Context

Tsetse transmitted human or livestock trypanosomiasis is one of the major constraints to rural development in sub-Saharan Africa. The distribution of the fly and its abundance play an important role in the epidemiology of the disease. Tsetse distribution and abundance forms the basis for intervention programmes hence there is need for up- to date and informative maps. It is also important to understand the interactions between tsetse and other factors within its environment, both biotic and abiotic.

Research Questions

1. What is the distribution pattern of tsetse flies in the Masoka Communal area of Mbire District?

2. What are the drivers of tsetse abundance in the Masoka Communal area of Mbire District?

Objectives

1.To model the spatial distribution of tsetse in Masoka communal land using the Maximum Entropy (MaxEnt) species distribution model

2. To determine the relationship between wild hosts, livestock, human populations and abundance of tsetse in Masoka communal land

The Study Area

The study will be conducted in Masoka, a community in the Mid-Zambezi Valley,Northern Zimbabwe. The settlement has got approximately 1632 inhabitants (Zimstat,2012) whose agricultural activities include cattle and goat rearing, small grainsproduction. Cotton is the major cash crop grown in the area. The community shares aboundary with Chewore Safari Area which is a protected wildlife area. An area of400km2 (16. 1° to 16.28° S and 30.1° to 30.28° E) were no tsetse interventionprogrammes have been implemented has been demarcated for the collection of tsetsedata which will be used to run the model.

Sampling Protocol

The study area will be subdivided into 5km x 5km grids and epsilon traps baited withmethyl ethyl ketone (MEK) and phenol sachets (3-n-propylphenol, octenol, 4methylphenol and acetone) will be used to survey for tsetse. There will be13 trap sitesin each 25km2 grid, taking into account the different strata defined by the supervisedclassification exercise as well as accessibility. Global positioning system (GPS)receivers will be uploaded with the coordinates of the trap positions and will be used innavigation. The traps will be monitored after 2 days and uplifted when there are tsetsecatches whilst those without catches will be left in position for a further 7 days.Information on trap location (geo-reference), altitude, date ofdeployment/monitoring/upliftment, type of surrounding vegetation and tsetse catcheswill be recorded on an entomological data sheet.

Climatic and Environmental Data Sets

Climatic and environmental data will be derived from a time series of MODIS with aspatial resolution of 250× 250 m. A standard deviation of the following variables willbe used to run the MaxEnt model:• Normalized Difference Vegetation Index (NDVI)• Land Surface Temperature (Air) (day and night)• Mid InfraRed (Ground Temperature)A supervised classification of SPOT images with a spatial resolution of 2.5m x 2.5mwill be conducted in order to obtain a land cover layer of the area.

Species Distribution Model

The tsetse presence data will be run using the Maximum Entropy Species Distribution(MaxEnt) model using climate and environmental layers as the habitat suitabilitydeterminants.70% of the tsetse presence data will be used to run the model whilst 30%will be used to validate the model. The performance of the model will be assessedusing the Area Under The Curve generated from the Maxent model – values closer to 1indicate a better model. The contribution of each variable will be assessed using theJackknife of Regularized Training Gain.

Tsetse Abundance Modeling

The abundance of tsetse will be computed as the average catch per trap per day. Thepredicted tsetse apparent density (ADT) per month per km2 will be computed using aspatio-temporal statistical model fitted against the monthly temperature (DLST),vegetation (NDVI) and Elevation (DEM). A negative binomial model with spatialrandom effects will be used. The abundance model will be overlaid with livestock,human and wildlife datasets and analysed for correlation.