie within the powerline clearing and near creek edges). Light availability was also the strongest determinant of the survival of seedlings of mid- and late-successional species. There was weak evidence for a decrease in herbivory pressure near the forest edge and no evidence of an effect of edge distance on seedling survival or growth. The majority of seedlings did not experience any growth in this experiment although the probability of seedlings experiencing positive growth rates increased with increasing light availability. The major differences observed in this study occurred in seedling survival rather than seedling growth, suggesting that the ability of seedlings to persist in the understorey may have been an important factor influencing the composition of the tree seedling community.

These data indicate that artificial linear clearings were associated with edge gradients in the abiotic environment of the adjacent forest and that these abiotic edge effects were the driving force behind changes observed in the plant community at the forest edge. The understorey of forest near the edges of highways and powerlines was brighter, warmer and drier than the forest interior although only slightly windier. In contrast, the edges of natural linear canopy openings did not cause elevated moisture stress but did experience elevated light availability. Analysis suggests that a combination of differences in the degree of edge ‘sealing’ among edge types caused by varying disturbance regimes in the clearings and differences in the physical properties of the linear canopy openings themselves (particularly in the level of evaporative cooling) were responsible for the observed biotic and abiotic edge gradients near natural and anthropogenic linear canopy openings.

Item ID: 1349
Item Type: Thesis (PhD)
Keywords: tropical rainforests, powerlines, watercourses, understorey vegetation, microclimates, seedling regeneration, rainforest clearings, canopy cover, environmental disturbances
Date Deposited: 24 Jan 2007
FoR Codes: 05 ENVIRONMENTAL SCIENCES > 0501 Ecological Applications > 050104 Landscape Ecology @ 50%
05 ENVIRONMENTAL SCIENCES > 0501 Ecological Applications > 050102 Ecosystem Function @ 50%
SEO Codes: 96 ENVIRONMENT > 9612 Rehabilitation of Degraded Environments > 961203 Rehabilitation of Degraded Forest and Woodlands Environments @ 50%
97 EXPANDING KNOWLEDGE > 970105 Expanding Knowledge in the Environmental Sciences @ 50%
Downloads: Total: 2236
Last 12 Months: 46
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page