Root Microhabitats Shift How Water and Microbes Support Desert Soil Multifunctionality

In a growing-season pot experiment on Eremopyrum distans, an annual desert grass native to northern Xinjiang’s Gurbantünggüt Desert, an investigative team from Xinjiang Agricultural University discovered that soil tightly adhering to root systems channels moisture effects indirectly through structured microbial networks. Conversely, soil located away from direct root contact relies predominantly on direct physicochemical drivers, particularly soil moisture and pH, to sustain simultaneous ecological processes. These findings demonstrate that root microhabitats reconfigure how belowground communities translate water pulses into functional outcomes.

Arid desert environments are characterized by scarce precipitation, where ecological activity is governed by intermittent moisture pulses rather than steady rainfall. In these moisture-limited ecosystems, water inputs determine water-film continuity, control solute diffusion rates, and trigger microorganisms to transition from dormancy to active metabolism. However, biogeochemical cycles do not necessarily recover in synchrony following rainfall events, because organic matter decomposition, nutrient mineralization, and microbial immobilization rely on distinct enzymatic systems, substrate sources, and stoichiometric constraints. Assessing individual soil processes in isolation often obscures how ecosystems respond to shifting hydrologic conditions. To capture these coordinated dynamics, soil multifunctionality integrates multiple nutrient stocks, microbial biomass metrics, and extracellular enzyme activities into a unified assessment framework. Furthermore, although desert research has historically emphasized bacteria and fungi, microscopic soil protists contribute substantially to nutrient turnover through predatory grazing and biotic interactions.

This specialized compartment contrasts with bulk soil, which experiences weaker root influence and lower inputs of readily accessible plant carbon. In extreme desert landscapes, the rhizosheath acts as a critical biological buffer, retaining moisture, stabilizing microclimates, and fostering dense communities of microorganisms in close proximity to active roots. In contrast, bulk soil located beyond immediate root contact experiences greater exposure to evaporative drying and fluctuating physical conditions. Under identical precipitation inputs, these two microhabitats establish markedly divergent microscale resource environments. Determining whether belowground communities and ecosystem functions respond differently within these separate soil compartments represents an essential step toward uncovering how vegetation alters dryland functional stability.

To test how changing moisture alters functional dynamics across these microhabitats, the research team established a controlled outdoor pot experiment spanning an entire growing season from October 2024 to May 2025. Soil was gathered from the upper thirty centimeters of sandy desert terrain at five representative field locations within the Gurbantünggüt Desert. The collected soil was combined into twenty-five field subsamples, passed through a two-millimeter sieve to remove debris, thoroughly homogenized, and air-dried before experimental setup. The investigators established five watering levels: a fifty percent reduction, a twenty-five percent reduction, an ambient control, a twenty-five percent addition, and a fifty percent addition. Each treatment consisted of twenty planted pots containing ten kilograms of desert soil and equipped with base drainage holes. Seeds of Eremopyrum distans were surface-sterilized and sown into the pots, with seedlings thinned to five uniform plants per pot following establishment.

Precipitation manipulation required specialized infrastructure to deliver natural hydrological patterns rather than artificial watering schedules. To simulate drought, researchers erected five independent transparent rainout shelters for each water-reduction treatment, with each shelter covering four planted pots alongside an unplanted correction pot. These shelters featured galvanized steel frames and V-shaped polycarbonate roofs providing eighty-five to ninety percent light transmission while diverting intercepted precipitation away. Unplanted correction pots were monitored after every rainfall and snow event to track moisture gains from edge splashing or snow drift. Ambient precipitation was collected on-site in clean vessels and redistributed within two hours of precipitation events to provide replacement water for drought treatments and supplemental water for addition treatments. Over the course of the experiment, cumulative water additions totaled 1018.8, 1528.2, 2037.6, 2547.0, and 3056.4 milliliters per pot across the five watering treatments.

Destructive soil sampling occurred in May 2025 when Eremopyrum distans reached seed-setting stage and possessed fully developed rhizosheaths. Intact root systems were harvested and gently shaken to dislodge loosely adhering soil, leaving only the firmly attached rhizosheath intact along the recovered roots. Bulk soil was collected simultaneously from the pot perimeter at a depth of zero to ten centimeters in zones lacking visible roots. For each watering treatment, pots were pooled into five composite replicates, yielding fifty distinct paired soil samples. To separate rhizosheath soil for molecular sequencing, root segments were vortexed in phosphate-buffered saline solution supplemented with Silwet L-77 surfactant and centrifuged, while samples for physicochemical and enzymatic assays were processed directly without surfactant. High-throughput amplicon sequencing on an Illumina NovaSeq platform targeted bacterial 16S ribosomal RNA genes, fungal internal transcribed spacer regions, and eukaryotic 18S ribosomal RNA genes to profile bacterial, fungal, and protistan communities.

Ecosystem performance was evaluated through sixteen functional indicators categorized into three distinct dimensions of soil functioning. The carbon dimension included soil organic carbon, microbial biomass carbon, and potential activities of four extracellular enzymes: cellobiohydrolase, xylanase, beta-glucosidase, and alpha-glucosidase. The nitrogen dimension comprised total nitrogen, available nitrogen, microbial biomass nitrogen, leucine aminopeptidase, urease, and beta-N-acetylglucosaminidase. The phosphorus dimension encompassed total phosphorus, available phosphorus, microbial biomass phosphorus, and alkaline phosphatase. Beyond individual dimensions, the authors calculated an integrated dimension-weighted multifunctionality index alongside threshold-based multifunctionality, which counted the number of indicators achieving twenty-five, fifty, or seventy-five percent of observed maximums. Microbial community attributes were simultaneously quantified, using Spearman correlations and single-sample LIONESS algorithms to determine topological network complexity and an integrated community-stability proxy combining resistance to disruption, average variation degree, and the balance of negative-to-positive co-occurrence links.

Analyzing functional outcomes across the experimental gradient revealed striking differences between evaluation frameworks and soil microhabitats. However, threshold-based assessments revealed a distinct pattern: when evaluating the highest functional threshold of seventy-five percent, the moderate twenty-five percent water addition consistently supported the largest number of simultaneous high-performing functions, particularly within bulk soil. Under higher water inputs of fifty percent addition, individual functions displayed divergent trajectories, preventing further gains in high-threshold multifunctionality. These patterns show that dimension-weighted metrics allow strong performance in one nutrient cycle to offset weaker performance in another, whereas high-threshold metrics reveal functional bottlenecks.

Statistical modeling using linear mixed-effects models and random forest machine learning demonstrated that distinct biological attributes govern specific functional dimensions. Carbon and nitrogen functioning were linked primarily to taxonomic diversity across bacteria, fungi, and protists, reflecting the diverse metabolic repertoires required to depolymerize varied organic substrates and regulate nitrogen transformations. Conversely, phosphorus-related functioning and high-threshold multifunctionality were associated far more strongly with microbial network complexity and community stability than with taxon richness alone. Because phosphorus remains relatively immobile in desert soils, mineral solubilization and phosphatase activities rely heavily on localized, highly interconnected microbial consortia. Notably, dominant compositional shifts captured by the primary axis of principal coordinates analysis exhibited minimal explanatory power for soil functioning.

Piecewise structural equation modeling uncovered contrasting association networks linking water inputs, environmental properties, and microbial attributes to multifunctionality in the two microhabitats. In this root-associated zone, the direct statistical path between soil moisture and multifunctionality was negligible. In sharp contrast, bulk soil models demonstrated that local environmental parameters maintained strong direct associations with functioning. Soil moisture exhibited a prominent positive direct link to bulk soil multifunctionality, while soil pH exerted a significant negative direct association. Formal paired-bootstrap comparisons confirmed that the direct path connecting soil moisture to multifunctionality differed significantly between the microhabitats, underscoring that plant roots shift dryland functional regulation from direct abiotic control toward microbially mediated networks.

The authors emphasized that several methodological constraints must be considered when interpreting these experimental findings. The investigation relied on sieved, air-dried, and repacked desert soil within a controlled pot setting over a single growing season, which does not fully capture the intact structural complexity or multi-year climatic variations of native desert landscapes. Additionally, rainout shelters inevitably introduced minor microclimatic differences in wind and temperature compared to unsheltered open pots, and the physical replication design nested drought pots under collective shelters. Because the study focused exclusively on Eremopyrum distans and did not directly measure root exudation rates or plant biomass allocation, unmeasured vegetative traits may have contributed to observed rhizosheath associations.

Subject of Research: Microbial and environmental drivers of soil multifunctionality across desert grass microhabitats

Article Title: Contrasting microbial and environmental associations link water availability to soil multifunctionality in the rhizosheath and bulk soil of Eremopyrum distans

Article References: Jiang, K., Wang, Y., Ma, S., Heng, Z., Zhang, Z., Lv, C., Zhang, Q., Zhang, L., & Li, P. (2026). Contrasting microbial and environmental associations link water availability to soil multifunctionality in the rhizosheath and bulk soil of Eremopyrum distans. Frontiers in Microbiology, 17, Article 2000297. https://doi.org/10.3389/fmicb.2026.2000297

Image Credits: AI Generated

DOI: 10.3389/fmicb.2026.2000297

Keywords: Eremopyrum distans, rhizosheath, bulk soil, soil multifunctionality, Gurbantünggüt Desert, microbial network complexity, community-stability proxy, piecewise structural equation modeling, soil moisture, soil pH, bacterial diversity, fungal diversity, protistan community, phosphorus cycling, nitrogen mineralization, carbon turnover, rainout shelters