Structural Characteristics, Ecology, and Dynamics of Plant-Communities in Toffo Forest Plantations (Benin, West Africa): Lessons Learnt for Forest Sites Identification, Forest Productivity, and a Sustainable Management of the Forest Resources

Abstract

Research works were carried out in the forest plantations of Toffo reserve (N 6°51′ to N 6°53′ and E 2°05′ to 2°10′). The objectives of the research were to identify and characterize the plant communities of the forest in relation to the ecological factors and the productivity of the forest plantation so as to enable a sustainable management of the forest and biodiversity conservation. In order to achieve those objectives, we used the approach of integrated synusial phytosociology in the study of vegetation; inventories of plantations were made in plant communities so as to identify the levels of productivity of the forest plantations within plant-communities. From the main results achieved, based on the spatial-temporal relationships of the twenty-one (21) synusia (elementary plant-communities) described, they were combined to describe and characterize nine (09) phytocoenoses (more complex plant-communities). The study of the productivity level of plantations within the undergrowth phytocoenoses helped to identify three (03) plantation productivity levels. Finally, the relationships between phytocoenoses, ecological factors, and plantation productivity enabled us to identify and map four (04) forest sites. Taking into account the potentialities and constraints of each forest site, we recommended silvicultural operations to enable sustainable forest management and biodiversity conservation.

1. Introduction

Forests are part of biodiversity which is of utmost importance for life on earth. Forests currently cover an area of 4.06 billion ha, or 31% of the land surface [1]. Forests provide humanity key ecosystem services such as provisioning ecosystem services composed of food, wood, roots, tubers, tannin, non-wood products (honey, meat, etc.) [2, 3]; religious and cultural ecosystem services of forests include ecotourism, cultural heritage etc.; regulatory services are related to mitigation of the effects of climate change and natural disasters (winds, tornadoes, hurricanes, etc.), carbon sequestration etc.; support services of forests are assured across water cycle, nutrient recycling, primary production, soil protection etc. Despite those important services of forests to people, they are facing alarming threats. Indeed, according to FAO [1], fires represent the main threat to forests; they caused a loss of 98 million ha or 4% of the total forest area in 2015. According to the same report, insects and diseases also constitute serious threats to the survival of forests. When it comes to forest losses, Africa comes first with a net annual loss of 3.9 million ha during the decade 2010–2020 [1].

Benin’s natural forest resources are quite limited. Indeed, since 1980, FAO inventory has clearly shown that forest formations (wooded savannahs, open forests, dense forests) covered at that time less than 12% of the country’s surface area, or approximately 1,337,440 ha. Since then, fires, overgrazing and clearing for agricultural purposes have caused advanced degradation of the limited forest resources at a rate of 50,000–70,000 ha/year [1]. In order to satisfy the needs of an ever-growing population, the governments of the country undertook teak (Tectona grandis L. f.) plantations since the end of the 1940s. As of today, the state’s teak stands cover around 20,000 ha and the private teak plantations also cover nearly 20,000 ha [4]. The sustainable management of the forest resources requires sound knowledge of their ecology and silvicultural performances. Sustainable management of forest resources in Benin is facing major constraints including poor knowledge of forest ecosystems; insufficient knowledge of forest phytocoenoses and their relationships with forest sites and forest productivity etc. The purpose of this paper is to fill the knowledge gaps identified so far: (1) identify and characterize the plant-communities of Toffo forest; (2) study the ecological factors related to plant-communities; (3) study the forest productivity in relation to the plant-communities identified; (4) highlight the relationships between plant-communities, ecological factors, and forest productivity; and (5) identify and characterize forest sites. Achieving these objectives made it possible to propose applied and sustainable forest management measures to support decisions of forest managers.

2. Study area

The study area is located in south-Benin in the Lama forest reserve, between 6°51′ and 6°53′ of North latitude and 2°05′ and 2°10′ of East longitude. The forest studied is located approximately 80 km north of Cotonou. It is bordered to the north by the forest plantations of the firewood project, to the west partly by the forest plantations of the firewood project and the Cotonou-Parakou railway, to the East by a path connecting the villages Koussi and Akpè and to the south by the Cotonou-Parakou railway (Figure 1). From an administrative point of view, the Toffo plantations are located in the municipality of Toffo. The study area is under the influence of a subequatorial climate with two rainy seasons: the main rainy season extends from March to July and the short rainy season covers September and October. The long dry season extends from November to February and the short dry season covers August and corresponds to a drop in rainfall between the long rainy season and the short rainy season. Rainfall recorded between 1961 and 2018 showed an average amount of 1030 mm. The average daily temperature is 27.5°C and the relative humidity of the air varies between 52 and 95%.

Figure 1.

Location of Toffo forest reserve. Map J.C. Ganglo.

Overall, we have two major types of soil: ferrallitic soils are found on the southern edge of Toffo forest plantations, upstream of the Agbé river. Black cotton soils (“Vertisols”) cover the northern part of the Agbé watercourse.

The original vegetation of Lama forest was floristically composed of Dialium guineense and Diospyros mespiliformis community and D. guineense and Afzelia africana community [5]. It has been degraded and resulted in a secondary forest characterized by a mosaic of Lonchocarpus sericeus and Ceiba pentandra community; L. sericeus and Anogeissus leiocarpa community and L. sericeus and Albizia zygia community [5, 6]. Fallows are dominated by pioneer species such as Chromolaena odorata, Combretum hispidum, Reissantia indica etc. [5, 7].

Forest plantations were intensively established in Toffo from 1949. Teak is the main species of reforestation and covers 72% of the forest area; the second reforestation species is Cassia siamea (16%). Other species include Gmelina arborea (3%), Khaya senegalensis, Cedrela odorata etc.

In Toffo forest, wildlife includes grass cutters, bushbuck (antelope), bush pigs, monkeys etc. There are also many birds (bee-eaters, hawks, etc.), reptiles (lizards, pythons, snakes) and many insects, including especially bees and the stinking locust.

The Municipality of Toffo, which includes the forest area, had a total population of approximately 101,585 inhabitants, including 49,068 men and 52,517 women [8]. Different ethnic groups are represented such as the Aïzo, the Fons, the Holli and, the Adja. The populations are mainly composed of farmers (at least 75%) whose uncontrolled activities are cause of forest degradation.

3. Material and methods

From Ganglo and de Foucault [9] we can summarize the following:

3.1 Phytosociological study of spontaneous vegetation

In order to carry out the phytosociological study, we used a 1/10,000 forest map for plant community mapping and orientation in the field. A Global Positioning System (GPS) was used to record relevant points and plant community contours; a SUUNTO compass was used for field orientation. In the study of the spontaneous vegetation, we used the synusial approach to phytosociology developed by Gillet et al. [10] and Gillet [11]. A plant synusia is an elementary plant community with homogenous species composition and a dominant type of biological, morphological and adaptation strategy [10]. We distinguished the following categories of synusia: annuals, low or high herbaceous perennials, shrubs, lianas and tree like synusia. The sample surfaces varied from 500 m² for the annual and herbaceous synusia to 1000 m² for the others. All flowering species and easily identifiable ferns were recorded in each sample. Each synusia was named for the species most frequently found and generally attached to its site conditions. Based on spatial and temporal relationships, the plant synusia were combined to describe more complex plant communities known as phytocoenoses. In this article, “plant community” refers to the phytocoenosis. Each plant community was named for its unifying component synusia; i.e., the one responsible for the structural unity of the plant community. We distinguished the same types of plant communities as of synusia, but spontaneous tree communities are not represented in the plantations.

3.2 Ecological factor study

The study of ecological factors involved the use of a soil drilling device to determine soil texture by touch. Soil color was determined with the Munsell code and a machete was used to refresh soil profile horizons and take soil samples. Polyethylene containers were used to carry soil samples to the laboratory for analysis. Slopes were measured with the SUUNTO slope meter device. Soil types were studied within each plant community, by appreciating tactile soil texture up to 50 cm in depth in at least four drill holes. In addition, at representative points of each plant community, two soil profiles 2 m in depth, 2 m in length and 1 m in width were dug for detailed soil descriptions. Topographical parameters such as slopes, topographical positions, and exposure were also noted. One soil sample was taken per profile in each of the two upper horizons for laboratory analysis. In the laboratory, phosphorus was extracted using the Bray I method: the solution used was hydrochloric acid in ammonium fluoride and the concentration was determined by means of standard curves. Soil pH was determined with a pH meter whose electrodes were immersed in a solution of 20 g of soil and 50 ml of distilled water. Soil texture was determined with the soil hydrometer method. Exchangeable cations were extracted with ammonium acetate and assayed. Carbon and organic matter were determined with the Walkley and Black method and the nitrogen content with the Kjeldalh method.

3.3 Forest plantation parameter study

Dendrometric parameters (diameter and height) were measured with a tape measure and a Blum-Leiss in representative parts of each plant community. To do so, at least four rectangular 300 m² sample plots (15 m × 20 m) were assessed in each plant community. The following parameters were recorded: diameter at breast height (1.30 m); height of the two largest trees in each plot in order to calculate top height; height of two trees in the mean diameter class in order to calculate the mean height. On the basis of productivity curves plotted for teak plantations in southern and central Benin [4], top heights were used to calculate the productivity indices of each plant community.

4. Results and discussion

4.1 The synusia (elementary plant-communities) described in the forest

There are a total of twenty-one (21) synusia described in Toffo forest reserve. They represent the different phases of vegetation dynamics taking place in Toffo forest. Indeed, they are concrete components of the stages of the successional vegetation series in Toffo forest reserve, although quite disturbed by human activities of forest plantations. There are three (03) annual synusia; two (02) ruderal synusia; three (03) low herbaceous synusia; three (03) high herbaceous synusia; four (04) shrubby, vinelike synusia; four (04) hygrophilous and hydrophilous synusia; and two (02) tree like synusia. In the table of Appendix 1 (see online material at https://docs.google.com/document/d/19LN-_P1-eNzVb1lHTDSl9KBBLCBztFJYkxRYhKI4jAM/edit?usp=sharing), we present a synthesis of the main characteristics of the synusia described.

4.2 Structural characteristics, ecology and silvicultural indicatory values of the phytocoenoses (complex plant-communities) described in the forest

4.2.1 High perennial herbaceous phytocoenosis of Chromolaena odorata

4.2.1.1 Synecology and indicatory value

The phytocoenosis is a highly heliophile post-cultural pioneer wasteland vegetation. Its presence is due to direct sunlight irrespective of types of soils and topography. Our results agree with those of Ganglo [7], Awokou et al. [12]; Noumon et al. [13] and Yévidé et al. [14] who described this phytocoenosis and its various variants in young non-closed canopy forest plantations and open environments in southern and central Benin. According to those authors, the Chromolaena odorata phytocoenosis colonizes various topographical positions and various soil types, direct sunlight being its only ecological determinism.

4.2.1.2 Synusial components

Chromolaena odorata represents the dominant synusia of the phytocoenosis with a frequency of 100% and a coverage of 42% (Figure 2). Companion synusia are quite diverse: annual, low perennial herbaceous, vinelike, shrubby, and tree synusia.

Figure 2.

View of the phytocoenosis of Chromolaena odorata. Photo J.C. Ganglo.

4.2.1.3 Silvicultural indicatory value

Due to the diversity of the forest sites it colonizes, the phytocoenosis does not have a reliable homogenous silvicultural indicatory value.

The phytocoenoses described in the study are presented on Figure 3.

Figure 3.

Phytocoenoses identified and described in Toffo forest reserve. Map J.C. Ganglo.

4.2.2 Shrubby phytocoenosis of Lecaniodiscus cupanioides

4.2.2.1 Synecology and indicatory value

The phytocoenosis is a shrubby vegetation of undergrowth of teak plantations. It mainly grows on leached ferrallitic soils.

4.2.2.2 Synusial components

Lecaniodiscus cupanioides represents the dominant synusia of the phytocoenosis with a frequency of 100% and a coverage of 58% (Figure 4). Companion synusia are quite diverse: annual, low perennial herbaceous, high perennial herbaceous, and tree synusia.

Figure 4.

View of the phytocoenosis of Lecaniodiscus cupanioides. Photo J.C. Ganglo.

4.2.2.3 Silvicultural indicatory value

Repeated measurements of the productivity index in the teak plantations of the phytocoenosis proved the homogeneity of the level of productivity across the phytocoenosis. Its biotope belongs to the fourth class of soil fertility where teak plantations are cultivated in South and Center Benin (Tables 1-3 and Figure 3) [4].

Order number of plots of productivity measurements	Non-pioneer phytocoenosis of Lecaniodiscus cupanioides	Non-pioneer phytocoenosis of Mallotus oppositifolius and Reissentia indica	Non-pioneer phytocoenosis of Paullinia pinnata and Combretum hispidum	Non-pioneer phytocoenosis of Cola millenii and Icacina tricantha	Pioneer phytocoenosis of Chromolaena odorata
1	20.0	26.1	23.4	19.0	34.9
2	18.8	25.8	24.7	19.4	25.8
3	18.8	26.1	24.7	16.7	18.5
4	20.0	25.2	25.5	20.0	27.6
5	18.1	23.2	24.1	—	25.5
6	18.1	26.7	20.4	—	19.7
7	23.2	27.1	21.4	—
8	21.5	—	—	—
Mean values (m)	19.81	25.74	23.41	18.78	25.3
Amplitude of productivity levels	5.1	3.9	4.7	3.3	16.4
Standard deviation (m)	1.79	1.28	1.94	1.44	5.9
Coefficient of variation (%)	9.03	4.97	8.29	7.67	23.32

Table 1.

Variation of productivity indices within phytocoenoses.

–	Degree of freedom	Mean squares	F	p
Factor	3	63.56614	22.96240	0.000001
Error	22	2.768271	—
Total	25	66.334411	—

Table 2.

Analysis of variance of productivity levels of non-pioneer phytocoenoses.

	Cola	Comb	Leca	Mal
	18.77500	23.41429	19.81250	25.74286
Cola		0.000330	0.289857	0.000167
Comb			0.001201	0.023597
Leca				0.000141
Mal

Table 3.

Newman-Keuls test for comparison of productivity indices between non-pioneer phytocoenoses; cola = phytocoenosis of Cola millenii and Icacina trichantha; comb = phytocoenosis of Paullinia pinnata and Combretum hispidum; Leca = Lecaniodiscus cupanioides phytocoenosis; Mal = phytocoenosis of Mallotus oppositifolius and Reisentia indica.

4.2.3 Shrubby and vinelike phytocoenosis of Mallotus oppositifolius and Reissantia indica

4.2.3.1 Synecology and indicatory value

The phytocoenosis is a shrubby-vine like thicket stage of vegetation on sandy soils of the flats of slopes in teak plantations of Toffo forest. Our results support those of Noumon et al. [13] and Yévidé et al. [14] who described several Mallotus oppositifolius phytocoenoses on black cotton soils and ferrallitic soils in various topographical positions. Sinasson et al. [15] described a vicarious phytocoenosis of Mallotus oppositifolius and Macrosphyra longistyla on flat and mid-slope sites, dominated by clay-sandy soils in the private teak plantations in southern Benin.

4.2.3.2 Synusial components

Mallotus oppositifolius and Reissantia indica represent the dominant synusia of the phytocoenosis with a frequency of 100% and a coverage of 58% (Figure 5). Companion synusia are quite diverse: annual, low perennial herbaceous, high perennial herbaceous, and tree synusia.

Figure 5.

View of Mallotus oppositifolius and Reissantia indica. Photo J.C. Ganglo.

4.2.3.3 Silvicultural indicatory values

Repeated measurements of the productivity index in the teak plantations of the phytocoenosis proved the homogeneity of the level of productivity across the phytocoenosis. Its biotope belongs to the second class of soil fertility where teak plantations are cultivated in South and Center Benin (Tables 1-3 and Figure 3) [4].

4.2.4 Vinelike phytocoenosis of Paullinia pinnata and Combretum hispidum

4.2.4.1 Synecology and indicatory value

The phytocoenosis is a vinelike thicket stage of vegetation on poorly drained black cotton soils of Toffo forest reserve. On the poorly drained black cotton soil of the Lama in central Benin, Noumon et al. [13] described, on similar sites conditions, a vicarious phytocoenosis of Paullinia pinnata and Reissantia indica.

4.2.4.2 Synusial components

Paullinia pinnata and Combretum hispidum represent the dominant synusia of the phytocoenosis with a frequency of 70% and a coverage of 30% (Figure 6). Companion synusia are quite diverse: annual, low perennial herbaceous, high perennial herbaceous, and tree synusia.

Figure 6.

View of the phytocoenosis of Paullinia pinnata and Combretum hispidum. Photo J.C. Ganglo.

4.2.4.3 Silvicultural indicatory values

Repeated measurements of the productivity index in the teak plantations of the phytocoenosis proved the homogeneity of the level of productivity across the phytocoenosis. Its biotope belongs to the third class of soil fertility where teak plantations are cultivated in South and Center Benin (Tables 1-3 and Figure 3) [4].

4.2.5 Hygrophilic phytocoenosis of Cola millenii and Icacina trichantha

4.2.5.1 Synecology and indicatory value

The phytocoenosis is a shrubby ticket stage in the successional vegetation series of Toffo forest growing along watercourses in slightly humid forest sites.

4.2.5.2 Synusial components

Cola millenii and Icacina tricantha represent the dominant synusia of the phytocoenosis with a frequency of 100% and a coverage of 48% (Figure 7). Companion synusia are quite diverse: annual, high perennial herbaceous, and tree synusia.

Figure 7.

View of the phytocoenosis of Cola millenii and Icacina trichantha. Photo J.C. Ganglo.

4.2.5.3 Silvicultural indicatory values

Repeated measurements of the productivity index in the teak plantations of the phytocoenosis proved the homogeneity of the level of productivity across the phytocoenosis. Its biotope belongs to the fourth class of soil fertility where teak plantations are cultivated in South and Center Benin (Tables 1-3 and Figure 3) [4].

4.2.6 Hygrophilic phytocoenosis of Mitragyna inermis and Berlinia grandiflora

4.2.6.1 Synecology and indicatory value

The phytocoenosis is a ticket stage of a shrubby vegetation on periodically flooded lowlands, heavy asphyxiating soil. A similar phytocoenosis of Mitragyna inermis and Nauclea latifolia, was described under the same site conditions in central Lama forest by Noumon et al. [13].

4.2.6.2 Synusial components

Mitragyna inermis and Berlinia grandiflora represents the dominant synusia of the phytocoenosis with a frequency of 100% and a coverage of 32% (Figure 8). Companion synusia are either helophytic or annual.

Figure 8.

View of the phytocoenosis of Mitragyna inermis and Berlinia grandiflora. Photo J.C. Ganglo.

4.2.6.3 Silvicultural indicatory values

The habitat of the phytocoenosis was not planted with teak because of the heavy humid and asphyxiating soil.

4.2.7 High helophytic phytocoenosis of Cyclosorus striatus

4.2.7.1 Synecology and indicatory value

The Cyclosorus striatus phytocoenosis is the helophytic fern vegetation of the backwaters of Toffo forest reserve.

4.2.7.2 Synusial components

Cyclosorus striatus represents the dominant synusia of the phytocoenosis with a frequency of 100% and a coverage of 80%. Companion synusia are either annual, high perennial, shrubby or tree synusia.

4.2.7.3 Silvicultural indicatory values

The habitat of the phytocoenosis was not planted with teak because of the heavy humid and asphyxiating soil.

4.2.8 Low perennial herbaceous phytocoenosis of Leersia hexandra and Alternanthera sessilis

4.2.8.1 Synecology and indicatory value

The phytocoenosis is the herbaceous vegetation of the lowlands and edges of watercourses of the Toffo forest reserve. Synusias close to its component synusia were described under the same sites conditions in the central Lama forest by Noumon et al. [13].

4.2.8.2 Synusial components

Leersia hexandra and Alternanthera sessilis represent the dominant synusia of the phytocoenosis with a frequency of 100% and a coverage of 70% (Figure 9). Companion synusia are either annual, high perennial, shrubby or tree-like.

Figure 9.

View of the phytocoenosis of Leersia hexandra and Alternanthera sessilis. Photo J.C. Ganglo.

4.2.8.3 Silvicultural indicatory values

The habitat of the phytocoenosis was not planted with teak because of the flooded heavy humid and asphyxiating soil.

4.2.9 Hidrophilic phytocoenosis of Nymphaea maculata

4.2.9.1 Synecology and indicatory value

The phytocoenosis is the hydrophilic vegetation of stagnant waters in the Toffo forest reserve. A similar plant community of Hygrophyla auriculata and Pseudoeriosema borianii, was described in the Rusizi plain on the edge of Lake Tanganyika in Burundi [16].

4.2.9.2 Synusial components

Nymphaea maculata represents the dominant synusia of the phytocoenosis with a frequency of 100% and a coverage of 50% (Figure 10). Companion synusia is reduced to the synusia of Leersia hexandra and Alternanthera sessilis.

Figure 10.

View of the phytocoenosis of nymphaea maculate. Photo J.C. Ganglo.

4.2.9.3 Silvicultural indicatory values

The habitat of the phytocoenosis was not planted with teak because of the flooded heavy humid and asphyxiating soil.

4.3 Synthesis of the spatial-temporal dynamics of the undergrowth vegetation of Toffo teak plantations

At Toffo, we identified two geomorphological units (south and north), with respectively north and south exposure to sunlight, separated by a shallow valley. This geomorphology and the resulting topography are the major determining factors, in the sense of Duvigneaud [17], Blasi and Frondoni [18], of the vegetation types (plant synusia and phytocoenoses) described in this work.

The southern geomorphological unit, with a northern exposure to sunlight, is dominated by a clayey-sandy ferrallitic soil [19, 20] while the northern geomorphological unit, facing south is rather dominated by black cotton soils (“vertisols”) that are more or less poorly drained [20]. Slope break points, forming flat areas, are noted on both sides in the two morphological units and are dominated by sandy soils.

On either side of the valley which connects the two geomorphological units, in open environments, on wastelands, clearcuts and young unclosed-canopy plantations, pioneer plant successions take place and result in series of vegetation successional phases in the sense of Blasi and Frondoni [18]. This pioneering series of vegetation are determined by more or total direct sunlight.

In the successional vegetation phases of Toffo forest we noted:

1. Annual synusia with Spigelia anthelmia and Phyllanthus amarus in full sunlight; annual synusia with Asystasia gangetica and Phaulopsis falcisepala in less sunny environments and annual synusia with Aspilia africana; the latter develops in hygrophilic sites along the banks of watercourses.

   Following the annual synusia and always independently of the geomorphological units, perennial herbaceous synusia take place, in connection with previous cultivations, the degree of sunshine and/or ruderalization. We noted:

2. The low perennial herbaceous synusia of Imperata cylindrica that colonizes soils exhausted and degraded by previous cultivation; the high perennial herbaceous synusia of Chromolaena odorata and the related phytocoenosis are observed in plantation gaps, clearcuts and wastelands; the high perennial herbaceous synusia of Panicum maximum and Andropogon gayanus develops in sites whose soils are leached and depleted, either by previous cultivations or by the low water retention capacity of sandy sites; ruderal synusia of Sporobolus pyramidalis develops alongside roads and forest paths while the ruderal synusia of Desmodium velutinum and Sporobolus pyramidalis develops in plantations, in compacted places where traces of ancient human activities (huts, old abandoned paths etc.) are noted.

   Following the pioneering series of annual and perennial herbaceous vegetation, we identified and described types of vegetation specific to each geomorphological unit.

3. On the southern geomorphological unit dominated by ferrallitic soils, the shrubby synusia of Lecaniodiscus cupanioides and its related phytocoenosis are the dominant undergrowth plant communities. These plant communities colonize leached and/or eroded soils, often located on plateaus and slopes with slight to moderate slopes; the shrubby synusia of Cola millenii and Icacina tricantha and its related phytocoenosis form the undergrowth plant communities of the hydromorphic ferrallitic soils of the bottom of slopes along the banks of watercourses; the tree-like synusia of Ceiba pentandra and Triplochiton scleroxylon represents the vestige of natural secondary semi-deciduous forests of on well-drained soils and indicates the final stage of the evolution of the vegetation series in the absence of human disturbance.

4. On the northern geomorphological unit dominated by black cotton soils (“vertisols”), the vine like synusia of Paullinia pinnata and Combretum hispidum and its related phytocoenosis, exclusively colonize the more or less poorly drained black cotton soils.

5. On the flats of slopes of the two geomorphological units, the shrubby vine-like synusia of Mallotus oppositifolius and Reissantia indica and related phytocoenosis, typically grow on soils with a sandy coverage.

6. In the valley which connects the two geomorphological units, the hydrophilic synusia of Nymphaea maculata and its related phytocoenosis grow sparsely in the micro-depressions and stagnant waters of lowlands; the Leersia hexandra and Alternanthera sessilis synusia and its related phytocoenosis are pioneer helophytic plant communities, indicative of the mud of the lowlands and the edges of watercourses; the Cyclosorus striatus synusia and its related phytocoenosis are hygrophilous plant communities that grow in more or less enclosed ravines, served by backwaters of watercourses; the synusia of Mitragyna inermis and Berlinia grandiflora and related phytocoenosis are hygrophilous shrubby plant communities characteristic of lowlands with heavy, asphyxiating and periodically flooded soils; the tree-like synusia of Cola gigantea grow on hygrophilic sites of the banks and beds of watercourses.

4.4 Floristic diversity of Toffo forest plantations

The most diverse genera of the forest are Combretum (7 species), Desmodium, Sida (5 species each), Albizia, Cassia, Cissus, Ficus, Ipomoea, Salacia (4 species each), Celosia, Corchorus, Diospyros, Euphorbia, Hibiscus, Indigofera, Ludwidgia, Panicum and Vernonia with 3 species each. The most diverse families are that of Fabaceae with 35 species, followed by that of Poaceae with 33 species.

The number of species per synusia varies from 02 to 130. The least diverse synusia are those found on selective/constraining site conditions. This was the case of hygrophilic and hydrophilic synusia, including in particular the synusia of Mitragyna inermis and Berlinia grandiflora and that of Nymphaea maculata which grow on asphyxiating hydromorphic soils and in stagnant water respectively; they have 02 and 03 species respectively. The most diverse synusia are notably the low perennial herbaceous synusia where juvenile individuals of vine-like, shrubby, and tree-like species are abundantly represented. This was particularly the case of the low perennial synusia of Scadoxus multiflorus (118 species) and that of the low perennial synusia of Imperata cylindrica (130 species).

The total number of spontaneous plant species inventoried in Toffo plantations is 380, distributed within 290 genera and 91 families, over a total forest area of 855 ha. The species’ diversity of the undergrowth of the teak plantations of Toffo forest represents 13.54% of the flora of Benin sensu Akoegninou et al. [21]. Forest plantations therefore play an essential role in the conservation of the residual biodiversity of Toffo plantation. Our results support those of Luke et al. [22] who obtained 120 plant species belonging to 41 families in the palm plantations in Sumatra (Indonesia); they then concluded that there is a substantial representation of biodiversity in the undergrowth of palm groves. The research results of Braun et al. [23] showed a reduction in biodiversity due to pine plantations (Pinus radiata, P. contorta and P. ponderosa) in Chile and Chilean Patagonia; Malysz et al. [24] also noted that natural Araucaria forests have significantly higher functional species richness than pine and Araucaria forest plantations in the subtropical zone of southern Brazil; in the same vein, Abella et al. [25] reported the effect of conifer plantation on understory diversity after 14 years of experience in northwest Ohio (USA); they noted that the richness of endogenous species per 0.05 ha was 34–50% higher on plots where conifers had been cut than on control plots in uncut plantations; they also noted the development of species characteristic of the habitat on plots where conifers were cut while none of these species developed on plots of uncut plantations; the work of Longworth and Williamson [26] also showed a significant difference in the species composition of natural regeneration between plantations of Hieronyma alchorneoides and Vochysia guatemalensis and secondary forests in Costa Rica; however, Bauhus and Schmerbeck [27] stated that the ecosystem services and functions of plantations are virtually the same as those of natural forests even though due to the difference in structure and composition, these ecosystem services and functions can be produced to different degrees; they also pointed out that forest productivity is positively correlated with its level of diversity due to the complementarity of niches and ecological assurance between the species present. In line with the above results, it is also evident from our work that the conversion of natural forests into forest or agroforestry plantations reduces biodiversity, however, the residual biodiversity is not negligible (more than 13% of the total flora of Benin).

Zhang et al. [28] have also shown in the plantations and forests of Beijing in China that the afforestation species and the sampling scales of data analysis must be taken into account in assessing the level of diversity supported by the plantations. In the same vein, it is important to emphasize that the effect of plantations on biodiversity must be analyzed in relation to the groups of plants in place and the inventory scales [29]. Indeed, according to these authors, in Ireland, on 4m² sampling plots, the analyzes revealed that the number of vascular plants is significantly higher in non-planted areas compared to plantations; on 100m² plots, on the other hand, the number of Bryophytes and lichens is significantly higher in planted forests than in non-planted areas. According to Bremer and Farley [30], the degree of biodiversity conservation by plantations in comparison with the original soil cover depends on whether we are dealing with the conversion of a pasture, a shrubby cover, a secondary forest or primary forest into plantations and whether the species used in plantations are exotic or endogenous. According to Humphrey et al. [31], Carnus et al. [32], Eycott et al. [33] and Hébert et al. [34], the stages of development of the plantations, their topographical positions, the silvicultural practices, the age of the plantations, their density, the origin of the planted species, the density of the canopy, the logging methods such as clear-cutting etc. affect the biodiversity of forest plantations and highlight the need to take into account the history of land use when assessing the species richness of forest plantations. Toffo forest plantations are timber production forests and silvicultural interventions such as weeding in young plantations, thinning in stands and clear felling at the age of exploitation are all practices that could affect the biodiversity of the forest.

4.5 Phytocoenoses, productivity and forestry sites

4.5.1 Degree of homogeneity of productivity levels within phytocoenoses

We remind here that within forest plantations, we identified five phytocoenoses:

• non-pioneer phytocoenosis of Lecaniodiscus cupanioides;

• non-pioneer phytocoenosis of Mallotus oppositifolius and Reissantia indica;

• non-pioneer phytocoenosis of Paullinia pinnata and Combretum hispidum;

• non-pioneer phytocoenosis of Cola millenii and Icacina tricantha;

• pioneer phytocoenosis of Chromolaena odorata.

The areas of extension of these phytocoenoses in Toffo forest are presented in Figure 3. In order to analyze the degree of homogeneity of teak plantations productivity within these phytocoenoses, we made repeated measurements of the productivity indices (Tables 1-3). It appeared that forest plantations present a notable variability of productivity within the pioneer phytocoenosis of Chromolaena odorata. Indeed, within this phytocoenosis, the amplitude of productivity levels was quite high (16.4 m); this resulted in a high standard deviation (5.9 m) in comparison with the values obtained in other phytocoenoses (1.28–1.94 m). The coefficient of variation in the Chromolaena odorata phytocoenosis is also the highest (23.32%) in comparison with the values obtained in the other phytocoenoses (4.97–9.03%). We deduced that productivity levels are quite heterogeneous within the pioneer phytocoenosis of Chromolaena odorata; therefore it does not have a reliable silvicultural indicatory value.

When we consider non-pioneer phytocoenoses, we made the following observations:

• the standard deviations of productivity levels were quite low (maximum 2 m);

• the coefficients of variation were also low (9% maximum);

• the amplitudes of the productivity levels were 5 m at most and showed that the productivity values measured within each phytocoenoses globally belong to the same class of productivity.

We deduced that the productivity of teak plantations was remarkably homogeneous within each non-pioneer phytocoenosis. They can therefore be considered as reliable indicators of productivity levels of teak plantations in Toffo forest.

4.5.2 Variation in productivity levels according to phytocoenoses

In order to highlight the variation in productivity levels according to phytocoenoses, we submitted the data collected to variance analysis with one criterion of classification (Table 2). The productivity indices obtained in the Chromolaena odorata phytocoenosis were not taken into account in the analysis because of their high variability. The analysis of variance was followed by the Newman-Keuls test for means comparison (Table 3).

From Tables 2 and 3, we can deduced that:

• the analysis of variance showed a significant difference between the productivity levels of plantations across phytocoenoses at the high probability threshold of 10⁻⁶ (Table 2);

• teak plantations of the phytocoenosis of Mallotus oppositifolius and Reisentia indica was the most productive. Indeed, its average productivity index of 25.7 m was very significantly higher than the productivity indices of the other phytocoenoses (Tables 1–3);

• the phytocoenosis of Paullinia pinnata and Combretum hispidum has an average productivity level of 23.4 m; this level of productivity is intermediate and also presented a significant difference with the productivity levels of the other phytocoenoses except that of Mallotus oppositifolius and Reissantia indica phytocoenosis at the 5% of probability threshold.

• the phytocoenosis of Lecaniodiscus cupanioides and that of Cola millenii and Icacina trichantha had respectively the average productivity index of 19.8 m and 18.8 m. These productivity levels are not significantly different from each other at the 5% of probability threshold and allowed us to confirm that these two phytocoenoses belong to the same biological equivalence class as defined by Dagnélie [35, 36].

4.5.3 Productivity classes identified in teak forest plantations

Based on the yield table of teak plantations in Benin [4], it appeared from the above analyzes that three productivity classes are represented by the phytocoenoses described in the teak plantations of Toffo forest reserve (Table 4). Figure 11 presents the productivity map achieved on the basis of the results of our work.

Phytocoenoses	Productivity index	Productivity classes [4]
Phytocoenosis of Mallotus oppositifolius and Reissantia indica	25.7a	2
Phytocoenosis of Paullinia pinnata and Combretum hispidum	23.4b	3
Phytocoenosis of Lecaniodiscus cupanioides	19.8c	4
Phytocoenosis of Cola millenii et Icacina trichantha	18.8c	4

Table 4.

Productivity of teak plantations across phytocoenoses.

In the same column, productivity levels followed by the same letters are not significantly different at 5% of probability threshold.

Figure 11.

Productivity levels of teak plantations of Toffo forest reserve. Map J.C. Ganglo.

On Figure 11, the productivity level of teak plantations in the pioneer phytocoenosis of Chromolaena odorata is deduced from field observations and the yield table of teak plantations of Benin [4].

4.5.4 Forest sites identified in the teak plantations

From the previous analyses, we can deduce that:

1. Undergrowth non-pioneer phytocoenoses are reliable indicators of ecological conditions;

2. In view of the low values of standard deviations, coefficients of variation and amplitudes of the productivity levels of forest plantations within each non-pioneer phytocoenosis, we infer that the productivity of forest plantations is fairly homogeneous within each non-pioneer phytocoenosis;

3. From conditions (1) and (2), we deduced that the biotopes of non-pioneer under-growth phytocoenoses represent forest sites in the sense of Maître [37]; Delpech et al. [38]; Rondeux [39]; Dupuy et al. [40] and Ganglo [4].

The forest sites identified in Toffo forest plantations are represented on Figure 11.

4.6 Forest management measures suggested on the basis of our results

4.6.1 Case of the forest site in the phytocoenosis of Mallotus oppositifolius and Reissentia indica

The forest site of this phytocoenosis belongs to the second productivity class of the teak plantations of South Benin [4]; it is the most productive site of Toffo forest plantations (Figures 3 and 11). The valuable local species that grow in the site are Antiaris toxicaria, Ceiba pentandra, Triplochiton scleroxylon, Afzelia africana, Milicia excelsa; they must be conserved with appropriate and timely silvicultural treatments like liana cuttings, enrichment to increase their stock, and thinning where appropriate to account for over density etc. The forest site of Mallotus oppositifolius and Reissentia indica did not show any particular constraints. We recommend to apply the yield table of teak plantations [4] in order to implement appropriate and timely cultivation operations including thinning of the teak plantations to enable the production of wood of good quality.

4.6.2 Case of the forest site in the phytocoenosis of Paullinia pinnata and Combretum hispidum

The forest site of this phytocoenosis belongs to the third productivity class of the teak plantations of South Benin [4]; it is a poorly drained site of Toffo forest plantations (Figures 3 and 11). The valuable local species that grow in the site are Antiaris toxicaria, Diospyros mespiliformis, Pterocarpus erinaceus, Milicia excelsa; they must be conserved with appropriate and timely silvicultural treatments like liana cuttings, enrichment to increase their stock, and thinning where appropriate to account for over density etc. The forest site of Paullinia pinnata and Combretum hispidum is poorly drained so that the production of teak wood of good quality requires plowing and ridging of the soil, however, as tillage is quite expensive, we recommend to promote, through appropriate silvicultural treatments, the regeneration and cultivation of the valuable local species that are already growing in place.

4.6.3 Case of the forest site in the phytocoenosis of Lecaniodiscus cupanioides

The forest site of this phytocoenosis belongs to the fourth productivity class of the teak plantations of South Benin [4]; it is the least productive site of Toffo forest plantations (Figures 3 and 11). The valuable local species that grow in the site are Antiaris toxicaria, Ceiba pentandra, Triplochiton scleroxylon, Afzelia africana, Milicia excelsa; they must be conserved with appropriate and timely silvicultural treatments like liana cuttings, enrichment to increase their stock, and thinning where appropriate to account for over density etc. The forest site of Lecaniodiscus cupanioides is often exposed to erosion due to its topographic position on the slope (82% of our observations). Given the potentiality and constraints of the site, it is possible, through the implementation of appropriate silvicultural treatments like weeding, liana cutting, thinning etc. to produce quality timber; the silvicultural treatments must also be extended to local species in order to promote their growth and conservation in the plantations.

4.6.4 Case of the forest site in the phytocoenosis of Cola millenii and Icacina trichantha

The forest site of this phytocoenosis also belongs to the fourth productivity class of the teak plantations of South Benin [4] and is therefore one of the least productive site of Toffo forest plantations (Figures 3 and 11). This site, established on a hydromorphic ferrallitic soil, exposes teak to a high risk of wind throw (47% of the observations) due the a weak rooting of teak; indeed, teak roots are unable to explore the deep asphyxiating soil horizons of the site. The valuable local species which develop in the site include Antiaris toxicaria, Ceiba pentandra, Cola gigantea, Sterculia tragacantha etc. They regenerate quite well in this particular site and it is important to promote them through appropriate silvicultural treatments (weeding, liana cuttings, thinning etc.)

4.6.5 Case of the forest sites in the phytocoenosis of Chromolaena odorata

The forest sites in the pioneer phytocoenosis of Chromolaena odorata are characterized by a heterogeneity of productivity levels. We suggest further data collection in the sites at canopy closure so as to identify non-pioneer phytocoenoses where, appropriate forest management measures can be applied to enable wood quality production and biodiversity conservation.

4.6.6 Case of forest sites in hygrophilous and hydrophilous phytocoenoses

The phytocoenoses concerned are phytocoenosis of Mitragyna inermis and Berlinia grandiflora; phytocoenosis of Cyclosorus striatus; phytocoenosis of Leersia hexandra and Alternanthera sessilis; and phytocoenosis of Nymphaea maculata.

Those phytocoenoses grow on heavy ill-drained asphyxiating soils that prevent teak cultivation. The valuable local species that can grow there are Ceiba pentandra, Cola gigantea, Sterculia tragacantha, Berlinia grandiflora, Mitragyna inermis, Pterocarpus santalinoides etc. It is therefore important to promote them through appropriate silvicultural treatments such as weeding, liana cuttings, thinning etc. Research work on the technological qualities of the woods of the local species will make it possible to better define their uses. From a non-timber production perspective, the species Pterocarpus santalinoides which fruits are well appreciated by local populations can be planted. Herbaceous species such as Alternanthera sessilis, Sparganophoros sparganophora are vegetables consumed by people. We recommend to promote their production in market gardens on the banks of watercourses.

5. Conclusions and recommendations

Phytosociology applied to the study of spontaneous vegetation of Toffo forest plantations helped achieve quite interesting results. Indeed, the description and characterization of the synusia made it possible to combine them into phytocoenoses on the basis of their spatial-temporal relationships. In total, 21 synusia were described and characterized in Toffo plantations. They made it possible to identify and describe 09 phytocoenoses. The study of ecological factors, in particular the soil and topography (topographic positions, slopes, etc.), made it possible to highlight the remarkable homogeneity of site conditions within non-pioneer phytocoenoses. As a consequence of this, non-pioneer phytocoenoses are quite indicative of the site conditions of their biotopes. Pioneer phytocoenoses are quite ubiquitous and poorly indicative of the site conditions. The study of the productivity of forest plantations within phytocoenoses revealed the remarkable homogeneity of the level of productivity within each non-pioneer phytocoenosis. Furthermore, levels of plantation productivity vary significantly across non-pioneer phytocoenoses. This study therefore made it possible to highlight the triangular relationship between ecological factors, phytocoenoses, and forest productivity. This very important relationship allowed us to identify Toffo forest sites and recommend useful management measures for the sustainability of forest production and biodiversity conservation. Research work must continue with periodic observations on permanent research sites to collect additional data in order to guide appropriate planning of silvicultural treatments well adapted to each forest site.

Acknowledgments

The author is grateful to Professor Bruno De Foucault, one of the brilliant specialist of the synusial phytosociology who initiated the author to that innovative method of vegetation inventory and description. The author tells also his gratitude the International Foundation of Science (IFS) as a former scholar of the Foundation for its financial support to the initial stage of this work.