Analysis and downscaling multi-model seasonal forecasts in Peru using self-organizing maps

Abstract

We present an application of Self-Organizing Maps (SOM) for analyzing multi-model ensemble seasonal forecasts from the DEMETER project in the tropical area of Northern Peru. SOM is an automatic data-mining clustering technique which allows summarizing a high-dimensional data space in terms of a set of reference vectors (cluster centroids). Moreover, it has outstanding analysis and visualization properties, since the reference vectors can be projected into a 2D lattice preserving their high-dimensional topology.