These tools will no longer be maintained as of December 31, 2024. Archived website can be found here. PubMed4Hh GitHub repository can be found here. Contact NLM Customer Service if you have questions.


Pubmed for Handhelds

PUBMED FOR HANDHELDS

Search MEDLINE/PubMed

  • Title: [Parameter estimation and verification of DSSAT-CROPGRO-Tomato model under different irrigation levels in greenhouse.].
    Author: Zhao ZL, Li B, Feng X, Yao MZ, Xie Y, Xing JW, Li CX.
    Journal: Ying Yong Sheng Tai Xue Bao; 2018 Jun; 29(6):2017-2027. PubMed ID: 29974713.
    Abstract:
    Based on the greenhouse experiment in Shenyang, the growth, development, and yield formation of tomato under different irrigation levels were simulated by growth model DSSAT-CROPGRO-Tomato. The optimal scheme of parameter estimation and model validation was determined. There were four treatments in this experiment. Irrigation upper limit of whole growth season was set as field capacity, while the lower limit was 50% (W1), 60% (W2), 70% (W3), and 80% of field capacity (CK), respectively. The relevant genetic coefficients were estimated by DSSAT-GLUE, a program package for parameter estimation in DSSAT. The differences between simulated and observed values of phenological phase, canopy height, shoot dry matter, tomato fresh mass, leaf area index (LAI), and soil moisture were analyzed to determine the accuracy of simulation. The results showed that the estimated value of genetic parameter of tomato (thermal time for final pod load appeared greater variability under optimal genetic coefficient of tomato, PODUR) had large variability, with the coefficient of variation being 11.5%. When the CROPGRO-Tomato model was applied to the greenhouse in different regions, the PODUR should be estimated adequately. Otherwise, the accuracy of simulation would be affected. In the process of model application, the observation data of sufficient irrigation treatment should be selected for estimating genetic parameters, which could improve the simulation precision. The absolute relative error and standard root mean square error were 8.7% and 10.5%, respectively. The simulation results of LAI and soil moisture showed that the higher the irrigation level was, the higher accuracy of simulation was. By leave-one-out cross validation, the overall error validation ranged from 10.5% to 12.5%. Our results indicated that the growth, development, and yield formation of tomato could be accurately simulated by DSSAT CROPGRO-Tomato model under different irrigation conditions in Shenyang greenhouse. 在沈阳地区日光温室试验的基础上,利用番茄生长模型DSSAT-CROPGRO-Tomato模拟了不同灌水水平条件下温室番茄的生长发育和产量形成过程,并确定了参数估计和模型验证的最优方案.试验设4个处理,全育期的灌水上限均为计划湿润层田间持水率,灌水下限分别为计划湿润层田间持水率的50%(W1)、60%(W2)、70%(W3)和80%(CK).利用DSSAT-GLUE参数估计模块得到遗传参数的不同估计结果,通过对比分析番茄物候期、冠层高度、地上干物质量、鲜果产量、叶面积指数(LAI)、土壤含水率的模拟值与实测值之间的差异,来确定该模型模拟精度.结果表明: 番茄遗传参数——最优条件下最终果实负载所需光热时间(PODUR)的估计值具有较大变异性,变异系数为11.5%,将CROPGRO-Tomato模型应用于不同地区日光温室时,应对此参数进行充分估计,否则会影响其模拟精度.在模型应用过程中,应选用充分灌水处理的观测数据进行遗传参数估计,可以提高模型的模拟精度.此时的绝对相对误差和标准均方根误差值分别为8.7%和10.5%.对作物LAI和土壤含水率动态模拟结果可以看出,灌水水平越高,模型模拟精度越高.留一交叉验证法的总体模拟误差在10.5%～12.5%.说明DSSAT-CROPGRO-Tomato模型可以较为准确地模拟沈阳日光温室不同灌水水平条件下番茄生长发育和产量形成过程.
    [Abstract] [Full Text] [Related] [New Search]