Downslope winds on Eastern Slopes of the Central Andes mountains of South America in a 20-yr climate simulation with the ETA Model

PABLO L. ANTICO; SIN CHAN CHOU; FEDOR MESINGER; CAROLINE MOURÃO

Santiago

Conferencia; 11th International Conference on Southern Hemisphere Meteorology and Oceanography; 2015

American Meteorological Society y Facultad de Ciencias Físicas y Matemáticas, Universidad de Chile

Downslope winds often blow east of the Andes mountains causing extremely warm and dry conditions, sometimes accompanied by strong gusts causing property damage and adverse health effects, among other problems. These winds, that are a local version of the alpine foehn, are called Zonda winds in reference to a river valley where they normally occur. However, the term Zonda is used to describe downslope wind to the east of the Andes mountains anywhere it occurs. Zonda wind is a semi-permanent meteorological feature in the high mountain, and it is referred to as high Zonda wind in opposition to surface Zonda wind. The Andes mountain range extends through tropical and extratropical latitudes along the western edge of South America. Only north of approximately 35ºS, mountains are high enough (above 4.000 m) to prevent the cold and wet air from the Pacific Ocean be lifted and then be dropped downslope to the east. However, under specific conditions, this situation may be overcome and then a Zonda wind episode occurs causing temperature rise and dryness due to adiabatic compression on the lee side of the mountains.In this study, the effect of Zonda-like winds is analyzed on the lee side of Los Andes for a 20-yr climate simulation for South America and results are compared with previous observational evidence of Zonda wind. We adopt the term 'Zonda-like wind (ZLW)' to differentiate the observed Zonda wind (ZW) from that produced by the model. We include an index for the zonal pressure gradient across the mountain, an index for the vertical movement and two relative humidity (RH) indices evaluated at different levels on the lee side to define ZLW episodes.The model performance is explained not only in terms of the vertical coordinate design but also by the slantwise adjustment and finite-volume like scheme for temperature and momentum vertical advection, as it surges from the analysis of individual ZLW episodes.The resulting annual distribution of ZLW occurrences show a preference for winter and spring with almost no occurrences during summer. For the surface ZLW, the higher frequency occurs during spring. When considering the total number of high ZLW events, there is almost no preference for a given start time. However, surface ZLW episodes have a marked preferred start time during the afternoon. Our results are mostly in concordance with previous studies exclusively based on observational data.