收藏设为首页 广告服务联系我们在线留言

干旱区地理 ›› 2021, Vol. 44 ›› Issue (2): 494-506.doi: 10.12118/j.issn.1000–6060.2021.02.20

• 生物与土壤 • 上一篇    下一篇

陇中黄土高原区旱地春小麦产量对干旱胁迫响应的模拟研究

王钧1(),李广1(),聂志刚1,董莉霞1,闫丽娟2   

  1. 1.甘肃农业大学信息科学技术学院,甘肃 兰州 730070
    2.甘肃农业大学农学院,甘肃 兰州 730070
  • 收稿日期:2020-05-29 修回日期:2020-09-24 出版日期:2021-03-25 发布日期:2021-04-14
  • 通讯作者: 李广
  • 作者简介:王钧(1982-),男,博士研究生,副教授,主要从事农业信息化研究. E-mail:julianwong82@163.com
  • 基金资助:
    国家自然科学基金项目(31660348);甘肃省自然科学基金项目(20JR10RA509);甘肃省重点研发计划项目(20YF8NA135);甘肃农业大学青年导师扶持基金资助项目资助(GAU-QDFC-2020-13)

Simulation study of response of spring wheat yield to drought stress in the Loess Plateau of central Gansu

WANG Jun1(),LI Guang1(),NIE Zhigang1,DONG Lixia1,YAN Lijuan2   

  1. 1. College of Information Science and Technology, Gansu Agricultural University, Lanzhou 730070, Gansu, China
    2. Agronomy College, Gansu Agriculture University, Lanzhou 730070, Gansu, China
  • Received:2020-05-29 Revised:2020-09-24 Online:2021-03-25 Published:2021-04-14
  • Contact: Guang LI

摘要:

为进一步探明陇中黄土高原区旱地春小麦产量形成对不同干旱胁迫的响应机制,依据甘肃省定西市安定区凤翔镇安家沟村2016—2018年大田控水试验数据以及定西市安定区1971—2018年气象数据,验证农业生产系统模拟(Agricultural production systems simulation,APSIM)模型模拟不同干旱胁迫旱地春小麦产量及产量构成要素的适宜性,基于APSIM模型分析不同生育期、不同程度干旱胁迫对旱地春小麦籽粒数、千粒重和产量的影响,利用多元逐步回归方程确定陇中黄土高原区旱地春小麦最佳灌水时间和灌水量。结果表明:(1) APSIM模型模拟陇中黄土高原区旱地春小麦生育期、籽粒数、千粒重和产量的均方根误差(Root mean square error,RMSE)均小于3.67 d、300.52个·m-2、2.56 g、267.43 kg·hm-2,归一化均方根误差(Normalized root mean square error,NRMSE)均小于3.89%、2.86%、9.71%、11.58%,模型有效性指数(Model effectiveness index,ME)均大于0.62、0.78、0.60、0.66,表明APSIM模型对模拟干旱胁迫条件下陇中黄土高原区旱地春小麦产量形成具有较好的适应性。(2) 不同生育期干旱胁迫下,拔节期干旱胁迫对小麦籽粒数影响最大,其次由大到小依次为出苗期、分蘖期、无胁迫、抽穗期、开花期和灌浆期;灌浆期干旱胁迫对小麦千粒重影响最大,其次由大到小依次为开花期、抽穗期、无胁迫、拔节期、出苗期和分蘖期;拔节期干旱胁迫对小麦产量影响最大,其次由大到小依次为灌浆期、抽穗期、开花期、出苗期、无胁迫和分蘖期。(3) 不同程度干旱胁迫下,灌水量300.00 mm旱地春小麦产量最大为4866.19 kg·hm-2,与其他4种灌水相比产量分别增加283.53%、39.65%、0.46%和15.58%。(4) 出苗后第1 d、47 d、60 d、82 d、86 d灌水,且灌水量达到343.09 mm时,旱地春小麦产量最大为5578.91 kg·hm-2。干旱胁迫发生时间和程度对研究区小麦产量形成具有明显的交互作用,分蘖期适度干旱胁迫有利于提高陇中黄土高原区旱地春小麦产量,而拔节期和灌浆期为旱地春小麦田间水分管理的关键生育期,小麦生长发育过程中应加强该生育期的水分管理以提高陇中黄土高原区粮食产量。

关键词: 干旱胁迫, 小麦产量, 产量构成, APSIM模型, 陇中黄土高原区

Abstract:

We investigate the response mechanisms of different drought stress on growth and yields of spring wheat in the Loess Plateau Region of central Gansu Province, China. In this study, we collected and analyzed field experimental data from the Anjiagou Fengxiang Town, Anding District, Dingxi City from 2016 to 2018, and meteorological data in the Anding District, Dingxi City from 1971 to 2018. The field observation data were used to verify the validity of the Agricultural Production Systems Simulation (APSIM) model under different drought stress. The effects of different growth stages and degrees of drought stress on growth, grain numbers, thousand-grain weight, and yields of spring wheat were analyzed using the APSIM model. Finally, the optimum irrigation time and quota were selected using the results from the multiple regression analysis. Our results revealed the root means square error values of the phenology dates, grain numbers, thousand-grain weight, and yield of the simulation model were less than 3.67 d, 300.52 grain·m-2, 2.56 g, and 267.43 kg·hm-2, respectively. The normalized root means square error values were less than 3.89%, 2.86%, 9.71%, and 11.58%, respectively. The model effectiveness index (ME) values were higher than 0.62, 0.78, 0.60, and 0.66, respectively. The results indicate that the model had a good fitting and adaptability under drought stress in the study area. The most influential factor for spring wheat grain numbers was drought stress at the jointing stage, followed by the emergence, tillering, no stress, heading, flowering, and grain filling stages. The most influential factor for the weight of wheat was drought stress at the grain filling stage, followed by the flowering, heading, no stress, jointing, emergence, and tillering stages. The most influential factor for wheat yield was drought stress at the grain filling stage, followed by the grain filling, heading, flowering, emergence, no stress, and tillering stages. Under different degrees of drought stress, the yield of irrigation quota of 300.00 mm was the maximum value of 4866.19 kg·hm-2. The yield of irrigation quota of 300.00 mm increased by 283.53%, 39.65%, 0.46%, and 15.58% for 100.00 mm, 200.00 mm, 400.00 mm, and 500.00 mm, respectively. The optimal irrigation times were 1, 47, 60, 82, and 86 d after emergence, and the optimal irrigation quota was 343.09 mm, and the yield of wheat reached maximum values of 5578.91 kg·hm-2. A distinct interaction existed between the intensity and occurring stages of drought stress. In general, proper drought stress at the tillering stage was beneficial to increase wheat yield, but the jointing and grain filling stages were critical periods of water demand for spring wheat under drought stress. Reasonable irrigation management is needed at these growing stages to increase wheat yield.

Key words: drought stress, wheat yield, yield component, APSIM model, the Loess Plateau of central Gansu