Detecting Onset of Rainfall Using Recurrence Quantification Analysis

Abstract:

The onset of rainfall is critical in some climatic change indices, hence, the determination is very important. This study uses a recently developed nonlinear analysis tool - recurrence quantification analysis to determine the onset of rainfall in a tropical station. The RQA has been used successfully in examining structures of several physical and natural time series data. It has the capability to detect hidden structures such as chaoticity in a time series. The data was obtained from ongoing meteorological campaign at Federal University of Technology, Akure. The data was obtained using the tipping bucket method for a period of four years. RQA tools such as determinism and laminarity are used. Also, nonlinear structures in the rainfall data for the region will be investigated. The results obtained are compared with those obtained using conventional means. This method proves useful as it accurately determine the onset of rainfall with a reasonable level of accuracy.