This study presents a 30-yr high spatiotemporal downscaling data set performed with the WRF model (1 km horizontal grid spacing, hourly, 1 September 1987 to 31 August 2017). A novel approach is used to investigate the climatology of Sundowner winds. Combined empirical orthogonal functions are employed to objectively characterize the variability of surface winds. It is shown that the first two modes (EOF-1, EOF-2) represent 60.5% of the total variance (45.3% and 15.21%, respectively). EOF-1 is characterized by large loadings in the western part of the domain and captures the intense winds in the western SYM and coastal jet. In contrast, EOF-2 represents strong winds near Gaviota and intense downslope winds in the eastern SYM (i.e., near Santa Barbara and Montecito cities) and San Rafael Mountains. The hourly frequency distributions of the first two time coefficients are used to identify the occurrence of significant downslope winds in the southern slopes of the SYM. Three types of Sundowner wind regimes are found depending on how winds project onto EOF-1 and EOF-2. The presentation will discuss the spatiotemporal variability of winds, temperature and moisture during Sundowner wind events.