Seasonal shifts in water-source partitioning and fine-root overlap in a Populus simonii-maize shelterbelt system

Farmland shelterbelts are critical for agricultural production in arid and semiarid regions, but they may compete with crops for limited soil water. This study aimed to quantify seasonal water-source use, soil moisture dynamics, and fine-root distribution in a Populus simonii-maize shelterbelt system.

The study was conducted in Zhangbei County, northwestern Hebei Province, China. Three land-use conditions were compared: a pure P. simonii shelterbelt (P), a P. simonii-maize intercropping system (PC), and a maize monoculture reference (C). Synchronous sampling was conducted in May, July, August, and October 2023. Stable hydrogen and oxygen isotope ratios of soil water, groundwater, and plant xylem water were measured. The MixSIAR Bayesian mixing model was used to estimate water-source contributions. Fine-root length density was measured to 160 cm depth and at four horizontal distances from tree trunks. Schoener's proportional similarity index, Pianka's niche-overlap index, and Levins' niche-breadth index were calculated.

P. simonii relied strongly on deep soil water and a groundwater-like endmember, with a combined contribution exceeding 55% during the dry season. Maize relied more on shallow soil water, with >60% of its water uptake from the 0-40 cm layer during the rainy season; during the dry season, its estimated contribution from 40-100 cm increased. Water-source overlap was higher in the dry season than in the rainy season. The strongest fine-root co-occurrence occurred in the 40-80 cm soil layer within 0-100 cm of the tree trunk.

P. simonii exhibited a broader and more stable water-source niche than maize. Maize showed a larger seasonal shift in water uptake, especially under intercropping, suggesting an asymmetric water-use response. These results identify the specific soil layers and periods where tree-crop water interactions are likely to occur. However, because the study was conducted in a single year and at one site, the findings should be treated as testable hypotheses for future multi-year, replicated irrigation and spacing experiments.