**2. Methodology**

#### **2.1 Study site**

This multiscale study encompassed the village area, the household and farming unit, the cocoa farm and homogeneous cocoa plantations. The study was conducted in the Centre Region, i.e., in the Evodoula subdivision (Latitude: 4° 04′ 60.00" N Longitude: 11° 11' 60.00" E) (**Figure 1**). Located in one of the oldest cocoa producing basin in a degraded forest zone, the Evodoula village (≈76 Pop/km2 ) is characterized by a high land use intensity.

The climate is hot and humid with average temperatures and relative humidity of 25°C and 75% respectively. The rainfall pattern is bimodal with a heavy rainy season running from mid-September to mid-November, and a severe dry season running from December to mid-March.

Pedologically, there is a wide variety of soils based on structure and texture. The soils are mostly ferralitic, sandy clay and hydromorphic in the lowlands found around certain places. These soils are 90% agricultural land, favorable to the cultivation of cocoa and to a wide range of food and market garden products and are intensively exploited because of the strong demography. The operating method is based on clearing, cutting, burning and plowing, which helps to strip the vegetation cover and exposes the soil to severe erosion and reduced fertility. Unfortunately, the amendments (organic fertilizers) recommended by the competent services are used very little [14].

Vegetation is semi-deciduous evergreen and degrades from the gradient Equator to north [11].

**Figure 1.** *Location of the study site in it agro-ecological zone.*

#### **2.2 Data collection**

Both qualitative and quantitative data were collected from August 2013 to March 2014. Two different methods were used for data collection: (1) semi-structured socioeconomic surveys with households, and (2) direct observations and measurements in cocoa plantations. A total of forty (40) cocoa growers, were selected randomly for the semi-structured socioeconomic interview. These were focused on cocoa plantation characteristics (plantation status, age, history since its initial planting, area, cocoa tree ages, etc.) and the identification/selection and ranking of 10 tree species with leaf litterfall of high fertility potential was done. This identification/selection and ranking was based on their ethno-botanical knowledge of associated indigenous species, and the productivity of cocoa stands around those species. A generalized farmers' ranking was obtained by calculating the mean value of the position occupied in the individual farmers' ranking. The empirical classification by farmers of the fertility potential of these ten species was then compared to the classification of the same species based on their respective nutrient contents (Test Ranking).

Following these interviews with the cocoa growers, field visits to plantations were organized to select fifteen (15) cocoa agroforestry systems, through an in – depth assessment, for specific farm characterization. In each of the fifteen cocoa agroforestry systems, a systematic inventory of all non-cocoa trees exceeding 1 m in height were inventoried over the total area of each cocoa plantation following the method of [15]. Each tree species (forest, exotic as well as palm tree) was counted, numbered, identified and their density per plot estimated. The species identifications were based on vernacular names in the 'Eton' language with the assistance of the farm owner and correspondences with the scientific names were established from literature review [16].

From the above fifteen cocoa farms, litterfall of the 10 trees species of high fertility potential as rank by farmers was collected daily. Here, every newly fallen leaf was collected systematically at the same time after every two (02) days for one (1) week. The collection of fresh litterfall was done by a random walk around the specific tree species studied and the distance covered was from the base of the trunk to the longest branch of tree when the sun is at the zenith. Litterfall was conditioned according to standard procedure and taken to the laboratory for compositional analysis of macro-nutrients (nitrogen, phosphorus, potassium, calcium, and magnesium) and analyses were carried in conformity with standard analytical procedures of [17].

#### *2.2.1 Data analyses*

The data collected from the questionnaire and inventory forms were checked, entered into Microsoft Office Excel 2007 software and were analyzed using the Statistical Package for Social Sciences (SPSS) version 12.0. These analyzes consisted of descriptive statistics (sum, frequency, percentage, tree species densities and cross-tabulations of results), interactive graphs, and the total litter primary macro-nutrient (PMM) contents of the tree's species; which was obtained by summing the proportion of the respective elements analyzed. This, enable us to established the Test ranking of trees species. Data obtained from the chemical analysis were analyzed as a one-way analysis of variance (ANOVA) using the Proc GLM IN in SAS version 9.0. Separation of means was done using the DUNGAN Multiple Range Test, to test for significant effects between the leaf litters nutrient compositions of the different tree species at 5% probability level.

*Farmer's Perception of Associates Non-Cocoa Tree's Leaf Litterfall Fertilizing Potential… DOI: http://dx.doi.org/10.5772/intechopen.100262*
