Spring drought prediction based on winter NAO and global SST in Portugal |
| |
| Authors: | João Filipe Santos Maria Manuela Portela Inmaculada Pulido‐Calvo |
| |
| Affiliation: | 1. Dpto. Engenharia, ESTIG, Instituto Politécnico de Beja, Portugal;2. Dpto. Engenharia Civil, SHRH, Instituto Superior Técnico (Lisboa), Lisboa, Portugal;3. Dpto. Ciencias Agroforestales, Escuela Técnica Superior de Ingeniería, Campus La Rábida, Universidad de Huelva, Palos de la Frontera (Huelva), Spain |
| |
| Abstract: | The aim of this paper is to test the ability of neural network approaches to hindcast the spring standardized precipitation index on a 6‐month time scale (SPI6) in Portugal, based on winter large‐scale climatic indices. For this purpose, the linkage of the spring SPI time series with the winter North Atlantic Oscillation (NAO) and the sea surface temperature (SST) was investigated by means of maps of the correlation coefficient for the period from October 1910 to September 2004. The results indicate that the winter NAO is a good predictor for the SPI6 of the spring (SPI6 finishing in April, May and June, SPI6April, SPI6May and SPI6June, respectively) for the northern, central and southern regions of Portugal. The winter SST1 (area of the Mediterranean Sea) must only be considered for the northern region, and the winter SST3 (area of the North Atlantic between Iberia and North America) only for the southern region. Spatial maps of predictive SPI6 for April, May and June were created and validated. The neural models explained more than 81% of the total variance for the SPI6April and SPI6May and more than 64% of the total variance for the SPI6June. Probability maps were also developed considering the values predicted by the neural methods for the spring months and all drought categories (moderate, severe and extreme). These maps indicating the probability of droughts can provide valuable support for the integrated planning and management of water resources throughout Portugal. Copyright © 2012 John Wiley & Sons, Ltd. |
| |
| Keywords: | artificial neural networks hindcasting standardized precipitation index climatic indices NAO |
|
|
