Application of Recurrence Quantification Analysis (RqA) to Climatic Studies

English Abstract:

Scientists generally agree that climate change is real but there is no consensus on the cause. Statistical tools such as correlation, averaging are generally used to detect causes of climate change. However, this will not give a true picture since atmospheric parameters is nonlinear and chaotic. Poincare introduced recurrence in 1890 to investigate fundamental characteristics of dynamical systems. It has been used by researchers in fields such as astrophysics, neurosciences, earth sciences, amongst others to detect structures in time series data. Over the years, the field has extended to include Recurrence Quantification Analsysis (RQA), which has led to the introduction of indicators such as Recurrence rates (RR), Determinisim (DET), e.t.c. The applications of RQA to atmospheric sciences include detection of change points in time series e.g. when did significant changes occur in time series data. Complex relationship such as effect of global warming indicators (e.g carbon(iv) oxide) on atmospheric parameters (e.g. rainfall) can be investigated through phase, lag, generalized or complete synchronization. RQA can also be used to analyse spatial data for relationship. This technique can be used to validate results using other techniques. Examples will be presented using NCEP Reanalysis data for selected locations across the world to show the efficiency of the technique.