Supervisor:
Working thesis:
Bayesian modelling of grape phenology.
Grapevine phenology is heavily influenced by environmental factors such as temperature fluctuations and varying levels of precipitation yearly. Accurate predictions are always vital in managing vineyards effectively and boosting overall grape quality. Traditionally, phenological data (such as data on proportion of open flowers, budburst timing, veraison) have been gathered manually by humans. However technological advancements offer the potential for faster, more accurate and more efficient data collection through sensors, drones, and automated weather stations.
This project applies Bayesian Statistical modelling (specifically, mixed- effects generalized linear models (GLMMs) to:
· Compare the precision and the use of manually collected data versus data collected with modern technology-driven methods.
· Develop predictive phenological methods capable of flexibly incorporating diverse data sources enabling, forecasting under diverse environmental conditions.
· Quantify uncertainty in phenological transitions to support adaptive viticultural practices in under different environmental conditions.
By integrating Bayesian methods, this project will provide novel models reconciling human observation with data obtained from modern technology, flexible decision-making tools prioritizing data collection methods on predictive performances and create a framework for the adaptation in vineyards, balancing precision, and feasibility.
Research interests:
Bayesian statistics, risk management and optimization.
Academic history:
Master of Science in Actuarial and Financial Mathematics ,Rheinland-Pfälzische Technische Universität Kaiserslautern-Landau (RPTU), Germany
Bachelor of Science in Actuarial Science, first class, Kwame Nkrumah University of Science and Technology, Ghana.
