Assessing climate risks in cocoa supply chain

: yield forecasting and implications for the cocoa and chocolate industry

Abstract

Cocoa is a key commodity for both producing and consuming countries, with over 70% of the world’s supply coming from West and Central Africa. Like any other agricultural crop, cocoa yields are highly sensitive to climate change, which can cause vulnerabilities in production. This study analyzes the impact of climate change on cocoa productivity in Ivory Coast, Ghana, Cameroon, and Nigeria, for Barry Callebaut9s cocoa supply chain. By using historical yield data with high-resolution ERA5 climate reanalysis, multiple machine learning models were tested and evaluated for their predictive performance. The best performing model, a Semiparametric Neural Network, was used to forecast cocoa yield up to 2035 using CMIP6 climate projections under three future climate scenarios (SSP1-2.6, SSP2-4.5, and SSP5-8.5). The results showed significant differences in yield trends across regions and how sensitive they are to emissions scenarios. While Cameroon and Nigeria presented stable and increasing yields, Ivory Coast and Ghana faced more interannual variability, especially under the high-emission scenario. Scenario-based simulations of changes in cultivation area showed that production is highly sensitive to land availability, emphasizing the strategic importance of land management. The findings highlight the importance of Barry Callebaut integrating climate projections into supply chain planning. Targeted investments and diversification of their sourcing strategy are recommended to mitigate future climate risks and ensure supply resilience.

Date of Award	25 Jun 2025
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Miguel de Oliveiros Vieira de Albergaria e Castro Nogueira (Supervisor)