高红霞　徐蔚军

摘要：在以日为尺度的兴安盟玉米（Zea mays L.）气候适宜度模型的基础上，利用突泉县农业观测站1982-2010年地面气象资料及同期的玉米生育期观测资料，建立了兴安盟玉米生育期预报生理指标及预报模型，并利用兴安盟3个玉米主产旗县2011-2015年的玉米生育期观测数据对模型的预报结果进行验证。结果表明，各生育期持续时间的预报值与实际观测值一致程度较高，各生育期预报值与观测值之间的平均绝对误差在1.7～4.5 d，均方根误差在1.2～3.7 d；营养生长阶段绝对误差在3 d以内出现的频次接近62%，生殖生长阶段绝对误差在3 d以内出现的频次约為64%，模型能较好地预测相关生育期（出苗、三叶、拔节、抽雄、乳熟和成熟期），预报精度基本可以满足需求。

关键词：玉米（Zea mays L.）；生育期；气候适宜度；预报模型；兴安盟

中图分类号：S513 文献标识码：A 文章编号：0439-8114（2017）19-3615-04

DOI：10.14088/j.cnki.issn0439-8114.2017.19.005

Abstract： Based on the climate suitability model of Maize in Xingan League， physiological indexes and prediction model of maize was established using dailymeteorological data and growth stage observation data of Tuquan County Agricultural Observation Station from 1982 to 2010，and the models prediction results were verified using the maize growth period observation data of three major maize grain-producing counties in Xingan League from 2011 to 2015. The results showed that there was good agreement between predictions and observations for the duration of each growth stage. The average absolute error between the predicted value and the observed value was 1.7 to 4.5 days，root mean square error was 1.2 to 3.7 days. The frequencies of the absolute error within 3 days was near to 62% during the vegetative growth stage，and the frequencies of the absolute error less than 3 days was about 64% during reproductive growth stage. The model could predict the development period well， and the forecast accuracy could meet the requirements.

Key words： maize（Zea mays L.）； growth stage； climate suitability； prediction model； Xingan League

进入21世纪，农业生产在实现高产、优质、高效的同时，还要兼顾生态安全并迎接气候变化等多重挑战。作物发育期体现了作物生育过程中外部形态变化的特征，是衡量生态系统对气候变化的相应指标之一[1]。对于大田生产，农业气象条件对作物的生长发育及产量起决定性作用[2]。研究作物发育规律与气象条件的关系，准确及时地做出生育期预报，对合理进行作物田间管理，减少农业气象灾害影响、增产增收具有重要意义。

20世纪60年代开始，国内外许多专家相继研究过多种作物生育期预测模拟模型[1-8]，这些模型中常用的发育指标主要有积温和生理发育时间。……