A Crop Modelling Strategy to Improve Cacao Quality and Productivity.

Abstract

Cacao production systems in Colombia are of high importance due to their direct impact in the social and economic development of smallholder farmers. Although Colombian cacao has the potential to be in the high value markets for fine flavour, the lack of expert support as well as the use of traditional, and often times sub-optimal technologies makes cacao production negligible. Traditionally, cacao harvest takes place at exactly the same time regardless of the geographic and climatic region where it is grown, the problem with this strategy is that cacao beans are often unripe or over matured and a combination of both will negatively affect the quality of the final cacao product. Since cacao fruit development can be considered as the result of a number of physiological and morphological processes that can be described by mathematical relationships even under uncontrolled environments. Environmental parameters that have more association with pod maturation speed should be taken into account to decide the appropriate time to harvest. In this context, crop models are useful tools to simulate and predict crop development over time and under multiple environmental conditions. Since harvesting at the right time can yield high quality cacao, we parameterised a crop model to predict the best time for harvest cacao fruits in Colombia. The cacao model uses weather variables such as temperature and solar radiation to simulate the growth rate of cocoa fruits from flowering to maturity. The model uses thermal time as an indicator of optimal maturity. This model can be used as a practical tool that supports cacao farmers in the production of high quality cacao which is usually paid at a higher price. When comparing simulated and observed data, our results showed an RRMSE of 7.2% for the yield prediction, while the simulated harvest date varied between +/-2 to 20 days depending on the temperature variations of the year between regions. This crop model contributed to understanding and predicting the phenology of cacao fruits for two key cultivars ICS95 y CCN51.

Plant phenotyping relevance

カカオ果実の成熟・生育（フェノロジー）を気象変数と熱時間から推定する作物モデルを構築・パラメータ化し、観測値と比較検証しており、植物状態の推定手法が研究の中心である。

Code and data availability

The authors' cacao crop model (kocolatl), the modified SIMPLE model code used for the paper's phenology/yield simulations, is stated to be available in a public repository with an authors' GitHub URL. The Fedecacao field phenotype data are not independently deposited; the handle.net URL is a cited reference, not a data