摘要
运用季节性叠加趋势模型、季节周期-ARMA组合模型对奇台绿洲地下水位动态变化进行了分析和预测,并通过了后验差检验。结果表明:两种时间序列模型均适合对地下水动态变化的预测,后验差比值c和小误差频率p分别为(0.22、0.99)和(0.21、1)。但经过综合对比,季节周期-ARMA组合模型的预测精度更高。应用其对井灌区2005-2015年水位埋深进行预测,未来10年中,地下水位埋深以年均1.18 m的速度下降,应及时加以控制。
Using seasonal trends superposition model,seasonal cycle-ARMA model odd combination,dynamic changes of Qitai Oasis groundwater was analyzed and forecasted,and adopted after test.the results show that the two time series models are suitable for the prediction of groundwater dynamics,a posteriori error ratio c and the frequency of small errors p were(0.22,0.99) and(0.21,1) respectively.However,after comparison of the overall combination the seasonal cycle-ARMA model has higher prediction accuracy.Application of the model to the forecast of graund water table from 2005 to 2015,the annual groundwater level will yearly decline at the rate of 1.18m in the next 10 years and should be controlled in time.
出处
《水资源与水工程学报》
2011年第1期1-6,共6页
Journal of Water Resources and Water Engineering
基金
国家自然科学基金项目(40771194)
北京市属高等学校人才强教计划资助项目(PHR200906125)资助
关键词
地下水
时间序列模型
地下水动态
奇台绿洲
ground water
time series model
groundwater dynamic change
Qitai oasis
作者简介
鲁魁锋(1986-),男,河南长葛人,硕士研究生,主要从事区域可持续发展研究。
通讯作者:熊黑钢(1956-),男,湖南湘乡人,教授,博士,现主要从事干旱区资源环境研究。