This paper investigates a signal modality analysis for the characterisation and detection of nonlinearity in crude oil markets. Given the nonlinear and time-varying characteristics of international crude oil prices, this study employs the recently proposed delay vector variance (DVV) method that examines local predictability of a signal in the phase space to detect the determinism and nonlinearity in a time series. In addition, wavelet transforms, which have recently emerged as a mathematical tool for multi-resolution decomposition of signals, is utilised. In particular, among the wavelet methodologies considered, the complex Morlet wavelet is useful and best at detecting the various phases of oil prices through the trajectory of market developments. The findings of this paper highlight the significant phases of the series and its relation to real-world phenomena with an indication of early warning signals of future significant events, thereby providing a guide for proper decision making and risk management practices of market participants.