Questions Studies in alpine environments have highlighted dominant facilitative effects of cushion species for diversity. Many fewer studies have assessed the effects of cushion shrubs in drier and less cold mountain habitats and how these effects vary within the shrub canopy and with increasing drought stress with exposure and disturbance by fire. Location A mountain steppe of the Golestan National Park in the semi‐arid climate of northeast Iran. Methods We quantified the cover of understorey species at the center and edge of the canopy of the thorny cushion shrub Onobrychis cornuta and in paired open areas, in north and south exposures and unburned and burned plots, two years after a fire. Soil chemical variables and moisture were measured in treatments. We quantified cushions’ center and edge effects on understorey species cover with the Relative Interaction Index (RII). Treatment effects on RII cover were analyzed with generalized linear mixed‐effects models (GLMM) and species composition with Correspondence Analysis. Results In unburned plots, competition for cover was very high at cushion centers and strongly decreased at cushion edges. Exposure stress did not affect competition in unburned plots. Fire slightly decreased competition but only at center cushion shrub position and at south exposure. Species composition was mostly affected by exposure and fire treatments, although there was subtle variation in species composition among patches due to fire. Cushion position, fire and exposure treatments affected soil moisture and chemical variables, but soil changes did not explain cushion effects on understorey cover and their variations along treatments. Conclusions The effects of the cushion shrub were highly negative and in particular at cushion center, likely due to interference and competition for light in a dry and not too cold environment. Fire only slightly decreased competition in the short term, likely due to the delayed response of understorey species to disturbance induced by fire.

Questions (a) Are there contrasting short‐ and long‐term effects of shrubs on understorey species that may balance at the community level? (b) Are these effects due to the responses of different understorey species groups? (c) Are these effects contributing to community composition? Location An Artemisia steppe community in Golestan National Park (North‐East Iran). Methods We used both the removal and observational methods to separate short‐ from long‐term effects of Artemisia sieberi Besser on the cover of 18 subordinate species and the relative interaction index (RII) to calculate species’ responses. We conducted a principal components analysis (PCA) on species RII and grouped species with a cluster analysis depending on their responses to the shrubs. Another PCA on understorey composition plots was conducted to assess the consequences of shrub effects for community composition. Results Artemisia had negative short‐term and positive long‐term community‐level effects that balanced each other to produce neutral net effects. Short‐ and long‐term effects were due to the responses of different species groups, dominantly annuals and perennial grasses, respectively. Although balancing at the community level, the different shrub effects significantly contributed to explain community composition. Conclusions Our study highlights that a dominant foundation shrub may have contrasting short‐ and long‐term effects on different groups of understorey species that contribute to explain community composition, although balancing at the community level.

No abstract

No abstract

No abstract