摘要
参照腾发量ET0的实时预测对实时灌溉预报很重要。通过对普通天气预报信息进行解析,取得可用的合理数据,利用Penm an-M on te ith方法估算了北京大兴试区近10年逐日参照腾发量,最后与由实测气象数据计算的结果进行了对比分析。结果表明:解析气象因子与实测数据中,日照时数的相关系数为0.99,风速为0.90;t检验值日照时数为376.9042,风速为122.4295,远远大于t分布相应临界值2.576(α=0.01),表明其可以认为是来自一个近似的总体样本。由日最低气温确定的实际水汽压和由实测相对湿度计算的实际水汽压,二者相关系数达到0.93,t检验值为153.3015。运用天气预报信息计算预测的ET0与实测数据用Penm an-M on te ith方法计算的ET0相比,相关系数达到0.9613,t检验值为209.1194,说明二者具有高度显著的线性相关性。如果日常天气预报准确度能够达到90%以上,用此理论预测参照腾发量将具有较大的参考价值和实际意义。
Real-time irrigation forecast relies on the prediction of daily reference evapotranspiration. The Penman- Monteith equation has been a global standard for reference evapotranspiration （ET0） estimation. In this study, an attempt was made to use the public weather forecast messages to predict daily ETo with Penman-Monteith method. The so-determined results were compared with those estimated by Penman-Monteith method, with the recorded daily weather information of an automated meteorological station in Daxing District, Beijing. The key factors of meteorological data and the difference of the calculated actual vapor pressures were compared between those from the weather forecast messages and those from the automated station records. Both sunshine hour and wind speed had significant agreement with the statistic analyses. On a daily basis, the regression coefficients （Rc） are 0. 9876 and 0. 8982, and the values of t-test （Tv） are 376. 9042 and 122. 4295, respectively. The two groups of estimated actual vapor pressures have good linear relationship （Rc and Tv being 0. 9314 and 153. 3015, respectively）. All of their Tv values were bigger than the critical value of t-distribution （which is 2. 576 as α= 0.01）. This indicates the high accuracy of the daily ETo predicted with weather forecast messages. The proposed method is quite reliable. High precision of estimated ETo depends on accurate weather forecast messages availability.
出处
《农业工程学报》
EI
CAS
CSCD
北大核心
2005年第11期11-15,共5页
Transactions of the Chinese Society of Agricultural Engineering
基金
国家"863"计划(2002AA2Z4071-10)
水利部科技创新项目(SCX2003-14)
国家自然科学基金资助项目(50479043)
