温室盆栽一品红生长发育模拟模型

该研究的目的是建立一个温室盆栽一品红（Euphorbia pulcherrima Willd.）生长发育模拟模型，为温室盆栽一品红生产中的光温精准调控提供理论依据与决策支持。以一品红品种‘中国红’（Euphorbia pulcherrima Willd. Red China）为研究对象，通过不同定植期和不同密度的试验，定量分析了一品红生长发育与光温的关系。在此基础上，以生理辐热积（Physiological product of thermal effectiveness and PAR，PTEP）为尺度，建立了温室盆栽一品红生育期模拟子模型；以冠层吸收的生理辐热积（Canopy intercepted PTEP, PTEPint）为尺度，建立了温室盆栽一品红干物质生产和分配模拟子模型；综合生育期模拟子模型与干物质生产和分配模拟子模型，建立了温室盆栽一品红生长发育模拟模型，并用独立的试验数据对模型进行了检验。模型对从摘心到短日处理、单苞、单蕾、多蕾和开花期的模拟预测值与实测值的符合度较好。模拟值与实测值基于1︰1线的决定系数R2为0.99，回归估计标准误差RMSE分别为0.7、3、3.5、0.7和2 d，预测精度明显高于以有效积温为尺度的发育模型（RMSE分别为8、4.5、3.8、2.8和7.2 d）。模型对单位面积总干质量、叶干质量、茎干质量和苞叶干质量的模拟值与实测值基于1︰1线的R2和RMSE分别为0.98、0.97、0.91和0.95；7.12、7.49、3.89和2.48 g/m2。模型对一品红单位面积总干质量的预测精度明显高于基于光合作用驱动的生长模型（R2和RMSE分别为0.77和35.06 g/m2）。该研究建立的模型能够较准确地预测温室盆栽一品红各生育期出现时间、干物质生产和各个器官干质量的动态，模型的预测精度较高、参数少且易获取、实用性较强。

Model for simulating development and growth of pot planted poinsettia (Euphorbia pulcherrima) grown in greenhouse

In order to optimize the climate management for pot planted poinsettia (Euphorbia pulcherrima Willd.) grown in a greenhouse, a model of pot planted poinsettia growth and development simulation was developed. Experiments with different planting dates and densities were conducted in a multi-span Venlo type greenhouse of Nanjing. The photo-thermal effects on the development and growth of poinsettia were quantitatively analyzed using the experimental data. Based on the quantitative analysis, a sub-model for predicting the development stages of poinsettia was developed using an integrated photo-thermal index physiological product of thermal effectiveness and PAR, PTEP. A sub-model for predicting the dry matter production and dry matter partitioning of poinsettia was developed using the canopy intercepted PTEP, PTEPint. A growth and development simulation model for pot planted poinsettia was developed by integrating the two sub-models mentioned above. Independent experimental data were used to validate the model. The results showed that the model could give predictions of the development stages, dry matter production, and organ dry weight of poinsettia crops satisfactorily. Based on the 1:1 line, the determination coefficient (R2) between the predicted and observed development stages is 0.99; and the root mean squared errors (RMSE) between the predicted and observed days from pinching to start of short day treatment, first visible cyathia, first visible bud, third visible bud and flowering date are, respectively, 0.7, 3, 3.5, 0.7 and 2 days. Based on the 1:1 line, the R2 and RMSE are 0.98 and 7.12 g/m2, for total biomass production, 0.97 and 7.49 g·m-2, respectively, for leaf dry weight, 0.91 and 3.89 g/m2, respectively, for stem dry weight, 0.95 and 2.48 g·m-2, respectively for cyathia dry weight. The model developed in this study gives more accurate predictions for development stages than the growing degree days (GDD) based model (with RMSE of 8, 4.5, 3.8, 2.8 and 7.2 days, respectively) does, and gives significantly higher prediction accuracies for total biomass production than the photosynthesis driven crop growth model (with R2 and RMSE of 0.77 and 35.06 g/m2, respectively) does. Based on the results obtained in this study, it can be concluded that the model developed in this study can give satisfactory predictions of pot planted poinsettia development and growth, hence, can be used for decision making for precision control of light and temperature in greenhouse poinsettia production.