欢迎访问《中国农学通报》，
中国农学通报

中国农学通报 ›› 2025, Vol. 41 ›› Issue (35): 97-102.doi: 10.11924/j.issn.1000-6850.casb2025-0774

• 资源·环境·生态·土壤·气象 • 上一篇    下一篇

甘肃省正宁县苹果始花期预报模型研究

李美瑜1(), 陈帆2, 张天峰3, 柳家祺1, 周忠文3, 王娟3, 张雪姣4   

  1. 1 甘肃省正宁县气象局，甘肃正宁 745300
    2 甘肃省镇原县气象局，甘肃镇原 744500
    3 甘肃省庆阳市气象局，甘肃庆阳 745005
    4 甘肃省合水县气象局，甘肃合水 745400
  • 收稿日期:2025-09-09 修回日期:2025-11-24 出版日期:2025-12-11 发布日期:2025-12-11
  • 作者简介:

    李美瑜，女，1989年出生，陕西佳县人，高级工程师，本科，研究方向：地面观测及农业气象。通信地址：745300 甘肃省正宁县气象局，E-mail：。

  • 基金资助:
    甘肃省气象局“飞天风云”青年基层人才项目“影响正宁县苹果始花期的关键气候因子分析及始花期预报研究”(2425rczx-D-QNJCRC-03)

Research on Forecast Model of Apple Initial Flowering Period in Zhengning County of Gansu

LI Meiyu1(), CHEN Fan2, ZHANG Tianfeng3, LIU Jiaqi1, ZHOU Zhongwen3, WANG Juan3, ZHANG Xuejiao4   

  1. 1 Meteorological Bureau of Zhengning County, Gansu Province, Zhengning, Gansu 745300
    2 Meteorological Bureau of Zhenyuan County, Gansu Province, Zhenyuan, Gansu 744500
    3 Meteorological Bureau of Qingyang City, Gansu Province, Qingyang, Gansu 745005
    4 Meteorological Bureau of Heshui County, Gansu Province, Heshui, Gansu 745400
  • Received:2025-09-09 Revised:2025-11-24 Published:2025-12-11 Online:2025-12-11

RichHTML

2

PDF (PC)

6

摘要：

本研究基于甘肃省正宁县1984—2024年苹果始花期及同期气象观测数据，分析了正宁县苹果始花期的长期变化趋势，并识别了关键影响因子。综合运用主成分分析和逐步线性回归方法，构建了苹果始花期预报模型。结果表明：（1）近41 a间正宁县苹果始花期呈显著提前趋势，显著突变点在1994年，线性倾向率为4.2 d/10 a，共提前约17 d。（2）影响苹果始花期的关键性气象因子为气温、积温、平均地面温度、日照时数及相对湿度，始花期与冬春季（≥0℃、≥3℃、≥5℃、≥10℃）的活动积温、3月气温、日照时数、地面温度均呈显著负相关，与3月相对湿度呈正相关。这表明热量累积越充分，相对湿度越低，则苹果始花期越早。（3）所构建的始花期预报模型具有较高准确率：始花期预报值与实测值误差为0~3 d内的概率达68%，0~4 d内达87%。2022—2024年试报结果显示，误差仅0~1 d，模型预报效果较好，具备业务应用价值。

关键词: 苹果, 始花期, 预报模型, 气候变化

Abstract:

Using the data of apple initial flowering period and meteorological observations from 1984 to 2024 in Zhengning County, Gansu, the study analyzes the trend of changes in apple initial flowering period in Zhengning. Through the correlation analysis, the dominant meteorological factors affecting apple initial flowering period are identified, and a prediction model for apple initial flowering period is constructed using principal component analysis and stepwise linear regression. The results showed that: (1) from 1984 to 2024, the apple initial flowering period in Zhengning showed a significant advance trend, with a mutation point in 1994 and a linear trend rate of 4.2 d/10 a. The initial flowering period was advanced by 17 days in 41 years. (2) The key meteorological factors that affected the apple initial flowering period were temperature, accumulated temperature, average ground temperature, sunshine hours, and relative humidity. They were significantly negatively correlated with active accumulated temperature of ≥0℃, ≥3℃, ≥5℃, ≥10℃ in winter and spring, temperature, sunshine hours, and ground temperature in March, and positively correlated with relative humidity in March. This indicated that the more sufficient the heat accumulation, the lower the relative humidity, the earlier the beginning of apple flowering. (3) A prediction model for the apple initial flowering period was constructed using principal component analysis and stepwise linear regression. After testing, the accuracy of predicting the initial flowering period with a difference of 0-3 days from actual measurements reached 68%, and the accuracy of predicting the difference with a difference of 0-4 days reached 87%. A trial report was conducted on the apple initial flowering period in Zhengning from 2022 to 2024, and the difference between the measured and predicted values was 0-1 day. The model demonstrates satisfactory forecasting performance and can be applied in daily operational work.

Key words: apple, initial flowering period, forecast model, climate change