Abstract

Long–term monthly precipitation series registered at 56 meteorological stations located within Mexico’s state of Zacatecas were analyzed in order to identify their trends. We analyzed the power spectrum signals of precipitation anomaly series in order to identify their important frequencies and its possible connection with periodic phenomena. We found negative linear trends for 28 out of 56 precipitation series and 28 of the 56 were increasing linear trends. Only one of the increasing trends was significant at p<0.05. Noise in 24 out of 56 of the monthly precipitation anomaly series was found to have persistent behavior, anti–persistent behavior was found in 26 out of 56 of the monthly precipitation anomaly series, and random walk noise was found in 6 out of 56 of monthly precipitation anomaly series. Additionally, we found important frequencies that could be related with yearly cycle, quasi-biennial cycle, ‘EL Niño Sothern Oscillation’ phenomena and sunspot cycle.

Key words: Linear trends, fractal dimension, power spectrum density, yearly cycle, quasi–Biennial cycle, ENSO, sunspot cycle.