Résumé:
The climate is having an increasing impact on greenhouse production. Frequent and unpredictable events lead to physiological adaptations of plants to the detriment of fruit quality. Tomatoes, for example, split or even burst in the ripening phase, often in the summer when a succession of hot periods occur and the fruit is not elastic enough to absorb the physical changes due to frequent irrigation. This causes important losses for the growers.
This project aims to implement and test a real-time connected fruit dendrometer in a soilless tomato culture, combined with a micro-climate analysis and machine learning algorithms. The objective is to detect a typical signature growing curve for tomato crop production and predict cracking event. The research has three main outcomes: (i) improve crop quality; (ii) optimize harvest timing; (iii) reduce water usage.
A field trial took place in 2022 at the research centre of Agroscope, Switzerland. The mechanical and electronic behaviour of 60 fruit dendrometers was tested in a soilless tomato greenhouse, together with the setup of a data transmission and storage system. Six micro-climatic stations allowed to study the dependence between climatic data and fruit growth. Phenological and physiological monitoring of tomato plants and fruit quality analysis allows the characterization of the different climatic environments that could affect the fruit cracking occurrence
The preliminary results revealed promising. The build fruit-growth models can predict the final fruit diameter and the best harvest time. The plants exhibit different transpiration levels and a certain texture variability in response to the different micro-environmental conditions.
Monitoring continues for generating additional data on the fruit growth behaviour. This contributes to develop precise fruit cracking model, alerting the greenhouse producers about a potential risk that may lead to losses in yield quality.
SWITCH edu-ID
Administration
