Modelling Productivity of Indigenous Zebu Cattle (Bos Indicus) under the Natural Field conditions on Rusinga Island, Kenya

Abstract

The general objective of this study was to develop predictive models which could be used to determine p roductivity (liveweight gain) losses attributed to various causal factors that are specific to livestock production systems as found in typical communities of small - scale peasant farmers in Africa. Factors of special interest were infestations by heliminths, arthropod vectors and their associated parasites, notably ticks and Thei1eria spp of parasites, tsetse flies and Trypanosoma spp . It was also noted that in the absence of diseases, nutrition was most important in determining calf growth and development. The goal of the study was to develop models which could help managers and policy-makers in undertaking economic evaluation and monitoring of projects and programmes on tick control strategies in the tropical Africa, with particular emphasis on cost-benefit analysis. Africa's livestock production system i s predominantly characterised by the zebu cattle (8. indicus) with the destructive ticks and tsetse flies as the common disease vectors. The study was conducted on Rusinga Island on the eastern edge of Lake Victoria in western Kenya. The Island is typical of many tropical regions of Africa in relation to livestock production systems and their associated causal factors that claim their toll on the animal's productivity. The major livestock ticks (Acarina: Ixodidae) are found on the Island. The farmers were randomly selected for the study. These farmers had a total of two hundred sixty five cattle by December, 1987. This study involved seventy four calves born in both 1986 and 1987. The ten farms were visited every month to collect data on the number of adult ticks on each calf categorized by species · and sex, calf weight, as well as pasture quality. Data on weather conditions was also monitored. Information on morbidity and epidemiology of cattle diseases on the Island were secured from other secondary sources. Generally, nutrition quality of pastures in terms of crude protein, phosphorus, potassium, calcium and magnesium was satisfactory in 1987. On the basis of crude protein, phosphorus and calcium contents of pastures, the ten farms could be divided into three broad nutrition regimes. Farms with the highest nutritious pastures were located to the southern and western parts while the poorest to the eastern and north-eastern sides of the Island. The dominant tick was Rhipicephalus appendiculatus, followed by Amblyomma variegatum, R. evertsi and Boophilus decoloratus in that order of abundance. The spatial distribution of the ticks was found to conform to the pattern depicted by the nutrition regimes especially crude protein contents. Strong association was found between certain nutrients and tick species. It was revealed that crude protein was the most important nutrient in increasing host resistance to R. appendiculatus and B. decoloratus; two very destructive vectors of tick-borne diseases. Also important were potassium and calcium. No significant differences were detected in birth weights between farms and sexes. Observed birth weights varied between 11 kg and 23 kg. Mean birth weights were 16.5 kg and 15.2 kg for bullocks and heifers, respectively. Calf birth weights were found to be associated with crude protein contents of pastures in which the respective dams were grazed. Cal f growth was found to be significantly different between farms (P < 0.0001). Calf growth was, however, consistent to calf birth weight, pasture quality as well as tick burden. Calves with highest birth weights experienced greatest growth rates and vice versa. The relationship between nutrition and host resistance to ticks was compatible to calf growth patterns. Poor growth was found also to be the result of interplay of the two causal factors. No significant differences were detected in calf growth between the sexes and also between the months of birth. Observed growth rates mostly varied between 0.0159 to 0.359 lbs per day, (i.e . 0.0072 kg to 0.1632 kg per day). This represented a mean annual growth of between 26 to 60 kgs. The lowest growth rate was 0.009 lbs per day (0.0041 kg per day) while the highest 0.409 lbs per day (0.1859 kg per xxviiiday). These calves experienced mean tick burden of between 65 to 91 ticks per calf. The upper economic threshold levels were also determined and found to be dependent on the age of a calf. In this study, the oldest calf was 33 months for which the upper economic threshold was found to be about 145 ticks per calf and which was comparable to a figure of 158 for B. taurus X B. indicus steers found in Australia. It was found that the best model for simulation of calf growth was the modified Gompertz model. The model was found to adequately describe the growth behavi our of B. indicus calves on the Island. The model was found to be flexible and comprehensively explained the major growth features such as growth rate, age at which maximum growth rate is attained, and maximum potential growth of a breed under given production system. A relatively new apprach based on the minimum liveweights of cal ves for given ages, was used to develop a survivorship threshold model. The survivorship threshold mode l is the lowest liveweight for which a calf would be under the risk of dying due to starvation and ill - health. The exponential model was found to provide the best fit to the data. Because of its functional invariance, the model would be a useful tool in monitoring cattle productivity in similar agroecological zones in Africa. The study showed that mean calf growth rate, rk+j, was related to the mean age, Uk, and mean adult tick burden, mk, ixof a herd of calves as follows: loge (n+j) = Oj + f3jUk + 'l:jm*k k=O, 1, 2, 3, .... , K j = 0, 1, 2, ........ . where rk+j is the predicted growth rate in j months to come given that the calves are aged k months today, m•k = {mk, K is the maximum age in months of calves in the herd. Several models were developed for specific tick burdens and nutrition conditions on the Island. It was found that the relationship was strongest when the lag variable, j, was two months, i.e. j = 2. The usefulness of the models were then discussed specifically with reference to their applications in microeconomic and cost-benefit analyses of tick control and management projects and programmes in Africa.