2.2 Experimental design
Our study was conducted under semi-controlled field conditions over 4 years in the arid valley of the Minjiang River. The experiment used a randomized block design with eight treatments as combinations of four plant species and four RFC (0%, 25%, 50%, and 75% volumetric content, v v-1). Each treatment had three replicates (plots), making a total of 48 plots (4 RFC × 4 species × 3 replicates) in the study. The four xerophytic species evaluated wereArtemisia vestita , Bauhinia brachycarpa , Sophora davidii , and Cotinus szechuanensis , which are native to the arid valley. They were chosen because of their ecological importance and divergent root performance: A. vestita , a fast-growing subshrub with shallow root distribution; B. brachycarpa , a non-nitrogen-fixing legume shrub with a thin and high branching root system (root branching density = 1.47); S. davidii , a nitrogen-fixing legume shrub with thick and low branching root system (root branching density = 1.07); and C. szechuanensis , a tall shrub with developed root hairs and mycorrhiza. For microhabitat heterogeneity, differences in plant adaptations are recognized based on aboveground measurements (Li et al., 2008; Wu et al., 2008; Hu et al., 2021), whereas the variation in rooting profile and its ecological indicator is not well known.
Each plot was represented by a pit, with dimensions of 1 m × 1 m × 0.5 m (length × width × depth) and 50 cm spacing between plots. The soil at a depth of 0–50 cm was excavated from each pit. Subsequently, the walls of each pit were lined with polyethylene film to prevent interference from external conditions, and the bottom of the plot was left unlined to allow natural drainage. Fine soil particles (< 2 mm in diameter) and rock fragments (10–20 mm in diameter) were collected, mixed uniformly, and filled back into each pit to obtain the desired RFC.
Soil from all pits was then air dried for a week, and fine soil particles were collected by passing the dried soil through a 2 mm sieve and all the soil was then mixed uniformly. Initial soil properties in the samples (n = 6) were as follows: total carbon, 15.3 ± 0.09 g kg-1; total nitrogen, 2.31 ± 0.02 g kg-1; and total phosphorus, 0.61 ± 0.01 g kg-1. Thin-bedded limestone (dominated by phyllite), which is commonly found in regional soil, was used for the rock fragments in this study. A sufficient amount of rock fragments was collected from the coarse soil part (≥ 2 mm) that had not passed through the sieve and from nearby land. The crushed rock materials were first passed through a 10 mm sieve and then a 20 mm sieve, leaving rock fragments with a particle size of 10–20 mm for use. The density of thin-bedded limestone was 2.56 ± 0.03 g cm-3 (n = 12), as measured by the water displacement method (Wang et al., 2017). After uniform mixing, fine soil particles and rock fragments were filled back into each pit at the desired RFC in April 2018. Each plot was irrigated with 100 L of water and left for soil to stabilize.
Sowing in each pit was performed at a depth of 0.5–1 cm with an equidistant interval pattern of nine points (25 cm equidistance between two points) in April 2018 (Hu et al., 2021). Seeds of the four species were collected from their natural habitats in the arid valley of the Minjiang River (31°42′N, 103°53′E, altitude range of 1600–1920 m) in August to October 2017, air-dried for 4–8 days, and stored at room temperature (10–25 °C). All the seeds were disinfected by immersion in 2.5% NaClO for 1 h and then sown in each plot. Seedlings were watered weekly after sprouting to prevent early losses. They were thinned 2 months after sprouting (Hu et al., 2021), leaving four average-sized seedlings per plot and an interval of approximately 50 cm between seedlings. The plots were weeded twice a week to ensure normal plant growth.