基于光温效应的温室小白菜生理指标动态模拟

Dynamic simulation of physiological index of Brassica chinensis L.in greenhouse based on light and temperature function

【目的】研究温室环境下温度和光照与小白菜生理指标的动态关系,以期为小白菜生产的精准管理提供参考。【方法】以华冠小白菜品种为试材开展营养液膜技术(Nutrient film technique,NFT)栽培,于2019年6—9月分3个时段进行试验,试验期间自动采集温室内温度和光照等数据,每2 d进行1次生理指标测定。以建模试验数据计算光温效应(LTF)、辐热积(TEP)和积温(GDD),将其与同时期测定的生理指标数据进行拟合,建立温室小白菜生理指标动态模拟模型,将模拟值与实测值进行比较,检验模型拟合效果。【结果】试验期间的日平均气温为33.26~34.51℃,日光合有效辐射为8.18~13.64 mol/(m^2·d)。小白菜叶片主要生理指标基本上随生长期间LTF的增加而增加,但可溶性蛋白含量在后期呈现略微下降趋势,硝酸盐含量则呈现升高、降低、再升高的变化。LTF模型的预测效果优于TEP和GDD模型;该模型对小白菜可溶性糖、可溶性蛋白、维生素C、纤维素、蔗糖、淀粉、叶绿素a、叶绿素b、总叶绿素、类胡萝卜素和硝酸盐含量及根系活力的预测结果回归估计标准误差(RMSE)比后两者低,其各指标的RMSE为TEP和GDD模型的9.88%~49.91%和5.86%~93.42%,表明模型的预测精度较高。LTF模型预测结果与实测值之间R2均大于0.950,表明模拟值与实测值的匹配程度较好。【建议】光温效应法适用于对小白菜主要生理指标的模拟,模型对特定小白菜品种、特定影响因子、特定时间段、特定温室环境下的小白菜模拟效果较好,后期需要加强对多品种、多因素和多时空适用模型的系统性研究。

【Objective】Studying the relationship between ambient temperature and light and physiological indexes of Brassica chinensis L.in greenhouse could provide reference for accurate management of facility cultivating B.chinensis.【Method】Nutrient film technique(NFT)cultivation was carried out with Huaguan B.chinensis as materials,data of temperature and light in greenhouse was automatically collected for three periods during June and September in 2019,physiological indexes were measured once every 2 d.The values of light and temperature function(LTF),thermal effectiveness and photosynthetically active radiation(TEP)and growing degree days(GDD)were calculated by using modeling experiment data,which were fitted with the physiological indexes measured at the same period,and then dynamic simulation models of physiological indexes of B.chinensis in greenhouse were established.The simulated values were compared with the measured values at the same period to verify the fitting effect of the models.【Result】The results showed that during the experiment,the daily average temperature was 33.26-34.51℃,and the daily photosynthetically active radiationwas8.18-13.64 mol/(m^2·d).The main physiological indexes of B.chinensis mainly increased with the increase of LTF during growing time;the soluble protein content decreased slightly in the later stage,while the nitrate experienced the change of increasing,decreasing and increasing.LTF model was superior to TEP model and GDD model.The standard error of regression estimation(RMSE)of predicted results of LTF model for soluble sugar,soluble protein,vitamin C,cellulose,sucrose,starch,chlorophyll a,chlorophyll b,total chlorophyll,carotenoids,nitrate contents and root activity were lower than the latter two.The RMSE of TEP and GDD models were 9.88%-49.91% and 5.86%-93.42%,indicating that the prediction accuracy was high.The R2 between the prediction results of the LTF model and the measured values was greater than 0.950,indicating that the simulated values and the measured values matched well.【Suggestion】The LTF method is applicable to simulate physiological index of B.chinensis.The simulation effects of the model on specific varieties,specific impact factors,specific time period and specific greenhouse environment are fine.It is necessary to strengthen the systematic research on the models of multi-species,multi-factor and multi-space-time.