|
Significance tests for the relationship between "El Ni?o" phenomenon and precipitation in MexicoKeywords: enso, precipitation, statistical significance, mexican territory, climatic variability. Abstract: we describe the behavior of the statistical significance of the differences in rainfall in mexico during the occurrence of "el ni?o", "la ni?a", or neutral conditions. this analysis, which had not been made before, reveals the areas in which the effect of enso is statistically significant. the analyses were carried out for the wet season (june, july, august and september), for the dry season (december, january and february), and for the whole year. straight lines were fitted using precipitation as the dependent variable and the value of the multivariate enso index (mei) as the independent variable. we also compared precipitation between different "el ni?o", "la ni?a" and neutral conditions using the wilcoxon - mann - whitney test. the results show that during the occurrence of 'el ni?o", precipitation decreases in the south and increases significantly in northern and northwestern mexico. by contrast, in years under "la ni?a", rainfall increases in the south and decreases towards the north. in central mexico differences with respect to neutral conditions are not significant. comparing the wet and dry seasons, their behavior is somewhat different. during the wet season, and under "el ni?o" conditions, the rainfall decreased in the southern and central areas, remaining roughly unchanged in the north. under "la ni?a" conditions and the wet season, the precipitation increased in the south. during the dry season and under "el ni?o" conditions, rainfall increases towards north and northwest, while with "la ni?a" it decreases southwards. the enso phenomenon explains between 3.7 and 8.9% of the variance of the precipitation; this may be due to the many weather phenomena affecting precipitation in varying and unknown proportions. nevertheless, in some important regions of the country the enso phenomenon influences the precipitation in a way that is statistically significant.
|