Articles
Comparative analysis of plant stress-reducing ventilation strategies using plant sensors and decision trees in different tomato greenhouses
Article number
1433_26
Pages
203 – 210
Language
English
Abstract
In spring and summer, tomato plants cultivated in greenhouses can experience environmental stress due to increased levels of irradiation and vapour pressure deficit during the day.
On such days, increased transpiration rates deplete the plant’s internal water storage pools (i.e., living cells), leading to elevated daily stress with the risk of irreversible damage to the plant or its fruits.
To prevent such risky events, Belgian and Dutch greenhouse farmers are applying a specific ventilation strategy, which we dubbed the “plant stress-reducing ventilation” strategy.
This scientifically understudied strategy, currently lacking established guidelines, is approached differently by individual greenhouse growers.
A comparative analysis was conducted involving two different Belgian tomato growers.
For both growers, tomato plants (Solanum lycopersicum L.) were equipped with sap flow and stem diameter variation sensors to continuously measure the plant response during the application of the plant stress-reducing ventilation strategy.
The plant response was classified and combined with data from the greenhouse climate computer to generate a decision tree using machine learning.
The strategic integration of a decision tree algorithm with plant sensors proved essential in unraveling the plant stress-reducing ventilation strategy in practical settings.
Additionally, it facilitated comparative analysis of the different approaches employed by the greenhouse growers.
On such days, increased transpiration rates deplete the plant’s internal water storage pools (i.e., living cells), leading to elevated daily stress with the risk of irreversible damage to the plant or its fruits.
To prevent such risky events, Belgian and Dutch greenhouse farmers are applying a specific ventilation strategy, which we dubbed the “plant stress-reducing ventilation” strategy.
This scientifically understudied strategy, currently lacking established guidelines, is approached differently by individual greenhouse growers.
A comparative analysis was conducted involving two different Belgian tomato growers.
For both growers, tomato plants (Solanum lycopersicum L.) were equipped with sap flow and stem diameter variation sensors to continuously measure the plant response during the application of the plant stress-reducing ventilation strategy.
The plant response was classified and combined with data from the greenhouse climate computer to generate a decision tree using machine learning.
The strategic integration of a decision tree algorithm with plant sensors proved essential in unraveling the plant stress-reducing ventilation strategy in practical settings.
Additionally, it facilitated comparative analysis of the different approaches employed by the greenhouse growers.
Authors
R. Vanbeylen, F. De Ridder, H. Marien, K. Steppe
Keywords
greenhouse climate control, machine learning, plant sensors, sap flow, stem diameter variation, solar radiation, vapour pressure deficit
Online Articles (41)