Increasing soil age drives shifts in plant-plant interactions from positive to negative and affects primary succession dynamics in a subalpine glacier forefield

Abstract

The stress gradient hypothesis predicts that plant-plant interactions switch between facilitation (positive) and competition (negative) along environmental gradients, with facilitation being more common under high abiotic stress conditions relative to more moderate abiotic stress conditions. Our aim was to reveal, whether the interactions between Populus purdomii Rehder and Salix rehderiana Schneider switch from positive to negative during the early stages of primary succession in the Gongga Mountain glacier retreat region. We also investigated, whether soil age is a major driving factor for the transformation of interactions between neighboring plants. We analyzed differences between intraspecific interactions and interspecific interactions of Populus and Salix under 20- and 40-year-old soil conditions, including plant biomass accumulation and allocation, nutrient absorption and utilization, relative competition intensity, non-structural carbohydrates, foliar carbon and nitrogen isotope composition, mesophyll cell ultrastructure, soil microbial biomass and community structure, extracellular enzyme activities, and soil organic carbon (SOC), soil total nitrogen (TN), soil ammonium (NH4+-N), and soil nitrate (NO3−-N) contents. We found that P. purdomii and S. rehderiana growing under interspecific interactions had greater contents of aboveground dry matter, belowground dry matter and total dry matter compared to intraspecific interactions in 20-year-old soil. Furthermore, in 40-year-old soil conditions, the phospholipid fatty acid (PLFA) analysis showed that Populus and Salix exposed to interspecific interactions exhibited lower amounts of gram-positive bacteria, fungi (18,1 ω9c) and actinomycetes, and lower levels of total PLFAs than those growing under intraspecific interactions. The redundancy analysis (RDA) results demonstrated that soil N was the most important parameter contributing to the composition of microbial communities. In addition, the 15N stable isotope labeling method showed that Populus and Salix growing under interspecific interactions had higher foliage δ15N derived from NO3− (δ15N-NO3−) than those growing under intraspecific interactions in 20-year-old soil. In summary, our results demonstrated that Populus-Salix interactions exhibited positive effects on survival in 20-year-old soil. Conversely, under 40-year-old soil conditions, Populus-Salix interactions presented negative effects in relation to nutrients and elimination by neighboring plants. Moreover, soil age is a major driving factor for plant-plant interactions that shift from positive to negative with an increasing soil age in the Gongga Mountain glacier forefield. In all, our results support the stress gradient hypothesis. Our findings improve understanding of plant-plant interactions and plant-soil feedbacks during the early stages of soil development, and of the construction of vegetation communities.