There is a considerable uncertainty in a hydrological projection, which arisen from the multiple stages composing the hydrological projection. Uncertainty decomposition analysis evaluates contribution of each stage to the total uncertainty in the hydrological projection. Some uncertainty decomposition methods have been proposed, but they still have some limitations: (1) they do not consider nonstationarity in data and (2) they only use summary statistics of the projected data instead of the full time-series and lack a principled way to choose the summary statistic. We propose a novel Bayesian uncertainty decomposition method which can alleviate such problems. In addition, the proposed method provides probabilistic statements about the uncertainties. We apply the proposed method to the streamflow projection data for Yongdam Dam basin located at Geum River in South Korea.

中文翻译：

---

水文预测的贝叶斯不确定性分解

水文投影存在很大的不确定性，这是由组成水文投影的多个阶段引起的。不确定性分解分析评估每个阶段对水文预测中总不确定性的贡献。已经提出了一些不确定性分解方法，但是它们仍然存在一些局限性：（1）他们不考虑数据的非平稳性；（2）他们仅使用预测数据的摘要统计信息，而不是整个时间序列，并且缺乏有原则的方法选择摘要统计信息。我们提出了一种新颖的贝叶斯不确定性分解方法，可以减轻此类问题。另外，所提出的方法提供了关于不确定性的概率陈述。