Spatio-temporal dynamics of land use and land cover, wildlife habitats and populations in the greater Serengeti ecosystem, Tanzania.

Abstract

Land use and land cover (LULC) change is a common phenomenon and of great concern to conservation in many terrestrial ecosystems, including the Serengeti Ecosystem (SE) in Tanzania. LULC changes can pose profound impacts on wildlife habitats, abundance and spatio-temporal distribution of wildlife species. This situation needs close monitoring, as it is not clearly known how the future ecological conditions of the ecosystem might be, if these changes remain unchecked. Previous studies on LULC changes, drivers, wildlife habitats and species distributions in the ecosystem are fragmented, focused either on specific habitat types or only on predicting spatial distribution and habitat suitability for particular wildlife species inside the protected areas (PAs). The above-mentioned studies provided limited information on the long-term prediction, imposing difficulties to infer the causes of wildlife populations fluctuation and observed changes in distribution pattern. Knowledge of dynamics of LULC and habitats quality, and the drivers of change is imperative for maintaining healthy wildlife populations and ecosystems integrity. In lieu of this, therefore, the study aimed to carry out a spatio-temporal dyamics of LULC, wildlife habitats and populations in the SE (1975-2015). Specifically, the study sought to: i) characterize LULC change; ii) assess drivers of LULC changes; iii) assess quality of wildlife habitat; and (iv) determine the dynamics of herbivore distribution and habitat selection. For objective one, the random forest classification algorithm was employed to classify the Multispectral Scanner (MSS), Thematic Mapper (TM), Enhanced Thematic Mapper Plus (+ETM) and Operational Land Imager (OLI) was used to characterize LULC into eightiii main classes and extracted quantitative data for assessing the corresponding changes during 1975-1995, 1995-2015 and 1975-2015. For objective two, LULC data for 1995 and 2015 derived from Landsat imageries (objective 1), and nine predictors of change (human population density, precipitation, distance from rivers, soil moisture, fire frequency, distance from roads, elevation, slope and elephant density) were used to ascertain their negative and positive influence for the changes using Binomial Logistic Regression. Drivers of change in LULC, have implications for wildlife habitat quality and spatio-temporal dynamics of wildlife species, therefore, for objective three, we mapped and evaluated changes in habitat quality (1975– 2015) using the Integrated Valuation of Environmental Services and Trade-offs (InVEST) model, whereas, in objective four, Bonferroni confidence interval, with the Chi-square goodness-of-fit test and kernel density were used to assess herbivores habitat selection and distribution for browsers (grant’s gazelles and giraffe), grazers (wildebeest, zebra and buffalo) and mixed feeders (impala and elephant). Results revealed that grassland, shrubland and woodland were the major LULC types throughout 1975-2015 with percentage coverages of 50.6%, 23.7% and 20.9% for 1975; 54.2%, 23.5% and 15.9% for 1995; and 57.0%, 23.8% and 8.9% for 2015. Woodland cover (-11.1%) was the most converted to other cover types during 1975-2015. Overall habitat quality declined over time (1975–2015) in unprotected and human-dominated areas surrounding the ecosystem, intermediate deterioration rates in less heavily PAs (Wildlife Management Areas (WMAs), Game Controlled Area (GCA), Game Reserves (GRs) and the Ngorongoro Conservation Area (NCA) and the least rate in the most heavily protected Serengeti National Park. Significant clustered distribution pattern was observed for alliv herbivores across space and time, with contracted distribution ranges for browsers and an expanded one for grazers and mixed feeders for 2015 in comparison to 1995. The obtained information on species distribution, habitat selection and use are useful in determining high priority areas for effective conservation practices. Generally, increasing human population size, agriculture, settlements and policy changes, fires and elephant browsing pressure are central to LULC and habitats quality dynamics in the ecosystem. The study recommends a more protection effort to halter LULC changes and habitats degradations in order to enhance quality habitat conditions for both browsers and grazers in the ecosystem. For less PAs (e.g. WMAs and GCA) improvement strategies are needed to strengthen conservation and management practices. Effective management of the key drivers of LULC and habitats change in the SE are of paramount importance. Wet and dry season herbivores coverage is needed to examine species guild’s spatio-temporal changes.