节点文献
基于气象因子的赣南脐橙产量模拟模型构建
Construction of Gannan Navel Orange Yield Simulation Model Based on Meteorological Factors
【摘要】 利用赣州市2000-2022年脐橙产量数据和气象数据,分离脐橙气象产量,确定影响脐橙产量形成的关键气象因子,利用多元线性回归法构建基于关键气象因子的脐橙相对气象产量模拟模型,并对模型进行验证。结果表明:(1)给定0.8权重的指数平滑法对2000-2022年赣州市脐橙气象产量分离较合理;(2)脐橙越冬期降水量、现蕾开花期平均气温、幼果生长期平均气温、果实膨大期降水量和着色成熟期日照时数是影响脐橙产量的关键气象因子;(3)构建了2000-2017年越冬始期12月1日-翌年9月30日以及12月1日-翌年11月30日的脐橙相对气象产量模拟模型,模型均通过0.01水平显著性检验,相关误差(RE)分别为5.52%和5.31%,均方根误差(RMSE)分别为604.85kg·hm-2和614.86kg·hm-2,2018-2022年模型验证的准确率分别为97.88%和97.84%。综合来看不同数据源所构建的模型均适用于赣南脐橙产量模拟与评估。
【Abstract】 In this study, the navel orange yield and the meteorological data from Ganzhou were collected for the period from 2000 to 2022, and the corresponding meteorological yield was separated. The key meteorological factors affecting meteorological yield were identified by fitting relationships between meteorological yield and meteorological factors at five growth stages. Finally, the relative meteorological yield model based on key meteorological factors was constructed using multiple linear regression method, and the model was validated to determine its reliability, stability and accuracy. The results indicated that:(1) the exponential smoothing method with a given weight of 0.8 was more reasonable for separating the meteorological production of navel oranges in Ganzhou from 2000 to 2022.(2) The key meteorological factors affecting the yield of navel oranges included precipitation during the overwintering period, average temperature during the budding and flowering period, average temperature during the young fruit growth period, precipitation during the fruit swelling period, and sunshine hours during the coloring and ripening period.(3) Two yield simulation models were constructed based on the key meteorology factors from December 1 to September 30 of the following year(i.e., overwintering to the end of fruit swelling) and December 1 to November 30 of the following year(i.e., overwintering to coloring and ripening), respectively. Both the model passed the 0.01 level of significance, with the relative error of 5.52% and 5.31%, and the root mean squared error of 604.85kg·ha-1 and 614.86kg·ha-1, respectively. The model validation accuracy from 2018 to 2022was 97.88% and 97.84%, respectively. Overall, the two simulation models are suitable for simulating navel oranges yield in southern Jiangxi.
【Key words】 Gannan navel orange; Meteorological factors; Yield forecast; Model constructed;
- 【文献出处】 中国农业气象 ,Chinese Journal of Agrometeorology , 编辑部邮箱 ,2025年06期
- 【分类号】S666.4;S162.55
- 【下载频次】65