This study assesses the effects of vegetation patterns and environmental factors on the abundance of natural tree and shrub regeneration in semi-arid forests of the Zagros Mountains, western Iran. We sampled 120 releves at different topographic positions in a protected area of the studied region. Floristic composition, slope, elevation and soil properties were recorded at each releve, and woody seedling density was measured. We have first discerned five floristic groups using two-way indicator species analysis (TWINSPAN) and detrended correspondence analysis (DCA) and then explored the relationships among the floristic group compositions, environmental factors and seedling densities. The indicator species of the 5 groups were Quercus brantii, Acer monspessulanum, Cerasus microcarpa, Rhamnus arvensis and Astragalus licyoides. Our results indicated that these groups were significantly affected by elevation and soil properties and the soil properties refer to: EC (electrical conductivity), N (nitrogen), K (potassium), OM (organic matter), and bulk density. Woody regeneration was composed of Q. brantii, A. monspessulanum, C. microcarpa, Amygdalus scoparia and Crataegus pontica seedlings. The highest density of seedlings was found for Q. brantii (97.14 (±48.00)/hm2) and the lowest for A. scoparia (2.28 (±1.50)/hm2). Quercus brantii was the dominant species and the seedling density was positively correlated with soil pH and P (phosphorus) values. A. scoparia regeneration was negatively correlated with elevation, and the seedling density peaked in C. microcarpa group. There was no significant variation in distribution of C. pontica seedlings among the groups, but the seedling density of this species was positively correlated with slope and potassium. Cerasus microcarpa seedlings were more abundant in the Q. brantii group than in other groups. This study showed that the regeneration of trees and shrub species was unequally distributed in different floristic groups for some species (A. scoparia and C. microcarpa) but not for other (Q. brantii and C. pontica) and was generally correlated with some environmental factors, particularly elevation, slope and soil nutrients (P and K). These results are a first step to implement future management and restoration strategies for promoting forest regeneration.