FORECASTING ZONDA WIND OCCURRENCE WITH ERA5 REANALYSIS

FEDERICO OTERO; DIEGO ARANEO

Buenos Aires

Congreso; CONGREMET XIV; 2022

Centro Argentino de Meteorólogos

Zonda wind is a typical downslope windstorm over the eastern slopes of the Central Andes in Argentina, which produces extremely warm and dry conditions and creates substantial socioeconomic impacts. In this study, the “synoptic fingerprints” of Zonda wind was obtained from statistical models based on Principal Components Analysis (PCA) applied to atmospheric diagnostic fields ERA5. A probability index was obtained to discriminate between Zonda and Non-Zonda occurrence in central Andes. This index allows to obtain a separation probability as well as an event-based and alarm-based probability. By using geopotential height and wind components for six pressure levels, the performance of the models was assessed with a contingency table that summarizes the occurrence of observed and predicted Zonda/Non-Zonda events. Two models were proposed in this work, one with a single variable/level and another with two combined variables/levels. The best models turned out to be those with low-levels variables. The main performance measures were the total efficiency (95.69%), probability of detection (99.13%), probability of false detection (0.87%), missing rate (0.87%), correct alarm ratio (99.08%), false alarm ratio (0.92%), and missed alarm ratio (0.95%). The synoptic conditions that best discriminates between Zonda/Non-Zonda events mainly depend on the number of components retained in the PCA for each particular model. Despite this, a common synoptic fingerprint is present. Studies of this kind of downslope windstorms are scarce in South America, so this work represents a substantial contribution to recognize synoptic-scale atmospheric structures leading to Zonda occurrence.