水肥耦合驱动下的番茄植株形态模拟模型
Simulation of tomato morphology growth with water-fertilizer coupling

【目的】建立土壤水肥影响下的番茄株高、单株展开叶数及单叶面积随辐热积的变化模型，为番茄施肥和灌溉提供依据。【方法】以灌水上限和施肥量为因素，采用二元二次正交旋转组合设计，建立基于辐热积的番茄株高及单叶面积的logistic模型和单株展开叶数的指数函数模型，并根据建立的单株展开叶数模型和单叶面积模型，模拟番茄叶面积指数。【结果】所建模型对不同水肥处理下的番茄株高、单株展开叶数、单叶面积和叶面积指数的预测效果较好，其预测值与实测值之间基于1∶1直线的决定系数R2分别为0.962 7，0.947 1，0.854 8和 0.926 3，相对误差（RE）分别为9.41％，7.50％，17.80％和18.20%。【结论】所建模型对番茄株高、单株展开叶数、单叶面积、叶面积指数4个指标的预测精度均达80%以上，能较好地预测灌水上限和施肥量对番茄形态的动态影响。
【Objective】This study established models for tomato stem length,number of unfolding leaves and single leaf area on product of thermal effectiveness and PAR (TEP) with water-fertilizer coupling to improve fertilization and irrigation of tomato roots.【Method】Taking the upper limit of irrigation and fertilizer amount as factors,two quadratic general rotational combination design was used to build logistic regression model to assess the relationship between stem length,leaf area and TEP and exponential model between number of unfolding leaves and TEP.A leaf area index model was also established based on number of unfolding leaves and single leaf area.【Result】Prediction accuracies (mean relative error) of tomato stem length model,number of unfolding leaves model,single leaf area model and leaf area index model were 9.41%,7.50%,17.80% and 18.20%,respectively.Their determination coefficients (R2) between simulated and measured values were 0.962 7,0.947 1,0.854 8 and 0.926 3,respectively.【Conclusion】The prediction accuracies of the established models on the 4 indexes were all larger than 80%,indicating that the model can well predict the variations of tomato morphology characteristics