Evolution of Air Temperature and Multiscale Characterization of Greenhouse Gases in Taiwan based on Multi-dimensional Ensemble Empirical Mode Decomposition and Noise-assisted Multivariate Empirical Mode Decomposition

Global warming has always been a topic of concern, mainly due to climate change caused by the greenhouse effect. Due to the impact of global warming, the snow cover on the mid-high latitudes has decreased, which has weakened the Asian winter monsoon, resulting in a high temperature increase on the East Asian coast. Therefore, Taiwan is one of the most sensible areas under global warming that the average temperature increase over the past 100 years is about 1.25 degrees, which is higher than the global average. However, human influence including the emission of greenhouse gases such as carbon dioxide, methane and nitrous oxide is seen as the main cause of warming observed since the mid-20th century. In this study, to explore the impact of global warming in Taiwan, we propose to analyze data of temperature and concentrations of greenhouse gases in Taiwan for decades by using the time frequency analysis method. Time frequency analysis is a tool that allows us to observe both the time and frequency domain information and to investigate the characteristic time scale and energy distribution of a signal. Most observations of the natural and man-made systems are likely to be nonlinear and nonstationary, however, traditional data-analysis methods such as Fourier transform is based on linear and stationary assumptions. Thus, the method of noise-assisted multivariate empirical mode decomposition (NAMEMD) and multi-dimensional ensemble empirical mode decomposition (MEEMD) for multidimensional data such as images are introduced in this study. From this study, it is expected to know whether the evolution of air temperature in Taiwan is mainly affected by these greenhouse gases.