A coupled model of leaf photosynthesis, stomatal conductance, and leaf energy balance for chrysanthemum (Dendranthema grandiflora)

•A dynamic greenhouse climate regime is energy efficient.•Incidence of extreme microclimate conditions and plant stress can occur in greenhouse.•A new approach of leaf temperature prediction and monitoring is proposed.•Model based plant stress monitoring and decision-making is put forward.•A multi-model approach with self-selective sub-models is suggested.

While dynamic greenhouse climatic regimes are often applied to achieve energy efficiency, dynamic mechanistic models can assist in climate control decisions, and to elucidate plant stress under extreme microclimatic conditions. The present study developed a couple model with three integrated sub-models to predict net leaf photosynthesis (Pnl), stomatal conductance (gs), and leaf temperature under different microclimatic conditions: (1) a C3 photosynthesis biochemical model; (2) a stomatal conductance model; and (3) a leaf energy balance model. Leaf photochemical efficiency and maximum gross photosynthesis using a negative exponential light response curve were modelled with different leaf temperatures, light levels, and CO2 concentrations. The stomatal conductance and leaf energy balance models were calibrated independently. Pnl, gs, and leaf temperature model predictions were validated with independent measurements and climate input data. Model performance was evaluated by a linear regression of predicted values relative to observed values. The coupled model estimated Pnl with a 2–12% mean difference between the observed and the model, and a 1.82 °C maximum leaf temperature difference between the observed and the model. An additional stomatal model was implemented for comparison, and tested against the model system. Our model showed a better fit to Pnl, leaf temperature, and stomatal conductance validation data. The coupled model was therefore a good predictor for crop growth and microclimate. We suggest a multi-model approach with self-selective sub-models to assist in decisions optimising light, temperature, and CO2 for maximum photosynthetic rates for climatic conditions applied in the model (i.e. high light, temperature, and CO2 concentration). Furthermore, the model leaf temperature prediction could be used for leaf temperature monitoring under unfavorable microclimatic conditions.