Factors affecting seasonal dispersal of the tsetse flies Glossina pallidipes and G. longipennis (Diptera: Glossinidae) at Nguruman, south-west Kenya

Abstract

Seasonal changes in the distribution of the tsetse flies, Glossina pallidipes Austen and G. longipennis Corti, along a transect from riverine thickets out into open plains were monitored along with tsetse density, climatic factors, vegetation and host abundance. Dispersal of tsetse into open country was quantified using the mean spread. During and after the rains both species extended their distribution out into open country up to at least 3.5 km from riverine thicket areas. The mean spread of G. longipennis was greater than that of G. pallidipes. The spread of males and females was very similar, as was the spread of different age categories of parous females, but nulliparous flies and females carrying second instar larvae were under-represented in samples from open areas. This seasonal dispersal can be accounted for by random diffusion with an average root mean square displacement of about 175 m per day. Observed degrees of spread were best correlated with humidity conditions prior to sampling, but multiple regression models suggested that host abundance, vegetation as measured by Normalised Difference Vegetation Indices (NDVI), and, in the case of G. longipennis, tsetse density, were also factors in determining the degree of spread. The significance of these findings in relation to tsetse control is discussed. 5-r © 9 © © 0-5 km N 4 Trap pair © -1'55'S 36 1 05'E

Seasonal distribution and abundance of tsetse flies (Glossina spp.) in the Faro and Deo Division of the Adamaoua Plateau in Cameroon

“…As a result, tsetse are more confined to those areas where wildlife is present. The association between the distribution of savannah species of tsetse flies and their hosts has been observed elsewhere (Brightwell et al, 1992;Rawlings et al, 1994;Van den Bossche & De Deken, 2002). In all these cases sudden changes in the distribution and density of cattle resulted in substantial changes in the distribution and density of tsetse flies.…”

A dynamic species distribution model of Glossina subgenus Morsitans: The identification of tsetse reservoirs and refugia

“…Efforts to control tsetse have been chronically hampered by identification of infested areas, reinvasion of tsetse into previously controlled regions, and substantial costs associated with the means of control (Brightwell et al 1992, Williams et al 1992a, Hargrove 2003. Traditionally, tsetse distributions have been aggregated and mapped into contiguous ''fly belts,'' which can contain one or more tsetse species with boundaries set by a wide variety of physical, biological, and anthropogenic barriers (Ford 1971, Ford and Katondo 1977, Rogers and Robinson 2004, Muriuki et al 2005.…”

“…More recent attempts to map tsetse distributions have implemented spatial modeling techniques, which have succeeded in producing maps with both higher spatial resolutions than the fly belt maps, and estimates on the probability of tsetse presence (e.g., Rogers and Williams 1994, Gilbert et al 2001, Wint 2001. Although an improvement over the fly belt maps, the modeled predictions thus far have provided little information about intra-and inter-annual fluctuations in tsetse distributions, despite numerous studies that discuss such phenomenon (e.g., Austen and Hegh 1922, Nash 1933, Bursell 1956, Brightwell et al 1992, Hargrove 2001, Odulaja et al 2001, Bett et al 2008.…”

“…Williams et al (1992b) reported a Zimbabwe dry season NDVI threshold value of 0.39 based on several statistical tests (a linear regression, a non-linear regression, discriminate analysis, k-nearest neighbor analysis, and experiments using a neural network). Given that the two species of tsetse studied in Williams et al (1992) are two of the moristans group species found in Kenya, the woody vegetation types found in Zimbabwe are similar to those found in Kenya (Vale et al 1988, Rogers and Williams 1994, Muzari 1999, and high synchronicity between NDVI and precipitation events in semiarid environments (Park 2009), we adopted 0.39 as the threshold value for NDVI-based moisture suitability. Furthermore, to distinguish our use of NDVI as a surrogate for moisture as opposed to previous studies that have used it as a land cover descriptor, the NDVI variable will henceforth be referred to as the moisture variable.…”

“…Normally in the dry season, tsetse population densities are low (Brightwell et al 1992); during prolonged droughts, tsetse densities are even lower. This severe drought likely reduced fly population densities to such a low level that a long-term intensive monitoring operation covering the whole reservoir area would be needed to conclusively test whether or not tsetse were present (Leak et al 2008).…”