ENSO and IOD analysis on the occurrence of floods in Pakistan

Abstract

Different techniques have been used to discuss the existence of significant relation between the El Nino Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD). Various studies present their interaction and influence on the natural disasters (i.e. drought, flood, etc.) over large parts of the globe. This study uses a Markov chain method to investigate the relation between the ENSO and IOD for the period of 62 years (1950–2011) and aggregates their influence on the occurrence of floods in Pakistan. Both data sets show similarities in the formation of transition matrices and expected number of visits from one state to another. The strong values of 2-dimensional correlation and high self-communication of the transition states confirm the existence of a possible relation between ENSO and IOD data. Moreover, significant values of dependency and stationary test endorse the applicability of the Markov chain analyses. The independent analysis shows that strong events of both data sets are co-occurred in the same flood years. During the study period maximum number of floods was observed during summer monsoon season. However, further analysis shows that after 1970, Pakistan observed the highest percentage of floods occurred per year during El Nino, Non-ENSO and positive IOD years. These observations and results demonstrate that climate variability especially ENSO and IOD should be incorporated into disaster risk analyses and policies in Pakistan.