Search

Aucoumea klaineana Pierre is the most exploited timber species in Central Africa. In natural forests, its regeneration is hindered by various factors, including limited light availability and pest attacks. To address these challenges and develop effective forest management practices, a comprehensive understanding of A. klaineana seedling ecology is necessary. This study investigated the light requirements and pest resistance of A. klaineana seedlings under different light conditions through an 18-month experiment conducted in Gabon. Six shade houses with varying light intensities were constructed, ranging from 1 % to 62 % of full irradiance, along with an unshaded platform representing 100 % light. Six-month-old seedlings were planted among shade houses and platform and monitored monthly for growth parameters (relative growth rate in height and diameter), morphological parameters (number of leaves, both total and compound) and mortality. Productivity (total seedling biomass), biomass allocation traits (leaves, roots, and stem mass ratios), number of branches, and symptoms of pests attacks were assessed after 18 months. reveal that light conditions influence seedling performance, with optimal levels for biomass and relative growth rate in diameter and height identified between 9 % and 62 % of relative irradiance. However, pest infestation, particularly by psyllids and black canker, poses substantial threats to seedling growth, health, and form, regardless of light conditions. Pest attacks had a significant impact on growth but not on survival, which remained high (97.1 %) even under extreme lights conditions (1 % and 100 % of relative irradiance). The findings underscore the importance of integrating pest management strategies and adapting silvicultural practices to meet the species' ecological requirements. We suggest planting A. klaineana in small groups (to allow root anastomoses), separated by other species (to limit pest contamination), under light canopy cover (around 9–62 % of relative irradiance, to promote sustained growth rates at the seedling stage). This approach should ensure the conservation of A. klaineana populations and the long-term viability of Central Africa's timber industry. • Okoumé seedlings can survive for at least 18 months in extreme conditions (1 and 100 % of relative irradiance). • Optimal light levels (9–62 % of relative irradiance) enhance seedlings growth. • Pest infestations threaten seedlings health and growth, despite high survival rates. • Ecological needs of the species should be better integrated in silviculture. • Emphasis should be placed on managing psyllids and black canker. [ABSTRACT FROM AUTHOR]

• The timber species Cylicodiscus gabunensis experiences local regeneration deficits. • Comparison of spatial, demographic, and genetic structures in different populations. • Contrasted population characteristics and heterogeneity of reproductive success. • Special precautions are necessary to ensure the sustainable management of the species. Most Central African rainforest canopies consist of light-demanding tree species that hold high commercial value but also suffer locally from regeneration deficits, raising concerns about the sustainability of logging. Regeneration is influenced by factors such as past perturbations (including human activity), mating systems, and seed/pollen dispersal processes that impact demographic, spatial, and genetic structures within populations. To gain a better understanding of these interactions, we studied the spatial distribution and trunk diameter structure of Cylicodiscus gabunensis (Fabaceae) - a wind-dispersed, insect-pollinated, timber species - in three plots ranging from 400 to 839 ha situated in various environmental contexts (e.g. forest types and elephant densities) across Central Africa. We also genotyped adults and juveniles using microsatellite markers to analyze the spatial genetic structure of each population and infer the selfing rate, seed and pollen dispersal capacities and selection gradients using the 'neighborhood model'. The selfing rate was low (3 – 4 %), and seed dispersal distances (ds = 184 m) were much shorter than pollen dispersal distances (dp > 2 km). The three populations displayed contrasted spatial, demographic and genetic structures. One population showed no spatial aggregation or genetic structure, and a multimodal diameter structure indicating pulses of regeneration events. Two populations showed strong spatial aggregation and genetic structures. One exhibited a unimodal diameter structure indicating one ancient pulse of regeneration, while the other displayed a 'reverse J-shaped' diameter structure, typical of ongoing regeneration. In the latter, reproductive success appeared leptokurtic, three mother trees accounting for over 90 % of the regeneration and no tree below the minimum cutting diameter implemented by logging companies had offspring. The idiosyncratic nature of population characteristics observed in C. gabunensis suggests that, for sustainable management, a nuanced approach is needed. This involves protecting productive seed trees in areas where natural regeneration is occurring and actively supporting regeneration in areas exhibiting deficits, especially in contexts with low elephant densities. [ABSTRACT FROM AUTHOR]

• Land use type influences canopy openness but not the disturbance gradient. • Land use type little impacts species composition. • Species densities and frequencies vary across vertical strata and land use type. • The seedling stratum is less similar to the tree stand than the sapling stratum. • Light-demanding species require silvicultural practices to ensure their long-term survival. The dynamics of tropical forests are highly dependent on the forest understorey. Understanding its diversity and composition is essential to ensure sustainable management of forest ecosystems. In this study, following a gradient of human disturbance, we analysed the density, diversity and floristic composition of the forest understorey in three land use types of south-eastern Cameroon: a protected area, a selective logging concession and community forests. Vegetation was inventoried in thirty-three permanent one-hectare plots. Woody individuals (tree, shrub, liana) were classified in four strata depending on their diameter: seedling (dbh < 1 cm), sapling (1–9.9 cm), small trees (10–29.9 cm), large tree (≥30 cm). A total of 4,159 individuals (dbh < 10 cm) belonging to 418 species and 232 genera were recorded. They mainly represent tree species (65%) and shade tolerant plants (77%), with diaspores disseminated by animals (74%). Life forms and modes of dissemination were significantly different between land use types. Abundance, species richness and diversity of juveniles varied significantly between strata and land use types: they were globally the highest in plots that have been logged less than five years before inventory. In terms of species composition, the different land use types were not clearly separated, indicating some resilience to disturbance. A low similarity between strata was observed. In all types, most canopy species are light demanding and cannot regenerate in the understorey. This indicates a natural evolution of the forest stands in the absence of silviculture. The presence of species that currently dominate the canopy of these forests will be gradually reduced without human intervention. [ABSTRACT FROM AUTHOR]

Highlights • Growth variability of T. scleroxylon is high but ignored in most management plans. • Logging induces a long-lasting growth stimulation for at least 10–15 years. • Harvestable timber stocks are formed by trees growing faster than population average. Abstract The sustainability of the polycyclic logging system in tropical forests has been increasingly questioned for a variety of reasons, and particularly in central Africa as commercial species, mostly light-demanding long-lived pioneer species, usually fail to recover a stable number of large trees after exploitation. Several factors are known to affect tropical tree demographic processes, like tree growth, survival and recruitment. Tree growth has particularly been showed to depend on ecological conditions, tree genetics, and competition with surrounding vegetation, as well as tree size or ontogeny. Yet, due to the paucity of available data, the importance of such factors is unclear and usually ignored when estimating future timber yields. To fill this gap, we chose to evaluate the variability in growth of one African long-lived pioneer and commercially very important species: Triplochiton scleroxylon K. Schum, gathering a broad dataset composed of tree ring data recorded in one site in Cameroon and periodic field inventory data recorded in seven sites across central Africa. In total, we analyzed 13,225 records of annual tree diameter increments recorded over 920 trees from seven sites in Cameroon, Republic of the Congo and Central African Republic. We evaluated (i) to what extent the average growth of trees that reach harvestable dimensions differs from population average and (ii) to what extent past perturbations influence the growth of remaining trees. We found the diameter growth of T. scleroxylon to be remarkably variable and this study provided an unprecedented quantification of the magnitude of some key growth determinants. In unlogged forests, the diameter increment of T. scleroxylon ranged between 0.40 cm year−1 in Southern Cameroon and 0.83 cm year−1 in South-Eastern Cameroon. The diameter increment was weakly related to tree size but increased twofold from unlogged to logged forests. Perturbation caused by logging stimulates growth of T. scleroxylon for at least 10–15 years. Finally, harvestable timber stock of large-sized T. scleroxylon was found to be constituted by trees that grew in average twice faster than trees of the entire extant population. As more and more inventory data become available, quantifying these effects could be replicated for other timber species and in other sites, to improve the accuracy of future timber resource estimates and improve forest management guidelines. [ABSTRACT FROM AUTHOR]

Highlights • New plots in Celtis and Manilkara forests reveals responsive tree allometry. • Basal area explains the between- and within-AGB variation. • Wood density is an important AGB determinant integrating between-site variation. • The relationships between architectural attributes and AGB are site-specific. • Compensation exists among structural, composition and architectural attributes. Abstract Tropical forests play a key role in regulating the terrestrial carbon cycle and climate change by storing a large amount of carbon. Yet, there is considerable uncertainty about the amount and spatial variation of aboveground biomass (AGB), especially in the relatively less studied African tropical forests. In this study, we explore the local-scale variation and determinants of plot-level AGB, between and within two types of forests, the Celtis and Manilkara forests, growing under the same climate but on different geological substrates in the northern Republic of Congo. In each forest site, all trees ≥10 cm diameter were censured in 36 × 1-ha plots and we measured tree height and crown size using a subsample of 18 × 1-ha of these plots. We developed height-diameter and crown-diameter allometric relationships and tested whether they differed between the two sites. For each 1-ha plot, we further estimated the AGB and calculated structural attributes (stem density and basal area), composition attributes (wood density) and architectural attributes (tree height and crown size), the latter being derived from site-specific allometric relationships. We found strong between-site differences in height-diameter and crown-diameter allometries. For a given diameter, trees were taller in the Celtis forest while they had larger crown in the Manilkara forest. Similar trends were found for the sixteen species present in both forest sites, suggesting an environmental control of tree allometry. Although there were some between-site differences in forest structure, composition and architecture, we did not detect any significant difference in mean AGB between the Celtis and the Manilkara forests. The AGB variation was related to the heterogeneous distribution of large trees, and influenced by basal area, height and crown dimensions, and to a lesser extent wood density. These forest attributes have strong practical implications on emerging remote-sensing technologies for carbon monitoring in tropical forests. [ABSTRACT FROM AUTHOR]

• Secondary logging roads and skid trails are not avoided by large mammals. • In specific conditions, forest elephants and bongos are more seen on secondary roads. • Forest buffalos prefer to travel on secondary roads. • Gorillas and chimpanzees evenly use secondary roads, skid trails, and elephant paths. • Closing roads after logging is crucial for the protection of wildlife. In the literature, roads are often considered to be barriers for large vertebrates. In central Africa, the creation of roads and skid trails by logging operations leads to changes in the structure of forest landscapes that could influence wildlife movements. To assess the use of logging roads by six emblematic species of the central African forests, we conducted a camera trap (CT) survey on three types of tracks (secondary roads, skid trails, and elephant paths as control) in a logging concession of south-eastern Cameroon. The relative abundance indices (RAI) of each species derived from the CT data were used in a mixed linear model to test the effects of four factors (type of track; time: day vs. night; gregariousness: alone vs. group; and time after logging: less than one year vs. between one and two years). The results showed no preference for any type of track for gorillas (Gorilla gorilla) and chimpanzees (Pan troglodytes). In contrast, significantly higher RAI on secondary roads were observed for buffalos (Syncerus caffer), forest elephants (Loxodonta cyclotis), and bongos (Tragelaphus eurycerus). For the last two, the difference was only significant if they were detected in the most recently logged areas, at night (for elephants), or alone (for bongos). We could not test for leopard (Panthera pardus) as we captured only two events. Although none of the species appear to avoid roads and skid trails, nor do they perceive them as a barrier, further studies should be conducted to increase sampling efforts over time and space to consider seasonality, vegetation growth after logging, geographical variability, and other anthropogenic influence. However, these first results reveal the importance of closing roads after logging to limit encounters between wildlife and humans and highlight the relevance of characterizing roads (type of road, width, surfacing, and canopy structure over the road) when studying their impact on wildlife. [ABSTRACT FROM AUTHOR]

This study evaluates the characteristics of soil seed bank in two types of central African rainforests: Celtis forest on clay soils and Manilkara forest on sandy soils. In each study site, 30 samples were collected per soil layers (litter, 0–5 cm, 5–10 cm and 10–20 cm depth). The species diversity and abundance of the soil seed bank were estimated after soil samples were brought to germination. We globally observed 297 seedlings of 53 species for the Celtis forest and 222 seedlings of 39 species for the Manilkara forest. The total densities of germinated seeds were 330 seedlings m −2 and 247 seedlings m −2 , respectively. Herbaceous species dominated with percentages of 41.0 and 45.3%, respectively in the Manilkara forest and the Celtis forest. Both forest types displayed a regeneration potential through the soil seed bank. However, this potential seems higher in the Celtis forest. Pioneer taxa were more abundant in the soil seed bank of the Celtis forest (13 woody pioneer species) than the Manilkara forest (9 woody pioneer species). The values of Sorensen similarity index between the standing tree vegetation and the soil seed bank in each site were relatively low: 11.0% for the Celtis forest and 8.8% for the Manilkara forest. But these similarity values were higher when only the pioneer species were considered: 46.8% in the Celtis forest and 38.9% in the Manilkara forest. The highest species richness were obtained in the first two soil layers (0–10 cm depth) while 21.8% and 21.4% of the species were exclusively found in the deepest layer (10–20 cm) in the Celtis forest and the Manilkara forest, respectively. Among the timber species found in the forest, only three were observed in the soil seed bank of the two sites: Nauclea diderrichii , Erythrophleum suaveolens and Staudtia kamerunensis . N . diderrichii was particularly abundant in the soil seed stock of the two sites (14.4–34.4 seeds m −2 ). Results suggested that to improve regeneration of timber species, planting in open forest habitats with seedlings coming from tree nursery should be more efficient. [ABSTRACT FROM AUTHOR]

• Seeds dispersed by duikers are orthodox pioneer colonisers. • No significant impact of selective logging on duiker occupancy. • Dispersal role of the duiker community is not altered by certified logging. • Studied frugivores play overlapping rather than complementary dispersal roles. • Duikers disperse species logged for timber production. Seed dispersal by frugivores is a key process which is necessary for the regeneration of plant communities. The decrease in the abundance of fruit-eating mammals due to anthropogenic pressures impacts on the proper functioning of forest ecosystems. In this study, for the first time we evaluate seed dispersal by duikers (Bovidae, frugivores and the most hunted animals of the rainforests of central Africa: Congo basin), using germination trials of seeds contained in their dung and rumen. The influence of selective timber harvesting on this dispersal was also examined by comparing duiker community occupancy between a logged and an unlogged site. Our results revealed a community of four duiker species Cephalophus silvicultor , Philantomba congica and two species of 'red' duiker (C. callipygus and C. castaneus). A total of 79 plant species (5481 seedlings) were found in the dungs of all duikers and a total of 37 species (5225 seedlings) were observed in the rumens of red duikers and P. congica. Phyllanthus sp. and Musanga cecropioides were the dominant taxa in dung and rumen contents respectively. Taxa of high commercial value such as Erythrophleum suaveolens , Milicia excelsa and Nauclea diderrichii were also observed. The lack of a clear difference between the plant communities dispersed by the duiker species (assessed by β-diversity) showed that they play redundancy rather than complementary dispersal roles. Duikers play an underestimated role in the restoration of environments disturbed by logging or shifting cultivation as most of the species observed are pioneer colonizing species. Furthermore duiker communities appear to be affected little by selective logging. [ABSTRACT FROM AUTHOR]

• Cylicodiscus gabunensis is an important timber species with a poor regeneration in evergreen forest. • This study aims to define the conditions needed for optimal regeneration and silviculture of C. gabunensis in Gabon. • Establishment of the studied population coincides with the abandonment of agricultural land 148 years ago. • Cylicodiscus gabunensis stems are located on soil with relatively high content in Fe, K, P and total nitrogen. Despite the implementation of management plans, commercial tree species densities are declining in the forests of Central Africa. In the region, Cylicodiscus gabunensis Harms (Fabaceae-Caesalpinioideae; common name 'okan'), is one such species most exploited, but its ecology remains poorly understood. The rarity of its regeneration in evergreen forest suggests that, like other commercial light-demanding species, the conditions that allowed populations to become established are no longer present. Using a combined archaeobotanical and pedological approach, the aim of this study is to identify the factors explaining the current distribution of C. gabunensis individuals at local scale. Within a plot of 1050 ha in a forest concession in south-eastern Gabon, we installed 40 archaeological pits equally divided between sites with and without C. gabunensis. The artefacts encountered were collected and analysed. Charcoal masses were quantified and 18 charcoals were dated. These ages were compared with the average age of the tree population, using growth data from 50 individuals and heartwood dating from 4 individuals. An analysis of the physico-chemical properties of the soil was carried out on composite samples from each archaeological pit. Pottery sherds were found in two pits while charcoal was present in all pits, suggesting widespread human occupation and fire throughout the study area. Human occupation occurred in two phases: between 2480 and 1010 BP and from 590 to 80 BP. The abandonment of agricultural land at the end of this second phase could coincide with the establishment of the C. gabunensis cohort whose average age has been estimated at between 90 and 148 years. Soil analyses showed that C. gabunensis individuals were located on soils that were comparatively richer in element potentially toxic (Fe) and in some plant nutrients (K, P) and total nitrogen. The current scarcity of young trees argues for the implementation of a silviculture that integrates the light requirements of the species as well as the chemical fertility of the soil. [ABSTRACT FROM AUTHOR]

Accurate height-diameter allometry is crucial for the estimation of forest biomass and carbon stocks. Tree height measurements over a large range of diameters and species are urgently needed in the tropics, specifically in central Africa, for the development of locally derived height-diameter allometric equations and the conversion of forest inventory data into biomass estimates, and for the validation of remotely sensed canopy height that mostly rely on a few specific field sites. In this study, we aimed to identify the variation in height-diameter allometry of tropical trees between forest types and among species in central Africa, and we examined the consequences for biomass estimation. Height and diameter were measured for a total of 521 trees over a large range of diameters in two forest types in southern Cameroon, 10–240 cm in the evergreen forest and 11–182 cm in the semi-deciduous forest. A total of ten allometric models including asymptotic and non-asymptotic models were fitted to the height-diameter data. Measured tree diameters, grouped into 10 cm wide diameter classes up to 150, from commercial forest inventory data (0.5 ha plots, n = 2101 and n = 5152, respectively in the evergreen and in the semi-deciduous forests) were converted into biomass estimates using general allometric models with and without including our site-specific height-diameter allometry. Though debated in the literature, our results supported a saturation of tree height with tree diameter both at site and species level, with asymptotic models better depicting the height-diameter allometry. Height-diameter allometry significantly differed between forest types and these local height-diameter equations also differed from published equations. For a given diameter, trees tended to be taller in the semi-deciduous forest than in the evergreen forest, as already reported between moist and wet forests in pantropical studies. Similar trends were reported within species for the three species shared by both forest types, suggesting an environmental control of tree allometry. Because of the low performance of the bioclimatic stress variable to predict tree height and of the slight soil differences between the two forest types, the environmental determinants of height-diameter allometry remain to be explored. In addition to tree allometry variation, structural differences (basal area and density) were also identified between the two forest types using commercial forest inventory data at genus level, and both allometry and forest structure (taller trees and denser stands) contributed to the greater biomass per hectare of the semi-deciduous forest. [ABSTRACT FROM AUTHOR]

Selective timber production is implemented over a large proportion of the world’s tropical forests. Within these forest ecosystems, the frugivore community contributes important ecological services through animal-mediated seed dispersal. However, there is no clear understanding of the impacts of logging on wildlife or of the extent to which seed dispersal is preserved in logged forests. Given its tendency for nesting in light gaps, the western lowland gorilla is likely to provide directed-dispersal services to a wide range of tree species. Production forests preserved from poaching have been reported to harbor high densities of gorillas, but the monitoring of gorilla populations subjected to logging is poorly documented. This study investigated gorilla density and nesting behavior after timber exploitation in a logging concession in southeast Gabon. Nest count censuses were performed on line transects, before and after the second felling cycle. Gorilla density dropped from 1.5 (116.7 nests km −2 , 95% CI = 83.4–163.5) to 1.0 (64.9 nests km −2 , % CI = 32.6–129.5) weaned gorilla km −2 4–6 months after logging, then rose to 2.6 (176.3 nests km −2 , 95% CI = 113.5–274.1) gorillas km −2 9–11 months after logging. A consistent preference for nesting in open canopy terra firma forest was observed during all censuses. This study demonstrates the short-term resilience of the western lowland gorilla to selective timber harvesting, and argues that they offer a continued contribution to directed-dispersal services within months after logging. Although a long-term monitoring of gorillas in a logged forest should be undertaken, the role of this species in logged forest recovery is thought to be important. The preservation of gorilla should receive particular consideration by forest managers. [ABSTRACT FROM AUTHOR]

The greater part of the semi-deciduous moist forests of the Congo basin has been given to logging companies for exploitation. In the next decades, very few of these forests will remain intact. In this paper, we aimed to identify large-scale variations in the structure, composition and functioning of African moist forests that could serve as a baseline for both management and conservation purposes. Commercial forest inventory data were assembled for 49,711 0.5-ha plots, covering an area of more than six million hectares, crossing the borders of Cameroon, Central African Republic and Republic of Congo. Floristic composition was analyzed for a subset of 176 genera reliably identified in the field. Three key functional traits of tropical trees: regeneration guild, leaf phenology, and wood specific gravity, were collected at the species level from various sources, and assigned at the genus level. We first investigated the main variations in forest structure and composition, and identified seven forest types based on these variations. Differences in the percentage of pioneer and deciduous stems, and mean wood specific gravity were tested between forest types. Most of the study area was composed of a mosaic of the structural variations of the forests characterized by the occurrence of Celtis (Ulmaceae) species, which are mostly composed of frequent and abundant genera that formed the common floristic pool of the region. Secondary Musanga (Moraceae) forest is located in repeatedly disturbed areas, along roads and around main cities; mixed Manilkara (Sapotaceae) forest covers a huge area in the southern Central African Republic and in the northern Republic of Congo; and monodominant Gilbertiodendron (Fabaceae) forest is sparsely distributed along rivers. The contrasted structure, composition, and functioning of the forest types imply pronounced differences in population and ecosystem processes, and call for adapted management and conservation strategies. [ABSTRACT FROM AUTHOR]

Highlights: [•] African tree species does not differ from common allometric patterns. [•] Pantropical multi-species allometric equations are valid in South-eastern Cameroon. [•] Using wood specific gravity from global data base little impacts biomass estimates. [•] Single-species models are not better than multi-species models. [Copyright &y& Elsevier]

• 21,180 trees have been monitored for 1–7 years in eight sites in Central Africa. • Diameter increments and mortality rates of 42 commercial tree species were estimated. • Trees that are close to the minimum harvestable size tend to show the fastest growth. • Selective logging may stimulate tree growth but it may start 1–2 years after logging. • Tree growth factors are identified; ignoring them can lead to poor projection. Tree growth and mortality are two central processes in mixed and structurally complex moist tropical forests, yet accurate estimates of the variables needed to model them remain sparse and scattered. It is thus still difficult to predict forest evolution at a local scale and build reliable management plans. To help fill this gap, for 1–7 years we annually monitored 21,180 trees belonging to 42 species exploited for timber production in Central Africa. We made new species-specific estimates of diameter increments and mortality rates, and investigated how tree growth varied with tree size and logging history. We compared our results with the legal values of diameter increments, mortality rates, and minimum cutting diameters used to build forest management plans in Cameroon. Diameter increment was found to vary with tree size for most of the species studied. The relationships between diameter increment and tree size were mostly humpback-shaped. The trees with diameters close to or lower than the reference minimum cutting diameter generally grew faster than the average. We also found that tree growth could slow for 1–2 years after timber exploitation and was then spurred for at least 5 years. The tree growth response to logging was nevertheless species-specific. This study provides new estimates of tree diameter increments and mortality rates that could help make more accurate forest projections and draw up sustainable management plans in Africa. [ABSTRACT FROM AUTHOR]

Abstract: In this paper we seek to identify the floristic determination biases contained in large-scale commercial inventories conducted by logging companies and to determine whether this impacts on the observed patterns of alpha and beta diversity. The study focused on floristic data recently collected by industrial timber companies in the tropical forests of the Central African Republic (28,229 0.5-ha plots spread over 14,000km2). A subset of these plots (n =1107) was later re-sampled for controlling purposes by experienced botanists. The proportion of agreement between the two samplings was assessed for each species and independently for small and large trees, and at genus and family resolutions. Unsurprisingly, large trees and common species were more accurately identified than small trees and rare species. We found that the quality of the floristic determination increased slightly from species to families. We also detected a significant variation between concessions in the quality of the floristic determination that was more dependent on working conditions during forest inventories than on field workers. Contrary to a widespread belief, we did not find a strong bias toward commercial species, showing that commercial inventory data could also be valid for non-commercial species in ecological studies. Finally, we found that both alpha and beta diversity patterns in commercial inventories were highly consistent with those of the re-sampled inventory. This latter result shows that commercial inventories are well suited to detect large-scale patterns of floristic variation. Large-scale commercial inventories could thus play an important role in the identification of large-scale patterns in tropical tree diversity. This could enhance our ability to manage tropical forests by designing representative reserve networks and developing management plans that integrate diversity patterns at the landscape scale. [ABSTRACT FROM AUTHOR]

Abstract: Studies of regeneration in African rain forests suggest that without silvicultural treatments, natural succession in logging gaps may not result in the establishment of timber species. In this paper we present the results of an experimental enrichment planting with moabi (Baillonella toxisperma Pierre), a valuable and important timber species harvested in Central Africa. Although forest gaps are generally considered as favourable for the regeneration of this species, a survey conducted in a forest concession in south-eastern Cameroon provided an estimate of only 12.7seedlingsha−1, suggesting that the species was, in fact, poorly represented in logging gaps within the study area. To further investigate the dynamics of the moabi in logging gaps, 795 seeds were sown in 15 logging gaps and 410 nursery-raised seedlings were planted in 15 other gaps. A biannual monitoring program over a 30-month period showed a lower survival rate for seedlings from sowing (75.9%) compared to that of nursery-raised seedlings (95.3%). Planted seedlings reached an average of 229.3cm tall whereas seedlings from sowing were 167.5cm tall, with the observed difference roughly corresponding to the average height of the nursery-raised seedlings at the time they were introduced to the logging gaps. After 30 months, the diameters of planted seedlings (16.8mm) were also greater than those of the directly sown individuals (12.5mm). Forest gap characteristics significantly influenced the growth of the plants. Factors accounting for the differences were total solar radiation, the soil content of coarse sand, the topographic position of the gap, the vegetation cover and the density of Macaranga spp. Whilst total solar radiation had a positive influence on growth, the remaining factors had impacted growth negatively. A streamlined technique was tested by planting 7 seedlings in 250 gaps. Without additional site maintenance, 29.3% of the moabi seedlings emerged naturally from the competing vegetation after 24 months. With biannual maintenance some 89.4% of seedlings became successfully established. Clearance operations had no significant influence on the height of plants whilst plant diameter was greater in cleared gaps. The total cost of the enrichment technique was 5.5EUR per gap without maintenance and 7.5EUR per gap with a single maintenance measure. Whilst long-term monitoring is needed, this study suggests a high survival rate of moabi introduced in logging gaps, and a growth rate 10 times higher than previously reported under canopy cover. These findings, combined with the low costs of the enrichment technique, support the use of silvicultural measures in logging gaps to restore the forest. [Copyright &y& Elsevier]