Journal of
Ecology and The Natural Environment

  • Abbreviation: J. Ecol. Nat. Environ.
  • Language: English
  • ISSN: 2006-9847
  • DOI: 10.5897/JENE
  • Start Year: 2009
  • Published Articles: 411

Full Length Research Paper

Floristic and structural traits of tree vegetation in three sites with different level of disturbance in dense humid forest of Cameroon

Jules Romain Ngueguim
  • Jules Romain Ngueguim
  • Institute of Agricultural Research for Development (IRAD). P. O Box 77 Limbe-Batoké Cameroon.
  • Google Scholar
Marie Caroline Momo Solefack
  • Marie Caroline Momo Solefack
  • Laboratory of Applied Botany, Department of Plant Biology, Faculty of Science, University of Dschang, Dschang, Cameroon.
  • Google Scholar
Jean Lagarde Betti
  • Jean Lagarde Betti
  • Department of Plant Biology, Faculty of Sciences, University of Douala, Cameroon.
  • Google Scholar


  •  Received: 17 February 2018
  •  Accepted: 23 October 2018
  •  Published: 30 November 2018

 ABSTRACT

This study characterizes the floristic and structural traits of trees in three sites of rainforest of Cameroon. The sites were exposed to different intensities of disturbances: Mangombe, highly and frequently disturbed; Bidou, moderately disturbed and Campo, undisturbed. Data collection were carried in a disposive comprising 65 plots of 20 m × 20 m randomly installed in each site for inventory, identification and measurement of diameter of trees greather than 10 cm at 1.30 m height. In total, 4717 plants belonging to 130 species and 43 families were recorded in all the sites. Basal area showed a declining trend with the increase in disturbance intensity while tree density, species richness and families increase with disturbances: Campo (87 m²/ha, 569 trees/ha, 75 species and 29 families); Bidou (54, 538, 88 and 32); Mangombe (49, 708, 91 and 38). The vegetation indices showed a high diversity in all the sites. The Shannon index (5.40 to 5.52) and generic diversity (1.10 to 1.16) had greater value. According to the sample, the floristic composition of Mangombe considerably differs from those of Bidou and Campo. The undisturbed site (Campo) contained young tree population showing a vigorous regeneration while in the highly disturbed site (Mangombe), tree density was scarce, with few big size trees having high cultural importance and low economic value. Low shrub density was recorded in Mangombe and Bidou due to frequent human disturbances. Canopy gaps favor direct sunlight which enhanced the abundance of Shrub in all the sites. More protection is needed for the restoration in the long term of forest cover in Mangombe, which can be done naturally due to high density of small trees composed of species generally found in the upper strata.

 

Key words: Cameroon, dense humid forest, disturbance, species richness, structure of population.


 INTRODUCTION

Forest ecosystem offers various ecosystemic services for humanity. Its  plays  an  important  role  for the survival of
 
rural communities. These anthropogenic disturbances in tropical forests are old and date from the beginning of their occupation by men (Chazdon, 2003). Therefore the dense humid forest of Cameroon presents heterogeneity in the spatial distribution of patches of disturbed and undisturbed forest. The studied sites were formerly inhabited and cultivated during a century, then gradually abandoned with old and even contemporary reoccupation; making disturbance old in this ecosystem (Letouzey, 1968). A selective cutting of wood and harvesting of non-timber forest product were made in the sites at different period, intensity and frequency. Mangombe was highly and frequently disturbed; Bidou, moderately disturbed and Campo, undisturbed. Various studies attest that anthropogenic activities constitute one of the major threats to the conservation of tropical forest. Soft logging exploitation does not exist, but reduces the diversity and richness of forest in wood species with high economic value. Logging exploitation also involves damage on the surrounding forest. These disturbances can affect the floristic and structural threat of ecosystems (Thapa et al., 2010; Naing et al., 2011). Other authors think that disturbances related to moderate human pressure cannot necessarily affect the species richness of forest because it create new environmental condition which sometimes increase the flora diversity (Budke et al., 2010). The comparison of the vegetation analysis would help in understanding the effects of disturbance on the composition and dynamics of forest community and also help in managing the Mangombe and Bidou secondary forests more exposed to anthropogenic disturbance.

 


 MATERIALS AND METHODS

Description of study area
 
This study was conducted in the southwestern dense humid forest of Cameroon, in the middle of the Biafran Rain Forest. The area is an important site of the Guineo Congolian regional centre of endemism. The sites especially Campo is composed of many forest types with species of high conservation priorities (114 endemic species of which 29 are only found in the area and rich fauna). Despite his importance, the area is under human pressure which leads to the degradation of coastal forest and depletion of lowland forest. The main conservation effort was the creation of community council forest in Mangombe and a national park in Campo (Tchouto, 2004).
 
Sampling and data collection methods
 
A set of 65 plots of 20 m × 20 m were randomly installed in each site, a total cover of 2.6 ha; for inventory, and measure of tree (using a meter) with diameter greater than 10 cm at 1.30 m height. The use of plots has many advantages; it helps to highlight the different phenomena of forest dynamics, to intensify the study of the milieu and study the composition and structure of a forest ecosystem (Picard, 2007).
 
Data analysis
 
Floristic analysis
 
Specific Richness (SR): The specie richness is the total number of species observed. The Area - Species curve and Hurlbert curve (Abundance - Species) show evolution of the species richness in relation with the sampled area and help to describe the forest. This approach was used by several authors to estimate extinction rate of species in rain forests (Reid, 1992; Jha et al., 2005).
 
Diversity of families and genera: The diversity of families is the expression in percentage of appearance of each of the inventoried families.
 
E / G                                                                                              (1)
 
where E represents the number of species and G the number of genera which helps to appreciate the floral diversity. A low value (close to 1) of this equation indicates a strong diversity of the flora, but does not inform about the species distribution between the various genera and families. Analysis of families’ spectrum is important, because rich ecosystems are characterized by few large genera and families rich in species.
 
Important Value Index (IVI): Important Value Index (IVI) helps to determine the place occupied by each tree species within the community according to Curtis's and Macintosh (1950) formula.
 


 RESULTS

Floristic richness, basal area and important value index (IVI) of taxas
 
A total of 4717 trees were counted in the 195 plots which cover 7.8 ha. The average density of trees is estimated to 605 individuals/ha. This density presents a highly significant difference between the three sites (F2, 192 = 20.95, P < 0.001). The density was high in Mangombe (708 individuals/ha), average in Campo (569) and low in Bidou (538). The inventoried trees can be grouped into 43 families, 110 genera and 130 species. Shannon index is high in all the sites. The specific diversity decreases with the level of disturbance. The floristic diversity value is high in Mangombe the most disturbed site (38 families, 77 genera and 91 species), comparing to Bidou (32, 81 and 88) and Campo (29, 68 and 75). The basal area is significantly low at Mangombe (49.13 m²/ha) with regard to Bidou (54.08 m²/ha) and Campo (87.06 m²/ha) (F2, 4714 = 93.21, P <0.001) (Table 1).
 
The Areas - Species curve and the Abundance - Species Richness curve of Mangombe and Bidou are very close. They belong to the same reliable intervals and both present a regular growth more important than that of Campo, characterized by the recruitment of a high number of species when the number of trees and the inventoried area increase. The site of Campo is the least disturbed and less diversified comparing to others, with reliable intervals which do not recover those of Mangombe (Figure 1). In terms of their IVI, the five most important families and species are respectively estimated at 57 and 31% in Mangombe, Bidou (54 and 34%) and Campo (61 and 30%) (Table 2).
 
Floristic composition
 
Abundant species are generally more frequent and more distributed. This group is composed in Mangombe by: Tabernaemontana crassa Benth (8%), Strombosia scheffleri Engl. (8%), Oncoba glauca (p. Beauv) Planch (6%) and Lophira alata Banks ex C.F. Gaertn (5%). In Bidou, this group includes Dialium guineense Willd (8%), Diospyros crassiflora H. Perrier (8%), Uapaca guineensis Müll Arg. (6%) and Keayodendron bridelioïdes (Gilg and Mildbr. Ex Hutch. and Dalziel) Leandri (6%), whereas in the site of Campo, we have K. bridelioïdes (11%), Anthonotha macrophylla P. Beauv. (6%), D. crassiflora (6%) and Polyalthia suaveolens (Engl. and Diels) Verdc (5%).
 
Table 3 presents the specific composition of genera; it shows that Kola is the most diversified with 5 species. The forests of Mangombe and Bidou are more varied in genus. The number of genus represented by single specie is estimated at 67 genera at Mangombe, 74 at Bidou and 61 in Campo. In all the sites, more than 88% of the genera were composed of single specie. The numbers of genera composed of several species decreases considerably while the number of species per genus increases.
 
 
 
In Mangombe, the more diversified genus are Dacryodes and Strombosia (with three species each), and Kola (with 4 species). In Bidou they are: Afzelia, Anthonotha, Kola, Hallea, Markhamia, Tabernaemontana, Xylopia and Zanthoxylum (with 2 species each) and at Campo the diversified genera are Afzelia, Anthonotha, Kola, Entandrophragma, Tabernaemontana, Xylopia and Zanthoxylum (2 species).
 
In Mangombe, the five abundant families are: Olacaceae (18%), Fabaceae (13% dominated by the Caesalpinioïdeae 10%), Annonaceae (11%), Apocynaceae (10%), and Euphorbiaceae (6%), while in Bidou the group is composed of Fabaceae (19% dominated by the Caesalpinioïdeae 16%), Euphorbiaceae (13%), Annonaceae (9%), Ebenaceae (8%) and Meliaceae (7%), whereas in Campo we have Fabaceae (20% dominated by the Caesalpinioïdeae 16%), Euphorbiaceae (14%), Annonaceae (13%), Meliaceae (9%) and Ebenaceae (6%).
 
The diversified families are among the more abundant. Fabaceae is the more diversified floral group in all the sites and are dominated by the sub family of Caesalpinioïdeae. Other richness families are Euphorbiaceae, Apocynaceae and Annonaceae.
 
Comparison of the floristic composition of Mangombe, Bidou and Campo
 
Comparison of Bidou and Campo: The sites of Bidou and Campo have 63 common species. These common species represent 91% (1275 trees) of the counted trees in Bidou and 94% (1397 trees) of the effective of Campo. With regard to the sampling, this information suggests that there is very little difference between the floral compositions of these two sites. Bidou, still have impact of logging exploitation. The timber species like L. trichilioïdes are little represented with regard to the site of Campo. P. suaveolens used as firewood is less frequent at Bidou than Campo. The population of P. nitida presents gaps between Bidou and Campo, due to exploitation. P. nitida is a medicinal plant, with the bark and the roots used to cure malaria, stomach pains and pneumonia. K. bridelioides is a mesophanerophyte generally abundant in semi deciduous forests, and can be found scattered in dense humid forests. D. guineense is more abundant in Bidou; it is a riparian species found in the undergrown of dense humid forest, and the seeds of this species are consumed by local communities. In the site of Campo, we found emergent trees with bigger diameters.
 
Comparison of Mangombe with Bidou-Campo: The vegetation of Mangombe (91 species, 1840 individuals) is very different from those of Bidou (88, 1398) and Campo (75, 1479), in particular as regards the density, structure of population and floral composition. The Mangombe vegetation has 30 species (574 individuals) different from those of Bidou and Campo, and 44 common species (814 individuals) representing 44% of the species present in Mangombe.  The   common   species  are  dominated  by those familiar to less disturbed forest; they are mostly commercial wood such as: Afzelia pachyloba (Doussié), Canarium schweinfurthii (Aïlé), Coula edulis, Didelotia africana, D. crassiflora, Duboscia macrocarpa, Irvingia gabonensis, Khaya ivorensis, Pachyelasma tessmanii, Pseudospondias longifolia, Saccoglottis gabonensis, Triplochiton scleroxylon, and U. guineensis. The species of secondary forests in this group are: Cleistopholis patens, Dacryodes macrophylla, L. alata, Markhamia tomentosa, Musanga cecropioïdes, Pycnanthus angolensis, Spathodea campanulata, Zanthoxylum gilletii, Zanthoxylum heitzii, and Xylopia aethiopica.
 
Structural analysis of tree population in the different sites
 
Figure 2 represents the ACP based on the projection of the structural parameters for all the 4717 trees listed and distributed in class (Cl) of individuals with amplitude of 10 cm diameter. Cl1 represents the class of the individuals 10 cm ≤ dbh ≤ 20 cm; Cl2: 20 cm < dbh ≤ 30 cm; Cl3: 30 cm < dbh ≤ 40 cm etc. These analyses show on axis 1 an opposition between diversified diameter classes, composed of abundant tree with small diameters, on one hand, and on the other hand classes of individuals with low density and big trees. This confirms the presence of forest patches at different growing stage.
 
In Mangombe, the vegetation is characterized by height trees with small diameters; this can be due to disturbance and canopy gap which favor the growth in height among trees which are in competition for light in the undergrowth. In all the sites, the first class of individuals Cl1 is distinguished from others by the presence of various small sizes trees, while the upper classes are composed of few big trees.
 
In Mangombe, the classes of big diameters (Cl11) distinguish themselves from others by their very low density (5 individuals and 3 species) and the presence of emergent species like Sacoglottis gabonensis (dbh = 135.35 cm), C. edulis (dbh = 318.18 cm). There is a big difference in the abundance of trees between the small size diameter classes Cl1 (1025 trees), Cl2 (429 trees) and Cl3 (202 trees).
 
In Bidou, the class of big trees Cl13 counts only 6 individuals composed of 6 species, such as Staudtia kamerunensis (156.13 cm), D. africana (162.69 cm) and Baillonella toxisperma (238.73 cm), which dominate the landscape by their sizes. The difference in terms of trees abundance and species richness explains the gap between Cl1 (702 ind), Cl2 (276 individuals) and Cl3 (143 individuals).
 
In Campo, the class of the superior diameters is composed of 3 species (3 individuals); they are emergent species like S. gabonensis (207.16 cm), Erythrophleum suaveolens (382.23 cm).
 
 
 
The structures of trees population of Bidou and Campo  are similar to an opposition between groups of plots composed of small trees, dense and diversified which doubtless find themselves in disturbed area; while on the other side there is a group of plots composed of big trees. Also, a group of intermediate plots was observed; and in Mangombe, the appearance of new altitudinal strata was observed.
 
Individualization of group of trees
 
Figure 3 is the HAC obtained from the presence/absence of  tree   species   censored   in  195  plots. It  shows  five groups of trees. The composition of groups indicates that: Group 1 is a mixture at equal proportion of plots of Bidou and Campo; the Group 2 represents the site of Bidou, the Group 3 can be linkened to the site of Campo and the Groups 4 and 5 belong to Mangombe.
According to their floral and structural composition, the groups can be classified in various types of forests.
 
Frequently disturbed patches of forest
 
Group 2 is vegetation in a dynamic state. The most abundant   families  are  Fabaceae  (20%),   followed   by  Euphorbiaceae (17%), Meliaceae (8%) and Annonaceae (8%), whereas the abundant species are in decreasing order: K. bridelioides (9%), D. crassiflora (8%) and D. guineense (7%).
 
Group 4 counts various shade-intolerant species like A. macrophylla, T. pachysiphon, P. suaveolens, D. macrophylla and earlier growing species such as C. patens. These species mark signs of past and present disturbances. The abundant families are: Olacaceae (24%), Fabaceae (15%), Annonaceae (9%), Ebenaceae (9%) and the abundant species are: S. scheffleri (10%), Diospyros kamerunensis (9%), and O. glauca (6%).
 
Moderately disturbed patches of forest
 
The plant grouping 5 is the densest. The abundant families are: Olacaceae (15%), Apocynaceae (12%), and Annonaceae (11%). While the abundant species comprise: T. crassa (11%), L. alata (7%), S. scheffleri (6%), and O. glauca (6%). The higher number of pioneers and shade-intolerant species is an indicator of disturbed forest.
 
Table 4 presents the average biological  parameters  of the groups; we can note that Groups 4 and 5 are diversified, with a strong density and an ascendancy of shrubby stratum (23 cm ≤ dbh ≤ 24 cm), these characteristics shows vigorous forest regeneration. Group 2 is composed of trees with average diameter (29 cm) whereas Group 3 and 1 have big trees.
 
 
Undisturbed patches of forest
 
In Group 1, the abundant ligneous families are: Fabaceae (18%), Annonaceae (10%), Euphorbiaceae (10%) and the frequent species are: D. guineense (6%), D. crassiflora (5%), and U. guineensis (5%).
 
Group 3 possesses the most important families, Fabaceae (20%), Annonaceae (15%), Euphorbiaceae (14%); whereas the frequent species are: K. bridelioides (13%), D. crassiflora (8%), and A. macrophylla (7%).
 
In these plant groupings, the abundance of Meliaceae in the understory indicates the importance of shade-intolerant species which are gradually replaced by tolerant species; this marks signs of past disturbances and the presence of Ebenaceae reminds one of an advanced maturity of the forest.

 


 DISCUSSION

Floristic traits
 
The studied sites are rich and diversified as indicated by their species diversity (75 - 91), Shannon's index (5.40 - 5.52), low value of the generic coefficient (1.10 ≤ E/G ≤ 1.16) and the ascendancy of families and mono specific genera. This wealth characterize rainforests recognized by their favorable environment necessary for their evolution (Stebbins, 1974), for the preservation of biodiversity (Leigh, 2008) and marked by the presence of a high number of genera represented by low number of species.
 
The number of species is high in Mangombe, a situation that results from logging exploitation and other human pressures which increased canopy gaps and favored pioneers and shade-intolerant species, which are added to the species existing in the forest before exploitation. Gaps in the canopy engendered by the death of trees modify the ecological conditions of the forest and constitute the driver of sylvigenesis. Other processes such as the substitution of emergent trees dead from foot and by the dominated trees also contribute to the mechanisms of renewal of the forest.
 
Generally, when environmental conditions are favorable, the natural regeneration which settles down after deforestations, can contain, a few years later, the seedling of all the forest species of the environment with however an important frequency and vitality of pioneers species able to mask the presence of other species (Chazdon, 2008). According to Sahu et al. (2008) and Chazdon et al. (2009), natural disturbances (tree blow downs) or anthropogenous pressures can contribute to biodiversity in the canopy gaps, or constitute a threat (Hanster Steege, 2010; Loreau, 2010). In disturbed area, certain species can easily settle down favoring a fast increase of the biodiversity; however, some of these species cannot sustainably stay in these environments. Therefore, the maturity of the forest can lead to the reduction of specific richness (Chave et al., 2003). Disturbances are not always negative to an ecosystem as it depends on their frequency and intensity. It can contribute   to   the   diversity   richness,   renewal  of  the ecosystem and release of nutrients.
 
The higher value of observed Shannon diversity index corresponds to the conditions where environment is favorable to installation of several species represented by few numbers of individuals. This tendency to a high specific diversity associated with a low density of species is a remarkable phenomenon in rain forest.
 
Specific richness and diversity
 
The Area - Species curve for all the sites shows a regular growth. It presents a classic structure of the recruitment of species when the number of trees increases. These curves are not stabilized; it indicates that additional inventories may increase the number of species. The comparison of the Area - Species curves shows a good correlation between the curve of Mangombe and Bidou where the number of species increase more quickly with the sample area comparing to the site of Campo. These tendencies find themselves in the Abundance - Species curve. The sites of Mangombe and Bidou seem richer and diversified than that of Campo, resulting from the importance and frequency of disturbances.
 
Important value index (IVI) and floral composition of the sites
 
This index shows a general representation of the composition of the vegetation in term of dominant species. Few numbers of species has an important contribution in IVI. Several wooden species are listed among the ten important species by their IVI in Bidou and Campo, viz; U. guineensis, Lovoa trichilioïdes, E. suaveolens, K. ivorensis, D. africana, Eribroma oblonga. At Mangombe, non wood species belong to those with important IVI such as C. edulis Baill., S. scheffleri Engl., O. glauca (P. Beauv.) Planch, T. crassa and L. alata Banks ex C.F. Gaertn which is a secondary forest wooden species.
 
At Mangombe, one can assume that wood exploitation which is selective because it takes only few big trees on a  reduced number of species, affects the specific composition of the forest and the process of reconstitution of the natural environment by increasing or strengthening the abundance of certain less appreciated species.
 
Family richness
 
Fabaceae, particularly the sub families of Caesalpinioïdeae and Mimosoïdeae is well represented in the whole forest sites. One of the fundamental characters of the African dense forests is their higher richness in Fabaceae who is closer to those of America and different from those of Asia (White, 1986). Small size trees are abundantly represented by Anacardiaceae, Annonaceae, Apocynaceae, Ebenaceae, Euphorbiaceae, Guttifères, Icacinaceae, Ochnaceaea, Olacaceae, Rubiaceae, Tiliaceae, Sapindaceae and Violaceae.
 
The floristic composition of lower and superior strata in Campo and Bidou differs from that of Mangombe. At Mangombe, we can observe in the undergrowth the presence of some species generally found in superiors strata like Alstonia congensis Engl, Ceiba pentandra (L.) Gaertn, C. edulis Baill., D. africana Baill., L. alata Banks ex C.F. Gaertn, Milicia excelsa (Welw) CC Berg, and Piptadeniastrum africanum (Hook f) Brenan.
 
Structural trait of the vegetation
 
Trees density
 
The density of trees in the studied forest is situated in the order of the average density of trees observed in dense humid forest estimated at 400 to 650 individuals with dbh ≥ 10 cm per hectare (Riéra, 2011). We can however note a higher density of trees in Mangombe compared with that of many tropical sites; this is due to the reconstitution of the forest after exploitation and the regrowth who settle down and develop after anthropogenic and natural disturbances. The disturbances have a direct influence on trees density, diversity and their specific composition (Hitimana et al., 2004; Kessler et al., 2005).
 
Basal area
 
Trees population in Mangombe is dense and the basal area low (708 trees/ha; 49.13 m²/ha) compared to Bidou (538 trees/ha; 54.08 m²/ha) and Campo (569 trees/ha; 87.06 m²/ha). This situation can be explained by the abundance of big trees particularly in the site of Campo. The low gap between the basal area of Mangombe and that of Bidou can be explained by the effect of the exploitation of trees which is more frequent in Mangombe than Bidou.  The  basal  area  in  Mangombe  is  however biased by the presence of two emergent trees B. toxisperma and C. edulis which have exceptional diameters (dhp = 373 and 318 cm) and contribute to 15% of the basal area.
 
In Mangombe, shrubs are abundant like in the floristic groups GR5 and GR4; in Bidou, trees have a litle big diameter (GR2); whereas in Campo we have many trees with big diameters (GR3). Basal area in Bidou and Campo are higher than the average value (30 - 50 m²/ha) frequently observed in rain forests (Riéra, 2011). Table 5 presents the distribution of basal area in some African forests. We can note that the basal area in Campo is exceptional and probably due to the type of ecosystem which is a forest rich in Fabaceae (Tchouto, 2004).
 
 
Tree and diameter class distribution
 
The structures of population reveal a strong presence of trees with small diameters and little number of big trees, the characteristic of vegetation with constant regeneration and in equilibrium. This forest structure is close to that of Rollet (1974), which showed that the distribution of trees in the diameter classes follows an exponential model. Mangombe forest presents a deficit of big trees due to anthropogenic cutting.


 CONCLUSION

The studied forest sites are rich and diversified; they distinguish themselves by a mosaic of forest patches with different development stages, structure and floristic composition. The study of flora and the vegetation group highlights the role played by anthropogenic pressures in the spatial distribution of numerous taxa’s and plant associations. The similarity in the floral composition of the various forest sites decreases with the gradient of anthropogenic disturbance.
 
The floristic composition and structure of population in Mangombe Forest differ from those of Bidou and Campo which are similar. In Mangombe, trees density is higher and the basal area is lower compared to that of the other sites; this can be due to a high number of small trees and shrubs, which is a sign of strong forest regeneration. In this forest site, a few numbers of species without a commercial value dominate by their IVI, showing that cutting of trees affected the specific composition and the process of reconstruction of the forest by strengthening the importance of little appreciated species. The presence in this group of typical secondary forests species (L. alata, T. crassa) mark the sign of the past and present anthropogenic disturbances. This group is composed at Bidou and Campo of various wooden species such as D. crassiflora, L. trichilioïdes, E. suaveolens.
 
Mangombe   Forest   can   be   compared  to  a   young secondary forest in phase of maturation. The restoration of the forest can be done naturally as shown by trees structure characterized by a high frequency of small size trees and deficit of big trees. Mangombe forest can recover its original structure and functions after a long term and in the absence of anthropogenic disturbance. These will also depend on the behavior of all the species and on the interactions between them and their environment.

 


 CONFLICT OF INTERESTS

The authors have not declared any conflict of interests.

 


 ACKNOWLEDGEMENTS

The authors thank the International Foundation for Science (IFS) who grant this work (D/4685), as well as National Herbarium of Cameroon for sample identification and the Laboratoire d’Ecologie General of the Muséum National d’Histoire Naturelle of Paris for the working environment necessary to accomplish this study.



 REFERENCES

Budke JC, Jarenkow JA, De Oliveira Filho AT (2010). Intermediary disturbance increases tree diversity in riverine forest of southern Brazil. Biodiversity and Conservation 19:2371-2387.
Crossref

 

Chave J, Condit R, Lao S, Caspersen JP, Foster RB, Hubbell SP (2003). Spatial and temporal variation in biomass of a tropical forest: results from a large census plot in Panama. Journal of Ecology 91:240-252.
Crossref

 
 

Chazdon RL (2003). Tropical forest recovery: Legacies of human impact and natural disturbances. Perspectives in Plant Ecology, Evolution and Systematics 6:51-71.
Crossref

 
 

Chazdon RL (2008). Chance and determinism in tropical forest successionIn W Carson and SA Schnitzer (Eds)Tropical Forest community ecology Oxford UK. pp. 384-408.

 
 

Chazdon RL, Harvey CA, Komar O, Griffith DM, Ferguson B, Martinez-Ramos GM, Morales H, Nigh R, Soto-Pinto L, Van Breugel M, Philpott SM (2009). Beyond reserves: A research agenda for conserving biodiversity in human-modified tropical landscapes Biotropica 41:142-153
Crossref

 
 

Hanster S (2010). Will Tropical Biodiversity Survive our Approach to Global Change? Biotropica 42(5):561-562.
Crossref

 
 

Hitimana J, Kiyiapi JL, Njunge JT (2004). Forest structure characteristics in disturbed and undisturbed sites of Mt Elgon Moist Lower Montane Forest western Kenya Forest Ecology and Management 194:269-291.
Crossref

 
 

Jha CS, Goparaju Laxmi Tripathi A, Gharai B, Raghubanshi AS, Singh JS (2005). Forest fragmentation and its impact on species diversity: an analysis using remote sensing and GIS? Biodiversity and Conservation 14:1681-1698
Crossref

 
 

Kessler M, Kessler PJA, Gradstein SR, Bach K, Schmull M, Pitopang R (2005). Tree diversity in primary forest and different land use systems in Central Sulawesi Indonesia. Biodiversity and Conservation 14:547-560.
Crossref

 
 

Leigh EG (2008). Tropical forest ecology: Sterile or virgin for theoreticians? In W P Carson and S Snitzer (Eds)Tropical rain forest community ecologyWiley-Blackwell Oxford UK. pp. 121-142.

 
 

Legendre L, Legendre P (1984). Ecologienumérique : le traitement multiple des donnéesécologiques Tome 1Masson Paris France 260p.

 
 

Letouzey R (1968). Etude phytogéographique du Cameroun Encyclopédie Biologique 69 Ed Paul Lechevalier Paris France 511p.

 
 

Loreau M (2010). Linking biodiversity and ecosystems: towards a unifying ecological theory. Philosophical Transactions of the Royal Societ B365:49-60.
Crossref

 
 

Magurran A (2004). Measuring biological diversityBlackwell Publishing.

 
 

Naing ZH, Nobuya M, Shigejiro Y (2011). Tree species composition and diversity at different levels of disturbance in Popa Mountain park Myanmar Biotropica 43(5):597-603.

 
 

Picard N (2007). Dispositifs permanents pour le suivi des forêts en Afrique Centrale: un état des lieux CIRAD.

 
 

Reid WV (1992). How many species will there be? In: Whitmore TC and Sayer JA (eds) Tropical Deforestation and Species Extinction Chapman & Hall London pp. 55-74.

 
 

Riéra B (2011). La forêttropicale: Architecture et Biomasse Conférence Université de Dschang 8 juin 2011.

 
 

Rollet B (1974). L'architecture des forêtsdenseshumidessempervirentes de plaineCTFT/CIRAD-Forêt Nogentsur Marne 298 p.

 
 

Sahu PK, Sagar R, Singh JS (2008). Tropical forest structure and diversity in relation to altitude and disturbance in a Biosphere Reserve in central India. Applied Vegetation Science 11:461-470.
Crossref

 
 

Stebbins GL (1974). Flowering plants evolution above species levelCambridge Massachusetts Belknap Press of Harvard University Press 399 p.
Crossref

 
 

Tchouto MPG (2004) Plant diversity in a Central African rainforest: implications for biodiversity conservation in Cameroon PhD thesis department of plant sciences Wageningen University the Netherlands 206 p.

 
 

Thapa S, Chapman DS (2010). Impacts of resource extraction on forest structure and diversity in Bardia National Park Nepal. Forest Ecology and Management 259:641-649.
Crossref

 
 

Trotsiuk V, Hobi ML, Commarmot B (2012). Age structure and disturbance dynamics of the relic virgin beech forest Uholka (Ukrainian Carpathians) Forest Ecology and Management 265:181-190.
Crossref

 
 

White F (1986). La végétation de l'Afrique avec cartes de la végétation Paris France.

 

 




          */?>