Author(s): Young-Hoon Jin; Chang-Ryol Oh; Sung-Chun Park; Kyong-Bum Roh
Keywords: Wavelet transform; Hydrological data; Water quality; Correlation
Abstract: Wavelet transform is applied for detection of relationship between hydrological and water quality data measured in two stations (JSD1 and JSD2) within Jangseong reservoir, Korea. Rainfall and storage amount are used for hydrological data, while water quality parameters for the present study include p H, Dissolved Oxygen (DO), Biochemical Oxygen Demand (BOD), Chemical Oxygen Demand (COD), Total Nitrogen (T-N) and Total Phosphorus (T-P). The period of monthly data for the respective parameters is between 1996 and 2010. A Daubechies wavelet function, namely‘db7’, is applied for discrete wavelet transform (DWT). Low frequency components which are called‘approximation’are separated from details which are high frequency components. The details can be considered as noise based on energy ratio. Due to the limited data number, decomposition for the raw data is carried out up to the second level. The approximation resulted from the application of DWT is used for calculation of correlation coefficients between the data used in the present study. The relationships estimated by approximation components are clearer and more significant than those between raw data, since the noise included in the raw data are segregated and removed by DWT.