3. Time Series Analysis

Time series analysis of hydrologic data helps understand the behaviour of a hydrologic variable over time. It aids in identifying long-term patterns and trends in the data, which may enhance the prediction of future water availability and flood or drought risk. The chapter introduces different types of time series data and discusses the decomposition of a time series into its constituent components like trend, cycle, seasonality, and irregularity. Methods to test the stationarity of a time series, e.g., the Autocorrelation function (ACF) plot and Augmented Dickey-Fuller (ADF) test, and approaches to converting data into stationary are presented. Similarly, methods to analyse trend, seasonality and periodicity, e.g., the Mann–Kendall Test, Sen's Slope Estimator, and partial autocorrelation function (PACF) plots, are included. The chapter also introduces the time series models like Autoregressive (AR), Moving Average (MA), Autoregressive Moving Average (ARMA) and Autoregressive Integrated Moving Average (ARIMA) models, which are typically used to model stochastic components of a time series.