Real-time application of crop transpiration and photosynthesis models in greenhouse process control

Mathematical models that incorporate knowledge of crop function and other greenhouse processes can inform the grower about the growing process. It can be assumed that models predict the behavior of a healthy, well-functioning plant. Hence, the difference between model-predicted values and accurate measurements can serve as a criterion of the status of the crop. A mechanistic transpiration model that predicts actual crop transpiration from the measured climate in the greenhouse was implemented in the process control computers of commercial tomato greenhouses to demonstrate the possibilities of this idea. In the same process, actual transpiration was available as measured with a ProDrain weighing gutter. Occasionally, measured transpiration was lower than predicted, which may indicate suboptimal crop function. When the difference exceeded a threshold value for a certain period, the process computer generated an alarm in order to warn the grower. Lower measured than predicted transpiration is probably caused by partial closing of stomata. To have an indication of the impact of stomatal behavior on crop production, a photosynthesis model was also implemented, and photosynthesis rate was calculated for theoretical as well as actual stomatal resistance. This study demonstrates that simulation models, when implemented in a greenhouse process computer and combined with the appropriate measurements, can automatically alert the grower of potentially damaging conditions, e.g., reduced crop performance or a system malfunction in the greenhouse, can provide the grower insight into factors affecting crop function and, finally, can help the grower reduce mistakes, optimize crop growth, maximize yield and save energy