Diversity and pollination activity of flower visiting insects associated with avocado along the slopes of Taita Hills in Kenya

Abstract

Avocado is a horticultural crop of increasing economic importance in Kenya and around the world. Avocado production has been declining partly due to inadequate pollination. This decline in avocado pollination is currently linked to habitat fragmentation and climate change that negatively affects insect pollinators, especially in high altitude areas such as Taita Hills in coastal Kenya. However, there is scarcity of information on avocado flower visiting insects in Taita Hills. The main objective of this study was to assess the diversity and pollination activity of flower visiting insects associated with avocado along the slopes of Taita Hills. The research was conducted in 18 sampling points along the slopes of Taita Hills. In each of the sampling points, 4 avocado trees were selected for sampling of insect diversity using sweep nets and aerial colored pan traps. Frequency of flower visits (i.e. the number of visits by an insect species at any given time), visitation rate (i.e. average time spent on an individual avocado flower and the average number of flowers visited in an observation quadrat) and quantity of avocado pollen grains removed from an insect body were used in determining the insect that was the most active flower visiting insect associated with avocado. Alternative food sources for avocado flower visiting insects were determined by capturing insects on two ~50 × 50 m plots using a sweep net and relating them to those captured on avocado. Statistical analysis was performed using SAS 9.1 software (SAS Institute Inc.). Monthly variation in insect richness and abundance was computed using a χ2 square test. The relationship between the overall insect diversity and species richness with altitude as well as insect abundance with altitude, time and weather variables was analyzed using a Pearson correlation coefficient in proc corr. Shannon diversity index for insect diversity was produced using R version 2.10.0. Frequency of pollinator visits in an observation quadrat were generated using proc freq. Means of time spent on an individual avocado flower and the number of flowers visited by an insect in an observation quadrat were computed using proc means while analysis of variance (ANOVA) computed using proc mixed with LS Means for mean separation. Pearson correlation determined the relationship between pollen load on an insect body with altitude, time, temperature and humidity. Frequency of bee visits to non avocado host plants was compared using a χ2 square test in proc freq. Altogether, 1081 individual arthropods were collected that belonged to at least 98 insect species (980 individuals), 1 Arachnid (Araneae) and 100 individual Dipteran fauna which could not be identified or grouped into morphotypes. The arthropods belonged to the orders Hymenoptera, Diptera, Coleoptera, Lepidoptera, Hemiptera, Mantodea and Neuroptera. Insect diversity was significantly higher in the mid zone (Hʼ=2.51) compared to the high zone (Hʼ=2.29). Furthermore, insect diversity decreased with increase in altitude (r=-0.5, p<0.05). Honeybee (Apis mellifera) had the highest quantity of avocado pollen bound on their body and was the most frequent flower visiting insect in a quadrat thereby considered as the most active flower visiting insect. Temperature inversely correlated with altitude and had differential influence on the diversity and pollination activity of avocado flower visiting insects. Information generated by this study will help in monitoring the effects of climate change on diversity and pollination activity of flower visiting insects associated with avocado.