中国农业气象 ›› 2020, Vol. 41 ›› Issue (01): 51-60.doi: 10.3969/j.issn.1000-6362.2020.01.006

• 论文 • 上一篇    下一篇

中国北方主产地苹果始花期与气候要素的关系

刘璐,王景红,傅玮东,栾青,李曼华   

  1. 1.陕西省农业遥感与经济作物气象服务中心,西安 710014;2.新疆农业气象台,乌鲁木齐 830002;3.山西省气候中心,太原 030006;4.山东省气候中心,济南 250031
  • 出版日期:2020-01-20 发布日期:2020-01-20
  • 作者简介:刘璐,E-mail:liululu128@163.com
  • 基金资助:
    中亚大气科学研究基金;陕西省重点研发计划(2018ZDCXL-N-24-3);秦岭和黄土高原生态环境气象重点实验室开放研究基金(2019Z-5)

Relationship between Apple’s First Flower and Climate Factors in the Main Producing Areas of the Northern China

LIU Lu, WANG Jing-hong, FU Wei-dong, LUAN Qing, LI Man-hua   

  1. 1.Shaanxi Meteorological Service Center of Agricultural Remote Sensing and Economic Crops, Xi’an 710014, China;2. Xinjiang Agricultural Meteorological Station, Urumqi 830002;3. Shanxi Provincial Meteorological Center, Taiyuan 030006;4. Shandong Provincial Meteorological Center, Jinan 250031
  • Online:2020-01-20 Published:2020-01-20

摘要: 选取福山、万荣、洛川、旬邑和阿克苏分别代表中国渤海湾、黄土高原和新疆苹果产区,利用1999?2018年各地富士苹果始花期物候观测数据,分析苹果始花期变化趋势,并利用偏最小二乘回归法,分析日尺度平均气温、平均地温、降水量、相对湿度和日照时数对苹果始花期的影响,明确影响苹果始花期的关键气候因子及其影响时段和强度,在此基础上,利用逐步回归法建立各主产地苹果始花期预测模型。结果表明,近20a来,中国北方主产地苹果始花期平均出现在4月7-20日,但变化趋势不显著。影响各地始花期的关键气候因子中,平均气温和平均地温为主导因子,其中关键影响时段内平均气温和平均地温每上升1℃,5个主产地苹果始花期将显著提前2.31~4.10d和2.34~4.96d;降水量每增加1mm,万荣、旬邑和阿克苏苹果始花期将显著推迟0.12~0.57d;平均相对湿度每增加1个百分点,旬邑和阿克苏苹果始花期推迟0.33d和0.51d;日照时数每增加1h,福山和万荣苹果始花期分别提前0.12d和0.07d。在明确影响各主产地苹果始花期关键气候因子的基础上,建立中国北方苹果主产地始花期预测模型,经回代检验,实测值与预测值相差小于5d的比例达80%~90%,模型可用于实际预测业务。

关键词: 富士苹果, 始花期, 偏最小二乘回归法, 预测模型, 气候变化

Abstract: Choosed Fushan, Wanrong, Luochuan, Xunyi and Akesu to respresent the Bohai Gulf, the Loess Plateau and Xinjiang apple producing areas, respectively, apple’s first flower data at the 5 sites during 1999?2018 were used to analyze their linear trend. Partial Least Squares (PLS) regression was applied to identify the impacts of mean temperature, mean geotemperature, precipitation, mean relative humidity and sunshine hours on first flower data at daily resolution. On the basis of which, forecasting models of first flower data were established by using stepwise regression. The results indicated that in the past 20 years, regional mean occurrence dates of apple's first flower were in April 7 to 20, and the interannual trend of first flower dates were not significant in all the five producing areas. Among the five climatic factors affecting the first flower period, temperature is the dominant factor. During the influence periods, the mean temperature and the mean geotemperature increase by 1℃, the first flower date will be significantly advanced by 2.31-4.10 days and 2.34-4.96 days; the precipitation increase by 1mm, the first flower date at Wanrong, Xunyi and Akesu would be postponed by 0.12-0.57 days; the relative humidity increase by 1 percent, the first flower date at Xunyi and Akesu would be postponed by 0.33 days and 0.51 days; the sunshine hours increase by 1 hour, the first flower date at Fushan and Wanrong would be advanced by 0.12 days and 0.07 days. On the basis of defining the key climatic factors affecting the first flower data of apple in the main producing areas of the northern China, forecasting models of first flower data were established, and the test showed that the ratio of the difference between the measured and predicted values less than 5 days was 80%?90%, which could be used for practical forecasting business and provide technical support for meteorological service of apple flowering period in China.

Key words: Fuji apple, First flower date, Partial least squares regression, Forecasting model, Climate change