农业气象台站玉米生育期的填补及对比分析

Filling and comparison of the growth period data of agricultural meteorological stations

农业气象台站记录的作物生育时期数据,广泛用于科研和生产活动,但是该数据集有较多生育时期缺失。为了高效、充分地利用已有生育时期,急需研究生育期填补方法,并分析与其他生育期数据的差异。本研究以黄淮海夏玉米区为例,结合历史气象数据,分析农业气象台站玉米关键生育阶段的积温比例,研究玉米缺失生育时期的填补方法;对比分析2002—2011年农业气象台站与国家玉米品种区域试验生育时期数据的差异。结果表明,本文提出的方法能有效地填补缺失生育时期,方法的均方根误差在1.82～5.20之间;除2006年、2007年以外,其余年份50%以上的农业气象台站数据与国家玉米区域试验生育期一致性好;就多年数据的平均水平而言,黄淮海夏玉米区内大部分农业气象台站与国家玉米区域试验生育期差异较小,约占农气台站总数的53%。该方法可广泛应用于其他玉米种植区、农作物的生育期数据填补和对比,为相关研究提供生育期数据选取、融合的参考。

Accurate data of growth period are the basis for crop growth and production process information service. The crop growth period data recorded by agricultural meteorological include a lot of growth stage period information missing. Taking the Huang-Huai-Hai summer maize area as an example, this study calculated the accumulated temperature distribution ratio combining with the daily average temperature, and filled the information of the missing stage from 2002 to 2011.Ultimately this method filled the original growth period data of 13.8%, and the error of filling method was characterized. The results demonstrated that the errors in different growth stages were inconsistent, of which the error in milky maturity stage and maturity stage was the largest. In order to compare the difference between the growth period data of agricultural meteorological stations and the national maize regional test data, a variety of interpolation methods were applied to the two sets of data. By comparing MAE(Mean Absolute Error) and RMSE(Root Mean Square Error), we chosen the Ordinary Kriging method. We compared the interpolated insults and defined difference values between [0, 5] indicating a small difference, large difference at(5, 10], and extremely difference at(10, +∞). Our result shows that, with the exception of 2006 and 2007, the majority of the two sets growth period data difference are limited, in which the difference between the two sets of growth period data was the least appeared in 2003, 2005. The difference between the two sets of growth period data was the largest in 2006. In terms of average of multiple years, there is not a significant difference in most parts of the study area, accounting for 53% of the total number of the agricultural meteorological stations. This study can be used in the filling and comparison of growth stages data in other maize growing areas or other crops.