Relationship between broadleaved mixed forest understory species groups with soil and elevation in a semi-arid Persian oak (Quercus brantii L.) ecosystem

Abstract

This research aimed to investigate the relationship between plant composition of Dinarkooh Protected Area (DPA), west of Iran (called Zagros forests) with elevation and soil properties. A total of 100 random sampling plots (25 × 25 m) at different elevations were sampled for their soil and collection of plant specimens and vegetation data based on Braun-Blanquet covered-abundance scale. Data were analyzed using TWINSPAN and multivariate statistical analysis packages. We recognized 142 plant species present in the study area identified under 105 genera and 29 families. Asteraceae, Papilionaceae and Poaceae were the most frequently encountered families. Bromus and Astragalus were also the largest genera in the study area. The plant species were grouped into three ecological groups based on elevation and soil properties: Group 1 with 20 species including Scorzonera sp. and Cerastium inflatum L. at lower elevation (1300-1500 m a.s.l.) and with clay-loam soil; Group 2 with 96 species, including Onosma trachytrichum Boiss., Eryngium noeanum Boiss. and Acer monspessulanum L. occurred at middle elevations with alkaline loam-clay soil; and Group 3 with 26 species, including Bromus sterilis L., Minuartia meyeri, Minuartia meyeri (Boiss.) Bornm., Lithospermum sp., Scariola orientalis (Boiss.) Sojak and Trifolium tomentosum L. at higher elevation with sandy-loam soil. Prevalent tree species included Oak (Quercus brantii Lindl. var. persica (Jaub. & Spach)) in Group 2 and Acer monspessulanum L. in Group 3. Low homogeneity of plant compositions at different elevations reflects the differences in habitat properties. Results showed that the elevation and soil properties played primary and complementary roles in vegetation spatial composition. Furthermore, diversity and richness of plant species was higher at middle elevations. More detailed investigation into biotic variables at the root level would complement current data to analyze this ecosystem.