湖北漳河灌区中稻气象产量变化特征及预测模型  

Variation Features and Estimation Model for Meteorological Yield ofMid-season Rice in Zhanghe Irrigated Area of Hubei Province

作　　者：余蕾[1] 邹志科 刘凤丽[1] 罗文兵[1] 王文娟 YU Lei;ZOU Zhi-ke;LIU Feng-li;LUO Wen-bing;WANG Wen-juan(Agricultural Water Conservancy Department,Changjiang River Scientific Research Institute,Wuhan 430010,China)

机构地区：[1]长江科学院农业水利研究所,武汉430010

出　　处：《长江科学院院报》2024年第2期82-90,共9页Journal of Changjiang River Scientific Research Institute

基　　金：国家自然科学基金委员会-中华人民共和国水利部-中国长江三峡集团有限公司长江水科学研究联合基金项目(U2040213);重庆市技术创新与应用发展专项重点项目(CSTB2022TIAD-KPX0198)。

摘　　要：准确的气象产量是正确评估气象条件对粮食产量影响的前提。为了探究湖北漳河灌区单季稻气象产量的时间序列变化规律,尝试通过三点滑动平均法、HP滤波法和一次指数平滑法、二次指数平滑法4种方法将漳河灌区1975—2020年的水稻单产数据分离为水稻趋势产量及气象产量,通过相关分析筛选水稻8个生育阶段的主要气象因子,然后与分离的气象产量构建水稻预测模型。结果表明:4种分割方法均能较好地反映气象产量序列与湖北省生产力发展水平的区域一致性特点,多年平均气象产量占总产量的比例约为3.39%,但2008年以后其占比达到10.1%。相关分析识别出抽穗开花期最低气温、拔节孕穗期最高气温、分蘖后期平均气温、返青期最低气温、乳熟期蒸发量和育苗最低气温是影响气象产量的主要因子,该模型在率定期(1976—2014年)和验证期(2015—2020年)的相对误差均在5%以内,模型的决定系数R 2为0.994。预测模型有助于研究未来气候变化下的区域水稻产量的变化。Precise estimation of meteorological yield is a premise of accurately assessing the impact of meteorological conditions on rice yield.This study delves into the time series variation of meteorological yield for single-season rice in the irrigated areas of Zhanghe,Hubei Province.Four methods,in specific,three-point moving average method,HP filtering method,single exponential smoothing method,and quadratic exponential smoothing method,were applied to decompose the rice yield data from 1975 to 2020 into trend yield and meteorological yield.Through correlation analysis,eight meteorological factors associated with rice growth stages were identified and used to construct a rice prediction model alongside the separated meteorological yield.The findings indicate that the four methods effectively capture the regional consistency between meteorological yield series and productivity level in Hubei Province.The annual average meteorological yield accounted for approximately 3.39%of the total output,and after 2008 the figure exceeded 10.1%.Via correlation analysis,the key factors influencing meteorological yield were identified as follows:minimum temperature at heading and flowering stage,maximum temperature at jointing and booting stage,average temperature at late tiller stage,minimum temperature at returning-green stage,evaporation at milk grain stage,and minimum temperature at seedling raising stage.During calibration period(1976-2014)and validation period(2015-2020),the model exhibited relative errors less than 5%,and a determination coefficient(R 2)reaching 0.994.The proposed model holds potential for facilitating the study of regional rice production under future climate change scenarios.

关 键 词：中稻气象产量 趋势产量 指数平滑法 滑动平均法 HP滤波法 漳河灌区 

分 类 号：S16[农业科学—农业气象学]