Relationship between precipitation anomalies and multivariate ENSO index through wavelet coherence analysis

A better understanding of the local and regional spatio-temporal variability of past precipitation is needed to contextualize climate change research.  Monthly precipitation data from 59 stations in the state of Zacatecas in Mexico were transformed into standardized monthly precipitation anomalies for the period 1976-2010 (34 years). Cluster analysis of the new time series was used to identify regions with similar precipitation regimes. Power-spectrum analysis was carried out to identify important frequencies. Wavelet coherence analysis was performed to detect significant relations between each of the regional monthly mean precipitation anomaly time series (SAMP) and the Multivariate El Nino Southern Oscillation Index (MEI). Three regions were identified: Semi-arid, Highlands and Canyons. In all the three regional SAMP series, the 0.5-, 1-, 2- and 3-year frequencies are present in the power–spectrum; the Highlands and Canyons regions showed important frequencies of 13 and 12 years, respectively. The 3-year frequencies may be linked to the El Nino Southern Oscillation phenomenon. Three periods with significant correlation between regional SAMP time series and the MEI were identified: 1993 to 2003 for the Semi-arid region, 1995 to 2003 for Highlands region and 1988 to 2003 for the Canyons region. The results suggest that precipitation anomalies vary over time and according to regions. The results also strongly indicate that precipitation is modulated by the MEI.