Comparison of multiple plant sensors aimed at early detection of drought stress in the greenhouse

With a growing world population and mounting pressure on natural resources, the need for efficient, sustainable food production is becoming increasingly urgent. A promising route towards improving agricultural productivity is to expand the use of sensors to monitor plants and their environment with high accuracy and temporal resolution. Data generated by such sensors is useful for optimizing irrigation, nutrition and illumination in the context of autonomous greenhouses, while allowing mitigation of plant stress due to pests, diseases and extreme climate conditions. We simultaneously tested ten different types of sensors for monitoring early signs of drought stress in mature, high-wire tomato plants grown in rockwool. Sensors ranged from high-density climate sensors to novel sensors for monitoring plant-specific parameters like acoustic emissions, stomatal dynamics, sap flow and stem diameter. Withholding water for two days led to a quick and complete depletion of water in the rockwool slabs, and strongly affected whole-plant transpiration, resulting in strong changes in: acoustic emissions, stomatal pore area, stomatal conductance, and stem diameter, all of which were found to be significant indicators of early drought stress. This work demonstrates the usefulness of these sensors in a greenhouse environment and provides a comparison between measured parameters in magnitude and onset time, aimed at the early detection of drought stress. Our study aims to facilitate sensor selection and implementation in precision agriculture.