日光温室黄瓜生长发育模拟模型

实现日光温室黄瓜生长发育动态模拟预测,可为日光温室黄瓜智慧生产管理提供技术支撑.本研究依据黄瓜生长发育的光温反应特性,以‘津优35’为试验品种,利用2年4茬分期播种试验观测数据建立基于钟模型的温室黄瓜发育模拟模型.依据温室黄瓜叶片生长与关键气象因子(温度和辐射)的关系,以辐热积(TEP)为自变量构建了黄瓜叶面积指数(LAI)模拟模型;依据单位叶面积光合作用对叶面积指数和日长的二重积分,结合黄瓜不同器官的呼吸消耗,构建了黄瓜干重生产分配模拟模型,结合器官含水量,构建了黄瓜器官鲜重模拟模块.基于各子模块构建了温室黄瓜生长发育模拟模型,确定了模型品种参数并进行检验.结果表明: 日光温室黄瓜移栽期-伸蔓期、移栽期-初花期、移栽期-采收初期和移栽期-拉秧期的模拟值与观测值的均方根误差(RMSE)在3.9～10.5 d,归一化均方根标准误差(nRMSE)在6.5%～28.6%,符合度指数(D)在0.79～0.97.LAI与TEP呈S型曲线变化关系,LAI模拟值与实际观测值的RMSE为0.19,nRMSE为17.2%,D值为0.90.根、茎、叶、花和果干重模拟值与实际观测值的RMSE在0.39～8.94 g·m^-2,nRMSE在10.9%～17.7%,D值均为0.98以上.表明模型能够较准确地模拟黄瓜关键发育期、叶面积和各器官干鲜重,定量化日光温室黄瓜生长发育过程.

Simulation model for cucumber growth and development in sunlight greenhouse

The dynamic simulation of cucumber growth and development in sunlight greenhouse can provide technical support for the intelligent management of cucumber production. According to the cucumber response characteristics to light and temperature, the cucumber development module based on the algorithm of clock model was established by using data from four-stage experiment with ‘Jinyou 35’ as experiment variety in two years. Based on the relationship between the leaf growth and key meteorological factors (temperature and radiation), leaf area index (LAI) module was established with the accumulated product of thermal effectiveness and photosynthetically active radiation (TEP) as independent variables. The simulation module of cucumber dry matter production was established by taking into consideration the double integral of LAI and daily length in photosynthesis per unit leaf area as well as the respiratory expenditure of different organs. Combined with water content of organs, fresh weight simulation module of cucumber organs was constructed. The whole cucumber development and growth simulation model in greenhouse was built based on each sub-module. The model parameters were calibrated and determined. The results showed that root mean square error (RMSE) of simulated values and observed values of four deve-lopment stages (from transplanting date to stretch tendril, to initial flowering, to early harvested and to uprooting), was 3.9-10.5 d. The normalized root mean square error (nRMSE) was 6.5%-28.6%. The coincidence index (D) was 0.79-0.97. The relationship between LAI and TEP was the regression of ‘S’ type curve. The RMSE of simulated and observed LAI values was 0.19. The nRMSE was 17.2%. The D value was 0.90. The RMSE of dry weight of root, stem, leaf, flower and fruit of the simulated values and observed values were 0.39-8.94 g·m^-2. The nRMSE were 10.9%-17.7%. The D values were all above 0.98. The growth and development model of cucumber could accurately simulate the key development period of cucumber, leaf area and the dry and fresh weight of various organs and quantify the growth and development of cucumber in sunlight greenhouse.