Study area
This study was carried out in the Atwima Mponua district of the Ashanti region of Ghana. The district is located in the south-western part of the Ashanti region and covers an area of approximately 894.15 square kilometres (http://www.ghanadistricts.gov.gh). The district is marked by double maxima rainfall seasons. The major rainfall period begins from March to July peaking in May. The average annual rainfall for the major season is about 170–185 cm per year. The minor rainfall period begins in August tapering off in November with an average minor annual rainfall of 100–125 cm per year. Mean annual temperatures of 27 °C are recorded in August and in March. The climate in the district is ideal for the cultivation of cash and food crops such as cocoa, kola, oil palm, maize, cocoyam, plantain, cassava, rice and all kinds of vegetables.
Even though the rainfall is adequate for agriculture, its erratic and unpredictable nature and concentration have adverse implications for rain-fed agriculture. The soil type in the district is the forest ochrosol. The vegetation is basically of the semi-deciduous type. The flora and fauna are diverse and composed of different species of both economic and ornamental tree species with varying heights and game and wildlife (http://www.mofa.gov.gh). The district has organic and conventional cocoa farmers located in several communities within the Tano Dumase Area Council. This study was conducted in two communities, namely Gyereso and Pasoro. The two communities have similar land forms, and so biases in comparison have been reduced.
Conceptual framework: linking flora abundance to income generation and food availability
There are state agencies/public sector institutions such as Cocoa Health and Extension Division (CHED), non-governmental organizations (NGOs) such as Agro Eco-Louis Bolk Institute (AE-LBI) and Rainforest Alliance (RA) and private cocoa beans buyers such as Armajaro Limited that are supporting the incorporation of organic cocoa farming. The organizations view organic farming practices as climate smart since it conserves biodiversity and also supports carbon sequestration. The expectation is that households that have adopted organic farming will have more diverse and abundant species. Therefore, the consumption and sale of such species will lead to increased food security, income and well-being as well as reduced vulnerability for organic farming households. The sustainable livelihoods framework (SLF) developed by the Department for International Development [5] captures the sense of the argument (Fig. 1). It is noted that the vulnerability context influences the level of livelihoods assets which include natural (N), human (H), financial (F), physical (P) and social (S) capital. The influence of transforming structures and processes leads to adoption of strategies that result in improved outcomes.
Method of data collection
Flora survey and farm selection
In August 2014, the ProEcoAfrica project randomly selected 150 households for farming systems’ study in Gyereso and Pasoro. The project is being implemented by the Research Institute of Organic Agriculture (FiBL), Switzerland, and University of Ghana to ascertain the profitability and productivity of organic and conventional cocoa farming systems. The ProEcoAfrica database was examined for relevance in January 2015. A total of 32 (32) cocoa farmers managing organic and conventional farms in the two communities were purposively selected from the database. Only farmers who managed cocoa farms that were aged between 2 and 3 years inclusive were selected. The canopies of these farms were not closed and therefore above-ground flora growth was in progress. After one month, those who were unwilling to participate in the experiment were replaced with willing farmers. For simplicity, there was an even distribution of the 32 farmers among the farming systems and among the communities: Eight each of organic and conventional cocoa farms were selected from Gyereso and Pasoro. Organic farms were selected and paired with conventional farms of similar size within 5 km where there were no adjoining fields between them. The pairing was done to minimize differences in landscape and elevation as well as soil types and characteristics [2].
Procedure for sampling the flora
In each cocoa farm, 25 m × 25 m main plots were marked out randomly. This plot was subdivided into five subplots of 5 m × 5 m out of which one was randomly selected to determine the quantitative abundance of herbaceous plants. The number of individual plant species in each plot was recorded. For each main plot, a species composition list was prepared to include trees, shrubs and herbs of the two farming systems. It was noted that the medicinal flora as well as crop flora like cocoyam (corm) and plantain (musa spp) were largely voluntary plants (that grow without cultivation). Species were identified as far as possible on site with the help of a professional (here, Taxonomist). Samples which could not be identified were collected and pressed for later identification at the herbarium in the Department of Botany, University of Ghana.
Household survey
Household heads of cocoa farms that were sampled were interviewed to assess the contribution of flora diversity on their cocoa farms to their livelihoods. Secondary data on the following variables were obtained from the ProEcoAfrica database: respondent’s biodata, household composition, farming activities, agrochemical application information, membership of a farmer-based organization, health status of farmers’ family and cost of production of farm produce. A face-to-face discussion was held with the individual farmers to validate the secondary data and collect new information through questionnaire administration (Additional file 1). The new data collected were on: frequency of consumption of identified flora species (as a measure of desirability or popularity of species), trends in sales, availability of flora species all year round (2014) and income from sale of flora. A focus group discussion was done to validate the information during the survey.
Methods of data analysis
The flora analysis to show biodiversity involved two steps:
-
1.
Measuring species compositional similarity using Jaccard index [19] and
-
2.
Measuring the diversity among species in the cocoa farm, using Shannon diversity index (H′) and Simpson diversity index (D) [3, 19].
The Jaccard index of similarity is a ratio that is used to determine the similarity between two farming systems. Hence,
$$J = \frac{a}{(a + b + c)}$$
(1)
where J Jaccard index of similarity; a the number of species found in both organic and conventional farms; b the number of species in only organic cocoa farms; c the number of species in only conventional cocoa farms and 0 ≤ J ≥ 1
Decision criteria: zero (0) denotes complete dissimilarity and one (1) for complete similarity.
The Shannon index is a ratio that is used to measure species abundance and evenness (see McCune and Grace [19] for formula). The Simpson index is a ratio used to measure species richness [19]. The calculation of ratios to obtain the Shannon (H′) and Simpson (D) indices for each of the farming systems was guided by the Biodiversity Professional Software (free version). The Mann–Whitney U test was used to test the difference in the indices obtained for organic and conventional farms.
Hypothesis testing of Shannon and Simpson indices (Mann–Whitney U test)
H
o 1
There is no difference in Shannon index of diversity of flora in organic and conventional cocoa farms. Hence,
$$H_{\text{OF}}^{{\prime }} - H_{\text{CF}}^{{\prime }} = 0$$
H
a 1
The Shannon index of diversity of flora in organic cocoa farms is greater than that of conventional. Hence,
$$H_{\text{OF}}^{{\prime }} - H_{\text{CF}}^{{\prime }} > 0$$
H
o 2
There is no difference in Simpson diversity index of flora in organic and conventional cocoa farms. Hence,
$$D_{\text{OF}} - D_{\text{CF}} = 0$$
H
a 2
The Simpson diversity index of flora in organic cocoa farms is greater than that of conventional
$$D_{\text{OF}} - D_{\text{CF}} > 0.$$
Livelihoods outcome measurement
In the household survey analysis, farmers’ livelihoods outcomes were measured:
-
Food security: This study assumed that edible flora species identified on the farms were nutritious and once available will be consumed by household members or sold. It was hypothesized that organic cocoa farmers will consume and sell more flora than their conventional counterparts. The t test of difference between means was used to ascertain the significance of the difference.
-
Well-being: The proportion of income from flora sale used to cover cost of health was measured.
Farmers who spent more than 40% of flora income to finance the cost of their health bills were considered to have better well-being (GSS 2015).
-
Income: The relative frequency of income ranges per annum of up to GHS400 was calculated. Any income (from sale of edible flora) above GHS150.00 was considered high; the difference in the proportion of organic versus conventional farmers who earned above GHS150 was measured, and the t test of difference between means was used to ascertain the significance of the difference.
-
Vulnerability reduction: This was considered as improved social inclusion. It was assumed that membership of farmer-based organization (FBO) improved a person’s social status. FBO is a platform for improving one’s voice in society, gaining and sharing knowledge and improving access to productive resources [15] A score of 1 was assigned to members and score of 0 was assigned to non-members. It was expected that organic cocoa farmers will have a higher mean score than the conventional; the organic farmers were considered more resilient than their conventional counterparts.
