ADVISOR: Dr. Henrique Barbosa

TITLE: Vegetation-Climate Dynamics in Africa Through the Lens of Machine Learning

ABSTRACT: Vegetation phenology plays a crucial role in regulating global carbon, water, and nitrogen cycles, thereby creating significant impact on terrestrial energy budgets and the exchange of atmospheric CO2. Through these cycles, vegetation phenological changes alter the climate system by modifying the exchange of energy and water between land surfaces and the atmosphere. Hence, adequate knowledge of how vegetation is changing, tracking the factors responsible for the changes, and their impact on the ecosystem is vital in gaining a better understanding of the interconnectivity between vegetation and climate system, which, in turn, will allow for improving the representation of vegetation in climate models, and reduce the uncertainty in climate change forecasts. Despite contributing significantly to the global carbon budget and being identified as a potentially highly vulnerable region to climate change impacts, Africa, which boasts 17% of the world's forest cover, remains relatively understudied when it comes to vegetation-climate interactions. This proposal aims to investigate such interactions using supervised and unsupervised machine learning (ML) techniques to analyze the time series of vegetation and climate variables derived from satellite observations for patterns, trends, and differences in vegetation phenology across different regions of Africa. I will investigate the complex relationships between climate and vegetation phenology in Africa, considering abrupt and continuous changes, as well as causal relationships. I expect the study to further our understanding of Earth's climate system, and will have practical applications for policymakers, as they will provide insights into how climate change affects vegetation phenology, which can help guide policy decisions related to agriculture, water management, and land use planning.