GAMLSS模型在黄河流域蒸发皿蒸发量序列统计建模中的应用研究

本文采用广义可加模型(GAMLSS),以年份序号作为解释变量拟合模型参数,对黄河流域蒸发皿蒸发量序列的非一致性特征进行研究。考虑到流域空间差异性和年内变化,从年、汛期、非汛期3个时间尺度来研究黄河流域29个水资源三级区面蒸发量序列的非一致性特征。研究结果表明,黄河流域各水资源三级区的面蒸发量序列在不同时间尺度上均具有较为明显的趋势变化;非汛期蒸发序列呈现出显著的趋势变化,与一致性模型拟合结果相比,黄河流域非汛期蒸发序列的AIC准则(Akaike Information Criterion)和SBC准则(Schwarz Bayesian Criterion)值减少明显,能更好地反映出非一致性变化的特征。

Application research of GAMLSS model in statistical modeling of evaporation sequence in the Yellow River Basin

In this paper, the generalized additive model (GAMLSS) is used to fit the model parameters with the year serial number as the explanatory variable to study the non-consistent characteristics of the evaporation pan evaporation series in the Yellow River Basin. Taking into account the spatial differences of the basin and changes within the year, the non-consistent characteristics of the 29 water resources tertiary area evaporation data series in the Yellow River Basin were studied from three time scales:annual, flood season and non-flood season. The research results show that the surface evaporation sequences of the three tertiary water resources of the Yellow River Basin have obvious trends on different time scales; the non-flood season evaporation sequences show a significant trend change, compared with the results of the consistency model The values of the AIC (Akaike Information Criterion) and SBC (Schwarz Bayesian Criterion) values of the non-flood season evaporation sequence in the Yellow River Basin are significantly reduced,which can better reflect the characteristics of non-uniform changes.