CONICET | Buscador de Institutos y Recursos Humanos

This paper examines the relationship between monthly rainfall and El Niño-Southern Oscillation (ENSO) phenomenon, explores a system to forecast monthly rainfall likelihood, based on the ?phase locking? of the Southern Oscillation Index, in comparison to ENSO events. The studied region, the Pampas, located in central-eastern Argentina, is one of the areas of greatest agricultural and farming potential throughout the world. This research intends to have direct application to agricultural decision support programs. Spatial patterns of several statistical properties of rainfall are analyzed for each month, considering cold and warm events. Differences between these properties are not always statistically significant, depending on the station and the developing stage of the ENSO event. A high-degree of regional variability is found; nevertheless, monthly rainfall tends to be above normal values in warm events and below the normal in cold ones. A remarkable feature is the reversal of the sign of the anomalies during February(1) in cold events and a weakening (and even a sign reversal) in some stations in warm events. The probability of exceeding the neutral-median is better defined: low probability-cold events and high probability-warm events. The ENSO-related rainfall signal is used as reference in the probabilistic rainfall forecast based on Southern Oscillation Index phases. The rainfall pattern in warm (cold) event is associated with the SOI phase 1 or ?consistently negative? phase (SOI phase 2 or ?consistently positive?). A lead-time of three to six months, especially for November (0), appears to be feasible. However, the lead-time found in SOI phases methodology does not improve the ?forecast? provided by the ENSO methodology occurrence; given ENSO event has already entered into the development stage. The high degree of regional variability found in ENSO-related rainfall has to be taken into account in any rainfall forecast for the region.

enviar mensaje