**2.2 Assessing impacts of soil degradation on phytodiversity**

Phytosociological surveys [24] were carried out in each sample as a mean to assess the floristic composition, discriminant species, species richness, species chorological types, species life forms and species dispersal types. Woody species were collected in the plots, while herbs were carried out on the sub-plots. All species were constituted as herbaria and were subsequently determined by the National Herbarium of the University of Abomey-Calavi.

The similarities in species composition between classes of soil degradation were assessed using the index of similarity of Jaccard (1901), which is given by the formula:

$$P\_j = 100 \, ^\star \, \frac{c}{a+b-c};\tag{1}$$

**77**

*The Impacts of Soil Degradation Effects on Phytodiversity and Vegetation Structure on Atacora…*

this study: the chorological index (*IC*), the life forms index (*IL*) and the dispersal types (of diaspore) index (*ID*). The objective was to understand how biodiversity indicators vary according to soil degradation classes, i.e. along degradation

These indexes were computed on the base of two main principles. The first one was the principle of biodiversity's indicators of disturbance. Along a gradient of disturbance, there were three major types of qualitative indicators of biodiversity (chorological types, life forms and dispersal types) which evolutions (in terms of number or cover) were negatively correlated. For example, widely distributed species, therophytes and sclerochory were assumed to be more abundant/dominant in the pioneer (more disturbed) stages and this trend decreased as less disturbed stages were reached. In the contrary, the number/cover of regional species, phanerophytes and sarcochory were assumed to increase from disturbed to stable communities [21, 29, 30]. The second principle is about the ratio or relative frequency used in Ref. [31] to calculate the phytogeographical index (*Ip*) which made it possible to compare and classify the different plant communities according to their level of affinity with the Sudanian or Guinea-Congolian region. On this basis, the

Where *IC* is the chorological index and *S, SZ, SG, Pt, PAL, AA, TA, PRA* are respectively the frequency of Sudanian, Sudano-Zambezian, Sudano-Guinean, Pantropical, Paleotropical, Afro-American, Tropical Africa and Pluri Regional in

> *L Ph I =*

*D Sarco I =*

where *IL* is the life forms index, *Ph* is the frequency of Phanerophytes and *Th* is

where *ID* is the dispersal types index, *Sarco* is the frequency of Sarcochory and

These indices calculated for each plot, compared the relative evolution of each pair of indicators between the different soil degradation classes. The higher the index, the greater the relative abundance of the biodiversity indicator in the numerator. The lower the index, the greater the relative abundance of biodiversity indicators at the denominator. Thereafter, the species richness (S), the chorological index (*IC*), the life forms index (*IL*) and the dispersal types index (*ID*) were submit-

The cover of each species was visually estimated within each plot. Braun Blanquet cover/abundance scale [33] was used: +: rare, less than 1% cover, 1: 1–5% cover, 2: 5–25% cover, 3: 25–50% cover, 4: 50–75% cover, and 5: 75–100% cover. The cover data of all inventoried species through the phytosociological surveys were grouped into an abundance matrix of 22 plots x 133 species and submitted to the Multi Response Permutation Procedures (MRPP). MRPP is a nonparametric procedure for testing the hypothesis of no difference between two or more groups of entities [34]. This procedure was used to pairwise compare the described soil

ted to discriminant analysis and ANOVA using R software [32].

**2.3 Assessing impacts of soil degradation on vegetation structure**

*S +SZ +SG I = ; Pt + PAL+ AA+TA+ PRA* (2)

*Th* (3)

*Sclero* (4)

*DOI: http://dx.doi.org/10.5772/intechopen.93899*

gradient.

indexes were computes as:

the frequency of Therophytes.

*Sclero* is the frequency of Sclerochory.

Africa species.

*c*

where *Pj* is Jaccard community coefficient, *a* is the number of species present in the community A, *b* is the number of species in the community B, and *c* is the number of species shared by A and B. In the study, soil degradation classes represented communities. The computation was automatically performed with the software CAP [25] on a presence/absence matrix consisting of a number of defined soil degradation classes and 133 plant species. This index has proved to be a consistently good measure of similarity for presence/absence data [26]. The values of *Pj* range from 0% for an absence of similarity to 100% for a complete similarity. Plant communities are dissimilar if *Pj* ≤ 50%.

Discriminants species of each degradation class was assessed and identified based on methodology as in Ref. [27]. Discriminant species of a particular group were species devoted to that group, exclusive to that group and never occurring in others groups. Dufrêne & Legendre's method produced indicator values for species within each group. These indicator values were tested for statistical significance using a randomization (Monte Carlo) technique [28]. P value of 5% was used to retain as discriminant species. All multivariate analyses were computed with PC-ORD for Windows Version 5 [28].

The impacts of soil degradation on phytodiversity were also assessed by using species richness (S), and three indexes of diversity that were developed as part of *The Impacts of Soil Degradation Effects on Phytodiversity and Vegetation Structure on Atacora… DOI: http://dx.doi.org/10.5772/intechopen.93899*

this study: the chorological index (*IC*), the life forms index (*IL*) and the dispersal types (of diaspore) index (*ID*). The objective was to understand how biodiversity indicators vary according to soil degradation classes, i.e. along degradation gradient.

These indexes were computed on the base of two main principles. The first one was the principle of biodiversity's indicators of disturbance. Along a gradient of disturbance, there were three major types of qualitative indicators of biodiversity (chorological types, life forms and dispersal types) which evolutions (in terms of number or cover) were negatively correlated. For example, widely distributed species, therophytes and sclerochory were assumed to be more abundant/dominant in the pioneer (more disturbed) stages and this trend decreased as less disturbed stages were reached. In the contrary, the number/cover of regional species, phanerophytes and sarcochory were assumed to increase from disturbed to stable communities [21, 29, 30]. The second principle is about the ratio or relative frequency used in Ref. [31] to calculate the phytogeographical index (*Ip*) which made it possible to compare and classify the different plant communities according to their level of affinity with the Sudanian or Guinea-Congolian region. On this basis, the indexes were computes as:

$$I\_c = \frac{\text{S} + \text{SZ} + \text{SG}}{\text{Pt} + \text{PAL} + \text{AA} + \text{TA} + \text{PRA}};\tag{2}$$

Where *IC* is the chorological index and *S, SZ, SG, Pt, PAL, AA, TA, PRA* are respectively the frequency of Sudanian, Sudano-Zambezian, Sudano-Guinean, Pantropical, Paleotropical, Afro-American, Tropical Africa and Pluri Regional in Africa species.

$$I\_L = \frac{Ph}{Th} \tag{3}$$

where *IL* is the life forms index, *Ph* is the frequency of Phanerophytes and *Th* is the frequency of Therophytes.

$$I\_D = \frac{\text{Sarro}}{\text{Sclero}}\tag{4}$$

where *ID* is the dispersal types index, *Sarco* is the frequency of Sarcochory and *Sclero* is the frequency of Sclerochory.

These indices calculated for each plot, compared the relative evolution of each pair of indicators between the different soil degradation classes. The higher the index, the greater the relative abundance of the biodiversity indicator in the numerator. The lower the index, the greater the relative abundance of biodiversity indicators at the denominator. Thereafter, the species richness (S), the chorological index (*IC*), the life forms index (*IL*) and the dispersal types index (*ID*) were submitted to discriminant analysis and ANOVA using R software [32].

#### **2.3 Assessing impacts of soil degradation on vegetation structure**

The cover of each species was visually estimated within each plot. Braun Blanquet cover/abundance scale [33] was used: +: rare, less than 1% cover, 1: 1–5% cover, 2: 5–25% cover, 3: 25–50% cover, 4: 50–75% cover, and 5: 75–100% cover. The cover data of all inventoried species through the phytosociological surveys were grouped into an abundance matrix of 22 plots x 133 species and submitted to the Multi Response Permutation Procedures (MRPP). MRPP is a nonparametric procedure for testing the hypothesis of no difference between two or more groups of entities [34]. This procedure was used to pairwise compare the described soil

*Soil Erosion - Current Challenges and Future Perspectives in a Changing World*

**Definition**

soils were observed.

remained as thin patches.

**2.2 Assessing impacts of soil degradation on phytodiversity**

*Characteristics of soil degradation classes on Atacora mountain range.*

National Herbarium of the University of Abomey-Calavi.

Phytosociological surveys [24] were carried out in each sample as a mean to assess the floristic composition, discriminant species, species richness, species chorological types, species life forms and species dispersal types. Woody species were collected in the plots, while herbs were carried out on the sub-plots. All species were constituted as herbaria and were subsequently determined by the

Light Soils characterized by a low level of soil compaction, few rills and no visible sheet

Moderate Soils characterized by compact red soils, with a thin clay crust on the surface. Sheet

High Soils red and very compact. They looked like ferricrete but remained friable. Sheet

Extreme Soils characterized by the presence of ferricrete (rich in iron, and hard) and red

less extended (only 23% of rod contacts).

erosion. On the topsoils black organic layer covered the entire surface and no reddish

erosion occurred on these soils, and rills were observed on the surface. Organic layers

erosion occurred. Rills covered a larger surface than on the other soils and were also deep. Organic layer remained as thin patches (less than 5 cm thick) anic layer were

soils. The presence of the ferricrete layer reduced the depth to which roots could grow. The organic layer remained only on small patches. The thickness of the organic layer rarely exceeded 10 cm. There was no visible evidence of sheet erosion, and the

presence of rills was very low because of the high level of compaction.

The similarities in species composition between classes of soil degradation were assessed using the index of similarity of Jaccard (1901), which is given by the

where *Pj* is Jaccard community coefficient, *a* is the number of species present in the community A, *b* is the number of species in the community B, and *c* is the number of species shared by A and B. In the study, soil degradation classes represented communities. The computation was automatically performed with the software CAP [25] on a presence/absence matrix consisting of a number of defined soil degradation classes and 133 plant species. This index has proved to be a consistently good measure of similarity for presence/absence data [26]. The values of *Pj* range from 0% for an absence of similarity to 100% for a complete similarity. Plant communities are

Discriminants species of each degradation class was assessed and identified based on methodology as in Ref. [27]. Discriminant species of a particular group were species devoted to that group, exclusive to that group and never occurring in others groups. Dufrêne & Legendre's method produced indicator values for species within each group. These indicator values were tested for statistical significance using a randomization (Monte Carlo) technique [28]. P value of 5% was used to retain as discriminant species. All multivariate analyses were computed with

The impacts of soil degradation on phytodiversity were also assessed by using species richness (S), and three indexes of diversity that were developed as part of

*<sup>c</sup> P = 100 \* ; a+b-c* (1)

*j*

**76**

formula:

**Table 1.**

**Soil degradation classes**

dissimilar if *Pj* ≤ 50%.

PC-ORD for Windows Version 5 [28].

degradation classes based on the cover data of their species lists. The analysis was computed with PC-ORD for Windows Version 5 [28].
