Within this thesis the damage caused in forests by winter storm „Lothar“ on December 26, 1999, is analysed with respect to soil characteristics, topographical, and meteorological influences. The automated detection of storm damage areas is based on high resolution (2 x 2 m²) airborne measurements of the so called normalised difference vegetation index (NDVI) combined with land use data. A case study for an area around Baden-Baden provides some insight into the factors controlling the storm damage pattern. Differences in shape, age and thinning of neighboured forest stands affect storm damage. Orographical flow effects arise from higher wind speeds over the top and the sides of mountains and causes extensive damage independently from forest stand properties. A quantitative analysis of storm damage in the whole area of investigation (~5000 km²) with respect to soil characteristics and topographic parameters shows orographic flow acceleration similar to the case study. Certain structures of the large scale damage pattern cannot be attributed to orographic or local flow effects, but are caused by mesoscale variability of wind fields. A comparison of the observed damage pattern with simulations of the numerical Karlsruhe atmospheric mesoscale model (KAMM) shows that the regions of highest simulated wind velocities broadly match with storm damage areas.