EDEM WISDOM ANYOMI

GENETIC DIVERSITY AND INDIGENOUS KNOWLEDGE OF TRAIT PREFERENCES FOR SHEA BUTTER GERMPLASM COLLECTION AND IMPROVEMENT IN GHANA

ABSTRACT

Shea butter is an important oil commodity in the global fat and oil industry for its immense use in the food, cosmetic and pharmaceutical industry. Being geographically restricted to West and East Africa, it is a commodity that offers huge potential for development especially for people located around the shea belt, through sales and other income generating activities. The shea tree is however an undomesticated tree species. Efforts to domesticate this important species requires in-depth knowledge of the traditional choices and usage, including problems associated with fruit gathering and butter extraction. Owing to its vast natural distribution and lack of selection, it is important to know the inherent diversity that exist across the different locations to inform breeding objectives,after which hybridization will be targeted at specific trait improvement. The objectives of this research were: (i) conduct a survey to understand the choices or preferences for selection of trees for butter production and germplasm conservation. (ii) assess the diversity and population structure of shea in Ghana using single nucleotide polymorphisms (SNPs). (iii) evaluate the established NAM for growth and vigour. (iv) characterize the fruits and nuts properties of the tagged shea germplasm A Participatory Rural Appraisal was undertaken in five key shea butter producing districts in the northern region of Ghana, to obtain information on shea value chain especially on farmers preference for butter extraction. The result indicated that shea nut collectors could identify their favorite trees by the fruit type (25.63%), fruiting time (14.84%), nut shapes (24.08%), leaf form (17.73%) and tree shape (17.73%). Shea fruiting season was observed to vary between regions, however due to the variations within the trees, the period spans almost four months in each area. It also revealed the depleting nature of shea trees across the belt with Bongo in the Upper East region found to have the lowest shea population with sparse vegetation. Among the most important issues raised, the rate of genetic erosion and the lack of government intervention to mitigate these occurrences were cited for the woes of the dwindling shea population. Agro-morphological diversity for fruit traits was analyzed using unweighted pair group method with arithmetic mean. The results showed at 58% dissimilarity a high diversity within populations and low diversity between populations, manifested two main clusters with many sub clusters. Fruit shape was categorized into 5 groups with ellipsoid shape being the predominant shape (69.5%). Most of the studied fruits had one nut per fruit (96.8%). Genetic estimate indicated high heritability for all traits studied including kernel weight (H2=0.78) and pulp weight (H2=0.78). Genetic advance estimates were also very high for all traits studied, favoring gains in crop improvement. There was also high correlation between fruit weight and fruit length (0.743), fruit width (0.67), pulp weight (0.921), dry nut weight (0.701) and kernel weight (0.678). Genetic diversity study was conducted using discriminant analysis of principal components (DAPC) from DART sequence single nucleotide polymorphism data. Three major clusters were observed with several subclusters, genetic diversity was Fst = 0.0022, observed heterozygosity was 0.034, while the expected heterozygosity ranged between 0.015 to 0.03, all pointing to the possibility of inbreeding within the shea germplasm studied. The first established shea population was evaluated for growth and vigor. There was significant differences between the populations for all growth traits studied. A Pearson correlation between the traits indicated significant positive correlation between stem diameter and plant height (0.95), stem diameter and petiole length (0.82), leaf blade length and leaf width (0.89). Clustering of the traits using UPGMA revealed two 36% dissimilarity separating the low vigor populations from the vigorous ones.