Generación de datos pluviometricos como contribución al conocimiento ecológico del noreste de la provincia de Buenos Aires

Resumen

Ecological studies using pluviometric data elaborated on the basis of homogeneous series, show sometimes the practical disadvantage of those series being interrupted or even incomplete. Thus, it arrises the need for methods to generate the lacking data.

The objective of the present work is to analyze the use of eigenvectors and eigenvalúes corresponding to a matrix, product of the matrix “Predpitation” times its transpose, to estímate lacking predpitation data. The results show a mean error of 10% or lower for monthly valúes, and up to 5% for annual valúes, when reconstructing the ten years recode for pluviometric stations not involved in the eigenvectors calculation.