K. Steppe
Greenhouse horticulture is a sector where timing, quantity and quality of production can be controlled to a large extent. The key to this challenging task is that we simply need to listen to what plants have to tell us and look for early-warning signs of problems. In an era of high-tech plant-based measurement technologies, with sap flow and stem diameter variations at the forefront, we are more than ever ready to decipher the plant’s story and get the dialogue with the plant started. But unambiguous interpretation of the huge amount of data continuously spit out by these various plant sensors and the interrelated plant processes they represent is challenging. Mechanistic plant modelling can assist us with exploring the full potential of plant-based measurements. Is it really realistic to build a magic box, consisting of one or more plant sensors and a mechanistic plant model, and get it out there in practice to assist growers in interpreting their plants’ behaviour? At first sight not an easy task, but absolutely feasible if done in a proper way. The pivotal step here is to build reliable mechanistic models that balance model complexity with data availability using state-of-the-art techniques, such as identifiability analysis. Magic will definitely come out of the box and these new technologies employing plant-based measurements and mechanistic modelling will help developing the sustainable and competitive greenhouse horticulture of the future.
Steppe, K. (2012). PLANT SENSORS AND MODELS: GETTING THE DIALOGUE STARTED. Acta Hortic. 957, 277-280
DOI: 10.17660/ActaHortic.2012.957.31
plant sensors, mechanistic plant models, PhytoSim, model calibration, model simulation, identifiability analysis, speaking plant, plant performance

Acta Horticulturae