Assessing the effect of land use transition, climate and vegetation anomalies on fire and encroaching species distribution in mountainous grassland

Dr Kayode Adepoju1, Dr Samuel Adelabu1

1University of The Free State, Phuthadijthaba, South Africa


Fire disturbance and recovery exert a strong influence on distributions and future spread pattern of encroaching species. Encroaching species and fire disturbance are likely to increasing in distribution and intensity due to climate change. However, it is uncertain whether the relationship between fire and encroaching species will converge or diverge under varying degrees of combined socio-ecological disturbance situation.  In this study, we applied random forest machine learning and Maxent species distribution modelling algorithms to estimate the distribution of fire and Serethium Plimosum  – an encroaching species, based on remote sensing and meteorological variables in mountainous grassland of South Africa.  To further understand the effect  and uncertainty of socio-ecological systems we (1) used landuse transition matrix, trends and anomaly in climate and vegetation datasets to determine preceding socio-ecological conditions leading to current fire and species distribution patterns (2) explored the effect of these variables on future distribution of fire and encroaching species in mountainous grassland of South Africa. We propose that the result of this study may contribute to a better understanding of the relationship between socio-ecological drivers of drought and fire across multiple years under global change.


Dr. Adepoju is a Postdoctoral Research Fellow in the Department of Geography,  University of the Free State, South Africa. He has a Ph.D. degree in Ecology and Environmental Science and he has worked as a geospatial data scientist with regional government and academia, specializing in the design and implementation of early warning decision support systems for food security and disaster management in the Sahel region. His main research interest is on the integration of stochastic conceptual models using machine learning approaches with system dynamical models for ecological risk assessments with a view to providing complementary insights for environmental conservation

About provides delegate registration, website and app solutions, and financial management for conferences, conventions and scientific meetings.

Terms & Conditions

All registrations and bookings are subject to our standard term and conditions.

Contact Us

Please contact the team at with any questions regarding the conference.
© 2017 - 2020 Conference Design Pty Ltd. is a division of Conference Design Pty Ltd.