Rainfall is the main resource of soil moisture in the semiarid areas, and the altered rainfall pattern would greatly affect plant growth and development. Root morphological traits are critical for plant adaptation to changeable soil moisture. This study aimed to clarify how root morphological traits of Bothriochloa ischaemum (a C4 herbaceous species) and Lespedeza davurica (a C3 leguminous species) in response to variable soil moisture in their mixtures. The two species were co-cultivated in pots at seven mixture ratios under three soil water regimes [80% (HW), 60% (MW), and 40% (LW) of soil moisture field capacity (FC)]. At the jointing, flowering, and filling stages of B. ischaemum, the LW and MW treatments were rewatered to MW or HW, respectively. At the end of growth season, root morphological traits of two species were evaluated. Results showed that the root morphological response of B. ischaemum was more sensitive than that of L. davurica under rewatering. The total root length (TRL) and root surface area (RSA) of both species increased as their mixture ratio decreased, which suggested that mixed plantation of the two species would be beneficial for their own root growth. Among all treatments, the increase of root biomass (RB), TRL, and RSA reached the highest levels when soil water content increased from 40 to 80% FC at the jointing stage. Our results implied that species-specific response in root morphological traits to alternated rainfall pattern would greatly affect community structure, and large rainfall occurring at early growth stages would greatly increase their root growth in the semiarid environments.

Aims Nitrogen (N) deposition is a global environmental problem that can alter community compositions and functions, and consequently, the ecosystem services. In this study, we assessed the responses of aboveground vegetation, surface soil properties and microbial communities to N addition, and explored the drivers of microbial community in a semiarid steppe ecosystem in northwest of China. Methods Thirty-six 6×10-m2 plots composed of six N addition levels and six replicates were distributed in six columns and six rows. Nine vegetation characteristics and seven soil properties were measured and calculated. Soil microbial characteristics were analyzed by 16S rRNA high-throughput sequencing. Results N addition positively affected aboveground vegetation traits such as the community weighted-mean of leaf nitrogen content (LNCWM). High N inputs significantly altered the microbial community assembly process from random to deterministic. The microbial community diversity and composition, however, were not sensitive to N addition. A piecewise structural equation model (SEM) further showed that the microbial community composition was affected by both aboveground vegetation and soil properties. The composition of bacterial communities was mainly regulated by the composition of plant communities and soil total N. In contrast, the composition of fungal communities was driven by soil pH and the community weighted-mean of specific leaf area (SLACWM). Microbial diversity and composition remained unchanged because their drivers were not affected by N addition. The results of this research improved our understanding of the response of grassland ecosystems to N deposition, and provided a theoretical basis for grassland utilization and management under N deposition.