fadzai zengeya - Academia.edu (original) (raw)

Papers by fadzai zengeya

Declaration 1: Originality I hereby declare that this thesis submitted for the Doctor of Philosop... more Declaration 1: Originality I hereby declare that this thesis submitted for the Doctor of Philosophy degree in Geography and Environmental Science (Spatial Ecology) at the University of Zimbabwe is my original work and has not been previously submitted to any institution of higher education. I further declare that all sources cited or quoted are indicated by means of a comprehensive list of references.

Ecology and Evolution

Understanding key overlap zones and habitats which are intensively shared by species in space and... more Understanding key overlap zones and habitats which are intensively shared by species in space and time is crucial as it provides vital information to inform spatial conservation with maximum benefits. The advent of high‐resolution GPS technologies associated with new analytical algorithms is revolutionizing studies underpinning species spatial and social interaction patterns within ecosystems. Here, using a robust home range estimation algorithm, the autocorrelated kernel density estimator (AKDE) equipped with an equally powerful home range overlap metric, the Bhattacharyya's coefficient (BC), we provide one of the first attempts to estimate and delineate spatial home range overlap zones for critically endangered African white‐backed vultures to inform conservation planning. Six vultures were captured in Hwange National Park using a modified cannon net system after which they were tagged and tracked with high‐resolution GPS backpacks. Overall, results suggested weaker average ho...

Transfrontier Conservation Areas, 2017

International audienc

The distribution of herbivores is hypothesized to be influenced by an interplay of several biophy... more The distribution of herbivores is hypothesized to be influenced by an interplay of several biophysical factors as well as human factors. However, our current understanding of herbivore movement dynamics and factors driving the movement and interaction between domestic and wild herbivores at interfaces between wildlife and livestock areas is limited. In this regard, the development of methods and approaches that allow spatially explicit (1) continuous monitoring, and (2) prediction of animal movements in relation to biophysical and human factors is critical for an improved understanding of livestock and wildlife distribution at these interfaces. The advent of Global Positioning Systems (GPS), remote sensing and improved spatial analytical frameworks has provided opportunities to accomplish this task. In this thesis, we develop novel methods and approaches using GPS and remote sensing to understand factors influencing the distribution of livestock, i.e. cattle (Bos taurus) and their s...

International Journal of Applied Earth Observation and Geoinformation, 2018

Malaria remains a major public health problem and a principal cause of morbidity and mortality in... more Malaria remains a major public health problem and a principal cause of morbidity and mortality in most developing countries. Although malaria still presents health problems, significant successes have been recorded in reducing deaths resulting from the disease. As malaria transmission continues to decline, control interventions will increasingly depend on the ability to define high-risk areas known as malaria hotspots. Therefore, there is urgent need to use geospatial tools such as geographic information system to detect spatial patterns of malaria and delineate disease hot spots for better planning and management. Thus, accurate mapping and prediction of seasonality of malaria hotspots is an important step towards developing strategies for effective malaria control. In this study, we modelled seasonal malaria hotspots as a function of habitat suitability of Anopheles arabiensis (A. Arabiensis) as a first step towards predicting likely seasonal malaria hotspots that could provide guidance in targeted malaria control. We used Geographical information system (GIS) and spatial statistic methods to identify seasonal hotspots of malaria cases at the country level. In order to achieve this, we first determined the spatial distribution of seasonal malaria hotspots using the Getis Ord G i * statistic based on confirmed positive malaria cases recorded at health facilities in Zimbabwe over four years (1996-1999). We then used MAXENT technique to model habitat suitability of A. arabiensis from presence data collected from 1990 to 2002 based on bioclimatic variables and altitude. Finally, we used autologistic regression to test the extent to which malaria hotspots can be predicted using A. arabiensis habitat suitability. Our results show that A. arabiensis habitat suitability consistently and significantly (p < 0.05) predicts malaria hotspots from 1996 to 1999. Overall, our results show that malaria hotspots can be predicted using A. arabiensis habitat suitability, suggesting the possibility of developing models for malaria early warning based on vector habitat suitability.

International Journal of Geographical Information Science, 2014

In this study, we used remotely sensed Enhanced Vegetation Index (EVI) as a measure of food abund... more In this study, we used remotely sensed Enhanced Vegetation Index (EVI) as a measure of food abundance to explain intraspecific home range overlap of Global Positioning System (GPS) collared semi-free range herbivores (Bos taurus). We then tested whether seasonality (estimated using the coefficient of variation of EVI) was related to intensity of home range overlap. We also tested whether the intensity of home range overlap was related to distance from water, particularly during the dry season. We determined the intensity of home range overlap using the Utilization Distribution Overlap Index. We then modelled the intensity of home range overlap as a function of seasonality. Results show that intensity of home range overlap varied with seasonality in a non-linear U-shaped manner. In addition, results showed that at both high and low levels of seasonality, intensity of home range overlap was high while it was low at moderate levels of seasonality. Results also indicated that the intensity of home range overlap increased with distance from water during the dry period. The U-shaped relationship obtained in this study conform to the behavioural theory that predicts similar relationships between food abundance and territoriality, indicating that remotely sensed productivity is an ecologically meaningful measure of food abundance. This further amplifies the utility of the combination of GPS animal movement data and remotely sensed data in spatial ecology.

Ostrich

Application of spatial data is becoming increasingly important as a tool for mapping and understa... more Application of spatial data is becoming increasingly important as a tool for mapping and understanding space utilisation patterns especially for critically endangered species. In this study, we tested how variation in the GPS fix interval may influence space utilisation patterns by African White-backed Vultures in Hwange National Park and the surrounding areas. In order to test our objectives, we collected data from six African White-backed Vultures using solar powered GPS-GSM units. The data were collected for periods ranging from five months to nineteen months (28 June 2016 to 30 January 2018), resulting in 89 809 total GPS fixes. We tested whether home range and core areas estimated at different fix interval differ significantly. We also tested whether African White-backed Vultures show significant site fidelity and at what sampling frequency the estimated home ranges tend to reach an asymptote. We used total individual data and seasonal data to test these hypotheses. In order to estimate home range size at different fix intervals, we used the Brownian Bridge Movement Model, because unlike other traditional estimators, it explicitly incorporates temporal sequence in GPS fixes, accounts for autocorrelation in movement data, as well as missing GPS data. Home range estimates differed significantly at different GPS fix intervals. Home range and core areas for five vultures were concentrated within conservation areas. African White-backed Vultures also showed significant site fidelity across seasons except for vulture three, which demonstrated nomadic tendencies. Results on home range asymptote were not conclusive, because of unique individual vulture space use patterns at different fix intervals. Our study was the first to track and monitor space utilisation patterns by mature African White-backed Vultures in the Hwange ecosystem. Accordingly, our results have critical conservation implications for the wide-ranging, critically endangered species. L'intervalle entre plusieurs points GPS montre des variations dans la manière chez le vautour africain Gyps africanus d'utiliser l'espace L'application des données spatiales devient de plus en plus importante en tant qu'outil de cartographie et de compréhension des modèles d'utilisation de l'espace, en particulier pour les espèces en danger critique d'extinction. Dans cette étude, nous avons testé comment la variation de l'intervalle de fixation GPS peut influencer les schémas d'utilisation de l'espace par les vautours africains dans le parc national du Hwange et les zones environnantes. Pour tester nos objectifs, nous avons recueilli des données auprès de six vautours africains à l'aide d'appareils GPS-GSM fonctionnant à l'énergie solaire. Les données ont été recueillies sur des périodes allant de cinq à dix-neuf mois (du 28 juin 2016 au 30 janvier 2018), ce qui a permis d'obtenir 89 809 positions GPS au total. Nous avons vérifié si le domaine vital et les zones centrales estimées à différents intervalles de fixation diffèrent de manière significative. Nous avons également testé si le vautour africain montre une fidélité significative au site et à quelle fréquence d'échantillonnage les domaines vitaux estimés ont tendance à atteindre une asymptote. Nous avons utilisé les données individuelles totales et les données saisonnières pour tester ces hypothèses. Pour estimer la taille du domaine vital à différents intervalles de points fixes, nous avons utilisé le modèle de mouvement du pont brownien car, contrairement aux autres estimateurs traditionnels, il incorpore explicitement une séquence temporelle dans les points fixes GPS, tient compte de l'autocorrélation dans les données de mouvement ainsi que des données GPS manquantes. Les estimations de la distance parcourue par le domicile diffèrent sensiblement selon les intervalles de temps entre les points GPS. Le domaine vital et les zones centrales pour cinq vautours ont été concentrés dans des zones de conservation. Les vautours africains ont également montré une fidélité significative au site au fil des saisons, à l'exception du vautour trois qui a montré des tendances nomades. Les résultats sur l'asymptote du domaine vital n'ont pas été concluants en raison des modes d'utilisation de l'espace propres à chaque vautour à différents intervalles de temps. Notre étude est l'une des premières à suivre et à surveiller les schémas d'utilisation de l'espace par les vautours africains adultes dans l'écosystème du Hwange. Nos résultats ont donc des implications cruciales pour la conservation de cette espèce à large distribution et gravement menacée.

The understanding of animal distribution in habitats located farther from water sources has not b... more The understanding of animal distribution in habitats located farther from water sources has not been dealt with adequately in the literature, yet this knowledge enables better prediction of species occurrence across an entire landscape. We tested whether elephant occurrence peaks away from water in addition to the known peak that is associated with water sources. We used the Maximum Entropy Modelling (MaxEnt) algorithm to predict the potential distribution of elephants in the Gonarezhou National Park, Zimbabwe. Elephant tracking data from Global Positioning System (GPS) collars were used as the response variable while NDVI (a proxy for forage quantity) and water sources data were the environmental variables. Results showed multiple peaks of elephant occurrence with increasing distance from water sources. Additionally, results illustrated that the peaks occur in high NDVI areas. Our findings emphasise the utility of GIS and remote sensing in enhancing our understanding of animal occu...

The coexistence of protected areas and neighbouring communal areas in Southern Africa is jeopardi... more The coexistence of protected areas and neighbouring communal areas in Southern Africa is jeopardised by negative interactions between wildlife and villagers and their livestock. The African buffalo (Syncerus caffer) has often been blamed for a significant proportion of the “human-wildlife conflicts”, mainly through the transmission of pathogens to cattle, competition with livestock for grazing and water resources, and also occasionally involved in crop destruction and threat to people's physical integrity. We present the results of several multidisciplinary studies carried out since 2008 on buffalo-cattle interactions at the periphery of Gonarezhou National Park, Hwange National Park and adjacent communal lands in Zimbabwe. Movements of sympatric buffalo and cattle have been described at various scales using GPS collars, revealing daily/seasonal patterns and occasional events of long-distance dispersal of young buffalo females. Habitat preferences, especially for open grassland ...

Cogent Environmental Science, 2017

African Journal of Range & Forage Science, 2020

We investigated application of MaxEnt, a one-class classifier, in mapping the spatial distributio... more We investigated application of MaxEnt, a one-class classifier, in mapping the spatial distribution of Colophospermum mopane, Dichrostachys cinerea and Salvadora persica using drainage, elevation, slope, soil and Normalised Difference Vegetation Index as environmental variables. Model performance was evaluated based on the area under the ROC curve (AUC), Kappa and Total Skills Statistic. The AUC results demonstrated the high predictive power of MaxEnt as test data values of all species, respectively, were 0.694, 0.754 and 0.998 (p < 0.05). Elevation contributed the most in explaining spatial distributions of all species. Results also indicated that several one-class species maps can be integrated into one species distribution map. We showed that C. mopane is likely to co-occur with D. cinerea and S. persica, whereas S. persica is likely to co-occur with D. cinerea. However, there was no habitat suitable for co-occurrence of all species. One-class species mapping can therefore be successful in heterogeneous agricultural landscapes.

Transactions of the Royal Society of South Africa, 2019

The African baobab (Adansonia digitata, L.) is one of the most important tree species in the Afri... more The African baobab (Adansonia digitata, L.) is one of the most important tree species in the African savannah, yet there is little knowledge of its spatial distribution and abundance. In this study, we modelled the spatial distribution of baobab in Zimbabwe as a function of key biophysical factors using Maximum Entropy (MaxEnt) and also estimated its abundance. Results show that altitude, mean temperature of the driest quarter and mean temperature of the coldest quarter are the key factors explaining baobab distribution in Zimbabwe. The population of baobabs in Zimbabwe was estimated to range from 3 980 070 to 4 441 697 at 95% confidence. The results suggest we can estimate the spatial distribution as well the abundance of baobabs. This implies that any sustainable utilisation of baobabs and their products should take into consideration their distribution and abundance.

Geocarto International, 2018

Malaria burden has considerably declined in the last 15 years mainly due to large-scale vector co... more Malaria burden has considerably declined in the last 15 years mainly due to large-scale vector control. The continued decline can be sustained through malaria risk stratification. Malaria stratification is the classification of geographical areas according to malaria risk. In this study, ecological niche modelling using the maximum entropy algorithm was applied to predict malaria vector habitat suitability in terms of bioclimatic and topographic variables. The output vector suitability map was integrated with malaria prevalence data in a GIS to stratify Zimbabwe into different malaria risk zones. Five improved and validated malaria risk zones were sucessfully delimited for Zimbabwe based on the World Health Organisation classification scheme. These results suggest that the probability of occurrence of major vectors of malaria is a key determinant of malaria prevalence. The delimited malaria risk zones could be used by National Malaria Control programmes to plan and implement targeted malaria interventions based on vector control.

Geocarto International, 2017

Abstract Waterholes are a key resource that influences wildlife distribution in semi-arid ecosyst... more Abstract Waterholes are a key resource that influences wildlife distribution in semi-arid ecosystems. Mapping waterholes can guide intervening decisions for supplementing water resources and managing wildlife distribution patterns. Although remote sensing provides a key to mapping distribution of waterholes, efficiency of existing remotely sensed methods for detecting waterholes have to be evaluated and even new ones developed. In this study, we evaluated performance of the Modified Normalized Difference Water Index (MNDWI) and Superfine Water Index (SWI) at selected optimum thresholds. Kappa results indicated that MNDWI detects waterholes better than SWI. We further validated MNDWI detected waterholes by testing response of waterhole area to temporal rainfall variability and waterhole persistence to spatial rainfall variability. Extent of MNDWI-detected waterholes varied in relation to temporal rainfall variability (p < 0.05). Waterhole persistence was not associated with spatial rainfall variability which could be explained by differences in waterhole types or low spatial rainfall variability.

Cogent Food & Agriculture, 2017

Rangeland productivity in semi-arid areas is adversely affected by increased variability in preci... more Rangeland productivity in semi-arid areas is adversely affected by increased variability in precipitation and frequency of droughts, coupled by increased livestock numbers. Knowledge on key rangeland resources that have capacity to increase resilience of livestock based rural livelihoods is critical for ensuring their sustainability. In this study, we identified key browse species used by livestock during the dry season, and determined their multiple uses in a semi-arid rangeland of Zimbabwe. Random sampling was used to select 138 respondents for participating in individual qualitative questionnaires, and seven key informants for a focus group discussion. The Cultural Significance Index was calculated to determine the importance of the key browse species identified. An index to determine risk associated with competitive use of key browse species based on individual species uses and relative abundance as an indicator for species sustainability was also introduced. Twenty-eight key species used as browse by livestock and wildlife, and for ethnoveterinary and human medicines were identified. Species that were common to all uses constituted 25% (n = 7) of the total. No species (n = 0) had a single purpose only or, were used for both medicines and firewood/timber. Therefore, screening key browse species facilitates their sustainability.

African Journal of Ecology, 2016

Applied Geography, 2016

In this study, we developed a stable and temporally dynamic model for predicting tsetse (Glossina... more In this study, we developed a stable and temporally dynamic model for predicting tsetse (Glossina pallidipes) habitat distribution based on a remotely sensed Normalised Difference Vegetation Index (NDVI), an indicator of vegetation greenness, and topographic variables, specifically, elevation and topographic position index (TPI). We also investigated the effect of drainage networks on habitat suitability of tsetse as well as factors that may influence changes in area of suitable tsetse habitat. We used data on tsetse presence collected in North western Zimbabwe during 1998 to develop a habitat prediction model using Maxent (Training AUC=0.751, test AU=0.752). Results of the Maxent model showed that the probability of occurrence of G. pallidipes decreased as TPI increased while an increase in elevation beyond 800 m resulted in a decrease in the probability of occurrence. High probabilities (>50%) of occurrence of G. pallidipes were associated with NDVI between high 0.3 and 0.6. Based on the good predictive ability of the model, we fitted this model to environmental data of six different years, 1986, 1991, 1993, 2002, 2007 and 2008 to predict the spatial distribution of tsetse presence in those years and to quantify any trends or changes in the tsetse distribution, which may be a function of changes in suitable tsetse habitat. The results showed that the amount of suitable G. pallidipes habitat significantly decreased (r 2 0.799, p=0.007) for the period 1986 and 2008 due to the changes in the amount of vegetation cover as measured by NDVI over time in years. Using binary logistic regression, the probability of occurrence of suitable tsetse habitat decreased with increased distance from drainage lines. Overall, results of this study suggest that temporal changes in vegetation cover captured by using NDVI can aptly capture variations in habitat suitability of tsetse over time. Thus integration of remotely sensed data and other landscape variables enhances assessment of temporal changes in habitat suitability of tsetse which is crucial in the management and control of tsetse.

African Journal of Ecology, 2016

Understanding factors affecting the distribution of the African elephant is important for its con... more Understanding factors affecting the distribution of the African elephant is important for its conservation in increasingly human-dominated savannah landscapes. However, understanding how landscape fragmentation and vegetation productivity affect elephant habitat utilization remains poorly understood. In this study, we tested whether landscape fragmentation and vegetation productivity explain elephant habitat utilization in the Amboseli ecosystem in Kenya. We used GPS (Global Positioning System) telemetry data from five elephants to quantify elephant habitat utilization. Habitat utilization was determined by calculating the time elephants spent within a unit area. We then used generalized additive models (GAMs) to model the relationship between time density and landscape fragmentation, as well as vegetation productivity. Results show that landscape fragmentation and vegetation productivity significantly (P < 0.05) explain elephant habitat utilization. A significant (P < 0.05) unimodal relationship between vegetation productivity and habitat utilization was observed. Results suggest that elephants spend much of their time in less fragmented landscapes of intermediate productivity.

Austral Ecology, 2016

Bush encroachment can have profound effects on the ability of savanna ecosystems to provide goods... more Bush encroachment can have profound effects on the ability of savanna ecosystems to provide goods and services to society. It is therefore crucial to understand the key drivers of bush encroachment in savannas. In this study, we test whether decadal changes in mean annual rainfall significantly explain changes in the dominant patch size as well as the density of bush patches at six protected savanna sites located along a rainfall gradient in Zimbabwe. We first performed Maximal Overlap Discrete Wavelet transform within the intensity-dominant scale theoretical framework on multi-temporal aerial photographs and high spatial resolution satellite imagery to objectively detect changes in the dominant patch dimension as well as the intensity of bush cover over a 40-year period at six test sites. We then pooled the data and performed regression analysis relating changes in dominant scale and intensity to decadal changes in mean annual rainfall in order to deduce a possible connection between dynamics of bush encroachment and rainfall variability. Our results indicate a significant nonlinear relationship between changes in the dominant scale and decadal changes in mean annual rainfall (R2 = 0.85, F13 = 35.96, P < 0.01). In contrast, the relationship between decadal changes in mean annual rainfall and changes in intensity was weak and not significant (R2 = 0.29, F13 = 2.69, P = 0.106). These results imply the importance of annual rainfall in explaining long-term changes in the dominant scale of woody patches. However, mechanisms other than rainfall probably explain changes in the intensity of bush cover, and this needs further investigation.

Ecology and Evolution

Transfrontier Conservation Areas, 2017

International audienc

International Journal of Applied Earth Observation and Geoinformation, 2018

International Journal of Geographical Information Science, 2014

Ostrich

Cogent Environmental Science, 2017

African Journal of Range & Forage Science, 2020

Transactions of the Royal Society of South Africa, 2019

Geocarto International, 2018

Geocarto International, 2017

Cogent Food & Agriculture, 2017

African Journal of Ecology, 2016

Applied Geography, 2016

African Journal of Ecology, 2016

Austral Ecology, 2016