Synthetic generation of daily streamflow sequences is one of the most critical issues in stochastic hydrology. In this study, a new wavelet transform method is developed for synthetic generation of daily streamflow sequences. Firstly, daily streamflow sequences with different frequency components are decomposed into the series of wavelet coefficients W
1(t), W
2(t),...,W
P
(t) and scale coefficients (the residual) C
P
(t) at a resolution level P using wavelet decomposition algorithm. Secondly, the series of W
1(t), W
2(t),...,W
P
(t) and C
P
(t) are divided into a number of sub‐series based on a yearly period. Thirdly, random sampling is performed from sub‐series of W
1(t), W
2(t),...,W
P
(t) and C
P
(t), respectively. Based on these sampled sub‐series, a large number of synthetic daily streamflow sequences are obtained using wavelet reconstruction algorithm. The advantages of this newly developed method include: (1) it is a nonparametric approach; (2) it is able to avoid assumptions of probability distribution types (Normal or Pearson Type III) and of dependence structure (linear or nonlinear); (3) it is not sensitive to the original data length and suitable for any hydrological sequences; and (4) the generated sequences from this method could capture the dependence structure and statistical properties presented in the data. Finally, a case study in Jinsha River, China, indicates that the new method is valid and efficient in generating daily streamflow sequences based on historical data.
