Skip navigation

Using Bayesian networks to model flowering in coffee plantations in Central America

Using Bayesian networks to model flowering in coffee plantations in Central America

Lara-Estrada, Leonel ORCID: 0000-0002-6562-9497 and Sucar, Luis Enrique (2022) Using Bayesian networks to model flowering in coffee plantations in Central America. In: 16th Bayesian Modelling Applications Workshop (at the 38th Conference on Uncertainty in Artificial Intelligence - UAI), 5th Aug 2022, Eindhoven, The Netherlands. (In Press)

Full text not available from this repository. (Request a copy)

Abstract

The coffee plant is climate-sensitive; extreme rainfall during a flowering day causes reductions in coffee yields. The coffee flowering’s intensity and date of occurrence are influenced by the water stress (dry season) and rainfalls during the dry and beginning of the wet season. Multiple flowering events could occur in a year, and there is a finite number of possible flowers per plant. This contribution introduces a Bayesian network model to infer multiple coffee flowerings for the Pacific Region in Nicaragua. The model structure and parameters were built based on previous related studies and data from coffee farms in the region, which included 55 years of flowerings (intensity and date of occurrence) and daily rainfall data from a coffee farm. Flowering data from 4 farms and rainfall data from two locations were used to validate the model using the metric spherical payoff: results were above satisfactory to infer flower intensity. Also, the model was able to depict expected phenological behaviors for single or multiple flowerings. We believe this model has the potential to evolve and support the development of an agricultural insurance to deal with yield losses because of extreme rainfall during flowering.

Item Type: Conference or Conference Paper (Plenary)
Uncontrolled Keywords: probabilistic modelling; crop modelling; Nicaragua; uncertainty; expert knowledge; machine learning
Subjects: Q Science > Q Science (General)
Q Science > QA Mathematics
S Agriculture > S Agriculture (General)
Faculty / School / Research Centre / Research Group: Faculty of Engineering & Science
Faculty of Engineering & Science > Natural Resources Institute
Faculty of Engineering & Science > Natural Resources Institute > Agriculture, Health & Environment Department
Related URLs:
Last Modified: 09 Feb 2024 16:31
URI: http://gala.gre.ac.uk/id/eprint/33723

Actions (login required)

View Item View Item