The Hellshire Hills is a tropical dry limestone forest located on the south coast of
Jamaica in the parish of St. Catherine. The forest has been affected by exploitation by charcoal burners as it provides wood for the production of charcoal and the unregulated actions of the charcoal burners over the last 1O to 20 years has resulted in a significant reduction of the forest area. In 1998, the Hellshire Hills was declared a protected area and presently there is a need to implement conservation strategies to protect, preserve, and replenish the remaining natural forest cover. This is important, as the area apart
from being one of the last remanding stands of primary limestone forest in the
Caribbean, is also the last remaining habitat for the endangered Jamaican Iguana
Cyclura collei.
One strategy being considered is the establishment of an outer buffer zone
surrounding a central core of undisturbed forest. The forest areas that have been
disturbed will make up the buffer zone and some amount of sustainable cutting will be allowed in the buffer zone. But nothing is known about the regeneration ecology of the forest and whether the forest can support any type of cutting.
The objective of this study is to contribute to this understanding by looking at the rates and identify the major pathway(s) of forest recovery (whether by seeds or by vegetative propagation) after different levels (intensity) of human disturbance. The study will attempt to identify species that are affected by disturbance and will determine the effects of different intensities of disturbance on the ability and rate of recovery of the forest community. It is expected to provide information that can be used to make generalizations about the patterns of response in the community so that predictions can be made about the response of the species that were not included within this study.
To address the objective of this study, a randomized block experiment consisting
of three treatments and four blocks covering a total area of 0.27 ha was set up to
investigate the short-term effects of different intensities of basal area removal on
regeneration in the Hellshire Hills. Before the application of the treatments a total of 1278 trees and 7863 seedlings, comprising 60 and 52 species respectively, were sampled throughout the experimental plots (0.27 ha). Pre-treatment assessments of forest stand structure revealed a reverse-J curve for adult (~ 2 cm DBH) and seedling ( <2 cm DBH) size class distributions, which indicated that the forest was probably regenerating adequately. Six regenerative species groups were formed using a correspondence analysis based on adult and seedling diameter class distributions, which was a good predictor of the ability of tree species to recover after disturbance.
Treatment application involved the removal (by cutting) of approximately 40%
and 100% of the basal area of one of the assigned treatment plots after which the
resulting dynamics: seed availability and dispersal, recruitment, mortality and the effects of moisture and light on the initial establishment of seedlings were assessed. Vegetative re-growth of cut stems was also assessed within all the cut plots to determine which of the two pathways of recovery (that is, by seed or vegetative re-growth) was the dominant mechanism in the cut sites and to identify the possible stage(s) at which regeneration by seed within disturbed sites may be limited.
Post-disturbance assessments revealed that coppicing was the primary
regeneration mechanism in the cut sites; it was a much more reliable and robust
regeneration mechanism. As a result, after 14 months, sprouts recovered 15% of the diameter of the pre-cut forest with the diameter growth increments of shoots being ca. 95% higher than that of seedlings in the cut plots. Also if the number of shoots and the number of seedlings alive were taken as a measure of recovery, then coppiced shoots accounted for 80% of the recovery of all plants while seedlings accounted for 20% of the recovery of all plants in the first dry season after cutting. Recovery by seed, however, resulted in a slight increase in the diversity of individuals (51 species precut to 52 post-cut), with recovery by shoots alone resulting in a reduction of species richness because
three species failed to produce coppice shoots (51 species precut to 48 species post-cut).
Regeneration by seed was found to be limited at the initial establishment stage of
the regeneration cycle with seed dispersal and seedling recruitment not being
significantly affected by this low intensity disturbance. Cutting significantly increased forest floor light intensities, which had a negative effect on seedling density within the disturbed plots. Mortality rates within the cut plots were significantly higher than in uncut plots and showed a significant decrease with increased shading three months after cutting. After six months this effect was no longer significant but the overall density of seedlings was shaped by this initial effect, where uncut plots maintained a significantly higher number of seedlings than the cut plots. Of the cut plots, partially cut plots maintained a significantly higher number of seedlings than clear-cut plots. Seasonality maintained this interaction throughout the study period by operating on differential rates
of mortality and recruitment within the cut and un-cut plots.