Time Series, Trend and Wavelet Analysis of Water Parameters

Abstract

Time series, trend and wavelet analysis of water quality parameters Chemical Oxygen Demand (COD), Biochemical Oxygen Demand (BOD), Dissolved Oxygen (DO) monitored at Nizamuddin bridge mid-stream of river Yamuna in India have been studied. In Auto Regressive Integrated Moving Average (ARIMA) model (p, d, q) value of ‘d’ is zero thus process is stationary. It is observed that Root Mean Square Error (RMSE) value are comparatively very low thus dependent series is closed with the model predicted level. The Mean Absolute Percentage Error (MAPE), MaxAPE, MAE, MaxAE, normalized Bayesian Information Criterion (BIC) are calculated and have low value. Trend is calculated by using Auto correlation function (ACF), Partial auto correlation function (PACF) and lag. The predictive model is useful at 95% confidence limits. The 1-D discrete and continuous Daubechies wavelet analysis explains that the parameters COD, BOD, DO have the maximum value 120, 50, 8; the value of a5 varies between 52 to 78, 10 to 30, 0.2 to 1.4 and the scale values of Db5, i.e., d5 ranges between -10 to 10; -5 to 5 and -0.5 to 0.5, respectively. It is concluded that as the value of COD, BOD increases and value of DO decreases, water is not fit for drinking, agriculture and industrial use.