Mature leaves of tree seedlings were exposed to high light in four experimental gaps in the Jamaican upper montane rainforest (UMRF). Two of the six species studied were light-demanders: Alchornea latifolia and Clethra occidentalis. Two were gap-favoured: Pittosporum undulatum (an invasive) and Palicourea alpina (a subcanopy shrub). One was intermediate: Hedyosmum arborescens, and one was shade-tolerant: Guarea glabra. After five months, the following significant changes occurred in shade leaves that were exposed to gaps (‘shade-to-gap’ leaves; values as % of those in the pre-gap shade): maximum rate of photosynthesis + 40% (Alchornea), +35% (Clethra), −34% (Pittosporum), +72% (Palicourea); dark respiration +120% (Alchornea), +140% (Clethra), +60% (Pittosporum), +233% (Palicourea), +175% (Hedyosmum), +100% (Guarea); leaf thickness +18% (Alchornea), +18% (Clethra), +14% (Palicourea); leaf mass per unit area +18% (Alchornea), +15% (Pittosporum). Leaves produced in the gaps were (as a percentage of total live leaf number) 74% (Alchornea), 71% (Clethra), 50% (Pittosporum), 71% (Palicourea), 62% (Hedyosmum) and 50% (Guarea). Photosynthetic rates of leaves produced in the gaps were 53–120% higher than ‘shade-to-gap’ leaves. Overall, shade leaves on the three native, more light-demanding species (Alchornea, Clethra and Palicourea) showed photosynthetic acclimation, while the more shade-tolerant species (Hedyosmum and Guarea and Pittosporum undulatum) showed little acclimation in shade-to-gap leaves.

Selective logging in tropical forests may lead to deforestation and forest degradation, so accurate mapping of it will assist in forest restoration, among other ecological applications. This study aimed to track canopy tree loss due to illegal logging of the important hardwood tree Ocotea usambarensis in a closed-canopy submontane tropical forest by evaluating the mapping potential of the very-high-resolution WorldView-3 multispectral dataset using random forest (RF) and support vector machine (SVM) with radial basis function kernel classifiers. The results show average overall accuracies of 92.3 ± 2.6% and 94.0 ± 2.1% for the RF and SVM models, respectively. Average kappa coefficients were 0.88 ± 0.03 for RF and 0.90 ± 0.02 for SVM. The user’s and producer’s accuracies for both classifiers were in the range of 84–100%. This study further indicates that vegetation indices derived from bands 5 and 6 helped detect canopy gaps in the study area. Both variable importance measurement in the RF algorithm and pairwise feature selection proved useful in identifying the most pertinent variables in the classification of canopy gaps. These findings could allow forest managers to improve methods of detecting canopy gaps at larger scales using remote sensing data and relatively little additional fieldwork.

African forests are under increasing pressure to supply local, regional, and international demand for timber. Much of this trade is unregulated, such that there is increasing concern regarding the ecological sustainability of this resource use. However, there is a lack of studies investigating the ecological impact of informal timber harvesting in African forests. While forest species have adapted to natural canopy gap dynamics, harvesting may alter natural disturbance regimes, with adverse effects on biodiversity. Information regarding harvest gaps, and concomitant impacts on habitat and biodiversity is thus essential to inform sustainable management. This study compared the frequency and nature of harvest gaps and natural gaps in a coastal forest in the Eastern Cape, South Africa, where informal selective timber harvesting occurs. Habitat condition and bird species richness and composition were compared across intact forest, natural gaps, and harvest gaps. Harvest-created gaps increased the number of canopy gaps by 30%, but were comparable with natural gaps in size. Habitat conditions in harvest gaps represented an intermediate state between intact forest and natural gaps, thereby increasing forest-scale habitat heterogeneity. While bird species richness was not affected, species composition differed across intact forest, harvest gaps and natural gaps, driven by changes in habitat condition, and mediated by species’ feeding and nesting traits. Specifically, insectivores, cavity- and ground-nesting species, and hawking and arboreal probing species were negatively affected by the habitat gradient from intact sites to canopy gap conditions, while nectarivores, omnivores and ball/cup nesting species were positively affected. Thus, while the single-tree selective harvesting method used by informal harvesters largely emulated natural canopy disturbances, the harvest-mediated increase in the frequency of canopy gaps may reduce the abundance of certain bird species sensitive to canopy gap conditions and reduce forest-scale beta-diversity.

The role of canopy gaps in tropical dry forest (TDF) dynamics remains unclear. Here, 75 canopy gaps, mostly formed by the fall of Bursera spp. and Pachycereus pecten-aboriginum individuals, are described, and their potential consequences for forest regeneration are analysed in a Mexican TDF. In 50 randomly selected gaps, understorey vegetation was sampled with a paired design (inside and outside gaps) and by distinguishing two plant height categories. In total, 1940 plants were recorded (63% in gaps and 37% in non-gap plots). Community attributes (density, community cover, taxonomic richness and Shannon diversity) were significantly higher for both height categories in gap plots. Conversely, neither an NMDS ordination nor a multinomial classification of 187 species by habitat affinities revealed floristic segregation between gaps and non-gaps; almost all species were classified as habitat generalists, with only a few opportunistic forbs (but no single tree species) being classified as gap specialists. The most important effects of gap formation are significant increases in plant abundance and species richness, but not a different species composition. Against earlier views that gap-phase dynamics is inconsequential for TDF dynamics, these results suggest a more active, albeit modest, role of treefall gaps in TDF, through promoting an abundant establishment.

Canopy gaps create a temporary spatial heterogeneity, often allowing pioneer species to establish and grow in mature forests. In this study, we asked whether the above model holds for tropical dry forests in the Florida Keys. Six hundred and forty-eight canopy gaps in an extensive Key Largo forest were identified with a LiDAR digital canopy model. The structure and composition of juvenile trees were examined in 45 selected gaps in three stands of known age, and weighted averaging calibration and regression were applied to the data to determine the successional age optimum for each tree species, and the inferred age for each gap based on its sapling composition. Less than 1% of the forest area was recorded as canopy gaps in the LiDAR model. The inferred stand ages were about 70 y greater in canopy gaps in young forest than in the surrounding, unimpacted forest. This suggested that gap formation advanced succession rather than reversing or resetting it. The apparent lack of recruitment by early-successional species may be due to the small size of canopy gaps in this forest, and the minimal contrast between gap and understorey environments; light and water conditions in the small gaps may favour late-successional rather than pioneer species. Establishment of pioneer species may not take place without intense, large-scale disturbances such as fires and hurricanes that remove the entire canopy and consume or erode soils.

Present theories of tropical rain forest regeneration dynamics suggest that different tree species specialize on the microclimates of different sizes of canopy gap. A detailed analysis of the microclimates of closed forest and 10 canopy gaps of different sizes was carried out in lowland dipterocarp rain forest, Sabah, Malaysian Borneo. It reveals that gaps exhibit considerable spatial and temporal variation in microclimate. As a consequence the relationship between microclimate and gap sizes is not simple. Gaps of the same size do not necessarily have the same microclimate and may favour the growth of different seedling species. Chance location and timing of gap creation play an important role in regeneration dynamics.

The effects of forest habitat, canopy light condition, vertebrate herbivory and species mean seed size on growth of tree seedlings were evaluated for six widely different species of North Queensland tropical rain forests. Two forest localities differing in size and rainfall intensity were used for the trial. In each forest, seedlings were transplanted three weeks after germination into small to medium-sized canopy gaps and into the forest interiors, with half protected by cages and the other half unprotected. Growth measurements were made over a period of 16 months. All growth parameters examined differed significantly between the six species. At the end of the study period, for most species, forest site and protection from vertebrates did not affect seedling biomass. For all species, growth was higher in gaps than in forest interior, but most biomass allocation patterns did not differ between the two habitats. This was attributed to the small difference in photosynthetically active radiation (PAR) levels between the two habitats (interior, 0.48–2.53% PAR; gap, 3.58–7.09% PAR). Between species, seedling growth in the forest interior and sensitivity to increasing light were significantly correlated with initial mean seed size. The growth ability of the six species in and out of canopy gaps is discussed in terms of regeneration status of forest tree species.

The canopy disturbance regime and the influence of gap methods on the interpretation of forest structure and dynamics were evaluated in a tropical semi-deciduous forest in south-eastern Brazil. We encountered a gap density of 11.2 gaps ha−1 and an average size which varied from 121 to 333 m2 depending on the gap delimitation method considered (minimum gap size was 10 m2). Although average size was slightly higher, the median value obtained (78 m2) was comparable to other tropical forest sites and the gap size-class distribution found supported the pattern described for such forest sites. Among 297 gap makers, snapping and uprooting were the most common modes of disturbance. The number and basal area of gap makers were good predictors of gap size. Almost 25% of all gaps suffered from repeated disturbance events that brought about larger gap sizes. Such processes, along with delimitation methods, strongly influenced the estimation of turnover rate and therefore the interpretation of forest dynamics. These results demonstrated the importance of further studies on repeated disturbances, which is often neglected in forest studies.

We examined effects of experimental manipulations of vegetation cover on recruitment, mortality and density of seedlings (20–100 cm tall) and saplings (≥100 cm tall) of woody growth forms over a 2.5-y period. We created four treatments in each of three 15–20-y-old tropical forest stands in Costa Rica: a large canopy gap (270–350 m2), a small canopy gap (50–100 m2), understorey vegetation removal, and an unmanipulated control treatment. Creation of canopy gaps, especially large ones, increased first-year recruitment and density, as well as overall mortality of seedlings. Saplings experienced lower mortality and more prolonged gap-enhanced recruitment and density than seedlings. Removal of understorey vegetation had little or no effect on tree seedling and sapling dynamics. Recruitment and density of lianas responded only to large gaps, whereas understorey species responded to both gap treatments and to spatial heterogeneity within gaps. Tree species exhibited diverse regeneration requirements, whereas liana and understorey species were more specialized to the high and low ends of the light availability gradient, respectively. Canopy gaps provide a critical mechanism for regeneration of lianas, and canopy tree species that dominate during the early stages of secondary forest succession. The choice of management system for these secondary forests can determine the direction and rate of succession.

The effect of natural canopy gaps on the performance of naturally occurring tree seedling populations was studied in a tropical rain forest in French Guiana. This was done at two levels of scale. Firstly, on a 20 m × 250 m forest transect intersecting four canopy gaps, it was investigated how patterns of recruitment, growth and survival of seedlings of Cecropia obtusa, Dicorynia guianensis and Pourouma bicolor differed between canopy gaps and closed forest. Secondly, for one large natural canopy gap, performance of seedling cohorts established before (‘pre-gap’ cohorts) and after (‘post-gap’ cohorts) gap formation was studied in relation to environmental heterogeneity. The direct site factor (DSF) was used as an indicator of light availability in the gap zone. Cecropia specialised in large gaps, and also in specific sites within the large gap: seedling performance was increased by light, and by dead wood. Pourouma mainly germinated under closed forest conditions, where it can survive for a long period until a canopy gap is formed nearby. Pourouma seedlings adjusted well to the new gap environment but they did not specialise in specific places within the gap. Seedling location of Pourouma seedlings was mainly determined by accidental positioning before gap formation. Dicorynia was able to germinate both in gaps and under closed forest conditions. Although the number of gaps studied in this study was low, the results show that seedling establishment, early growth and survival of the three tree species were affected by canopy gaps and by the environmental heterogeneity within a large gap.