Ecological and Behavioural Studies of Mosquitoes in Mwea Tebere Irrigation Scheme, Kirinya District, Kenya with special reference to Anopheles Arabensis (Diptera : Culicidae)

Abstract

During this investigation which took place between April 1989 and February 1991, an attempt was made to determine the impact of rice irrigation practices on 1) mosquito species diversity and their relative population density; 2) malaria infection rates in both the human and mosquito populations. The studies also investigated the relative importance and attractiveness of various hosts for different mosquito species. Mosquitoes were sampled from two study villages, Mbui Njeru and Mathangauta which are located in the Mwea Irrigation scheme. The two villages were selected mainly because their positions relative to irrigated areas offered a good contrast for comparative studies on mosquitoes and their hosts. Mbui Njeru lies in the centre of the irrigation scheme and is thus surrounded by paddies from all sides. Mathangauta on the other hand lies at the periphery and is encroached by irrigated paddies on only one side. The two villages are also accessible throughout the year. Mosquito sampling was carried out in each of the four houses in ea.ih village. Sampling replicat.od daily fo: seven days each month in every village. Sampling methods for adult; mosquitoes included collections from daytime indoor and outdoor resting sites using battery rawered aspirators and pyrethrum knock-down spray. Mosquitoes entering houses for feeding or resting purposes were also collected using miniature light traps. LBrvae were collected from flooded rice paddies and pools of stagnant water using standardized dipping ~methods. Mosquitoes collected by the various paddies the relative numbers of An. arabiensis decreased. In Mathangauta village the two vector species alternated in their predominance. When the numbers of An. arabiensis was high that of An. funestus declined and vice versa. The increase in An. arabiensis coincided with the preparation of nurseries and seedling transplantation while the increasG in An. funestus coincided with the draining of water from the paddies and harvesting. Irrigation water released from paddies during harvesting found its way into drainage canals with dense submerged vegetation, thus forming suitable breeding sites for An. funestus. Variation in the numbers of An. arabiensis in Mbui Njeru was influenced by seasonal rainfall pattern in addition to rice cultivation cycle. Mosquito numbers were low during the rainy season, probably due to wash off effects of breeding sites by the rain water, but increased during the dry season. The mean counts between the two vector species An. arabiensis and An. funestus were also significantly different with the former being the most abundant and present throughout the year. The species from both indoor and out dnor ites were also significantly difference more mosquitoes recorded indoors than outdoors. Suprisingly, it was also noted that the incidence of malaria was high when the relative numbers of An. arabiensis was low-compared.to the other months. It is therefore likely that the main vector responsible for transmission of malaria in Mwea Tebere is An. funestus. Investigations on the sporozoite rates of dissected female mosquito vectoLs showed. that the monthly average sporoLoite rate was not significantly different between the two villages (x2=0.303;p,<0.05) The difference in the sporozoite rate with respect to the seasons was also not significant (X 2 = 2.25; P>0.05 All the mosquitoes (n = 4594) tested for sporozoite rate using the Enzyme-linked Immunosorbent Assay (ELISA) were negative. This confirmed earlier observations of low sporozoite rates through dissections of salivary glands. Studies on resting behaviour of female Anopheles arabiensis and Anopheles funestus showed that freshly fed mosquitoes preferred to rest indoors. A comparison of the resting behaviour of An. arabiensis and An. rufipes showed that fed females of the latter rested predominantly outdoors. Results on feeding behaviour determined through blood meal analysis showed that An. arabiensis fe'd predominantly on bovine hosts (79%). The difference in the numbers feeding on human and bovine hosts was significant ((P< 0.05) with more An. arabiensis feeding on bovine hosts. Malaria parasite prevalence rate in the human population for Mbui Njeru and Mathangauta was less than 8.7% for all age and sex gemodiuin tal c iperum was the predominant .,.sr>ecies comprising 100% of the infections recorded. The malaria prevalence rate for Mbui Njeru was higher than that of Mathangauta village (Females 3.7% = 12.63; P (F" = 6.27; P<O.Ol There was also) signif icant difference in the malaria infection rate among the various age-groups. The age-group with the highest infection rate was the 10-14 year olds. The overall outcome of this study was that the numbers of mosquitoes were enough to maintain malaria transmission throughout the year in the area studied. However, the parasite infection rate in both vector and human population was Lowe r than would be expected in the presence of such high numbers of vector mosquitoes. This situation was possibly due to the fact that the vector species were predominantly feeding on bovine hosts than on human hosts. The tendency by mosquitoes to feed more on animals than human beings is referred to as zoophile. Such feeding behaviour may form the basis of zooprophylaxis, an important method through which man-vector contact could be minimised. Zoo prophylaxis is a practical malaria control. method in which the community can participate by being encouraged to keep at least a few cows outside their houses. However, there is need for further studies to determine the optimal densities, directions and distances at which such barrier animals )loved, to ensure that they on the other hand do not worsen the situation by attracting large numbers of potential vectors to human habitations.