Solar Panels Alter Desert Soil Microbial Networks Across Depth Profiles
Installing solar photovoltaic panels across arid desert landscapes alters soil microbial composition, functional profiles, and ecological assembly down through the soil profile, with bacterial communities showing significantly greater sensitivity than fungi, according to an extensive field investigation conducted at a utility-scale solar facility in Xinjiang, China. Researchers evaluated the belowground ecological consequences of large-scale renewable energy infrastructure by analyzing soil bacterial and fungal communities across five distinct vertical soil layers spanning 0 to 100 centimeters. The investigation compared three specific microhabitats created by the solar arrays, contrasting the open environment of bare land against the shaded microhabitat directly beneath the solar panels and the intermediate microclimate found between adjacent panel rows. The resulting empirical observations demonstrate that photovoltaic panel installation substantially modifies belowground biological communities, but these environmental alterations are heavily depth-dependent and attenuate progressively toward deeper soil layers.
Microorganisms residing in desert soils act as primary drivers of terrestrial biogeochemical cycling, organic matter transformation, and soil structural stability within ecologically fragile hyper-arid environments. The rapid global deployment of large-scale photovoltaic power stations across arid regions has introduced novel environmental pressures that alter natural surface regimes. Photovoltaic panel arrays intercept direct solar radiation, reducing photosynthetically active radiation by an estimated 60 to 80 percent beneath the modules, while concurrently altering surface temperature patterns. Furthermore, the physical panels intercept natural rainfall and redirect runoff toward their drip lines, creating localized water redistribution and suppressing evapotranspiration. To determine how these localized microclimatic alterations affect microbial ecology vertically through the soil column, the research team established sampling transects at the 34th Regiment Photovoltaic Power Station in Tie-men-guan, located within the Taklimakan desert region characterized by aeolian sandy soil, an average annual precipitation of 43 millimeters, and annual evaporation reaching 2,700 millimeters.
The solar facility under study utilizes fixed-tilt monocrystalline silicon panels oriented due south at an inclination of approximately 30 degrees, featuring a mounting structure 4 meters high at the rear edge, 1.6 meters at the front edge, a horizontal panel width of 4 meters, and an inter-panel spacing of 8.3 meters. Having achieved operational status in 2023, the facility had been functioning for approximately three years when soil cores were gathered in May 2026. Systematic grid sampling collected composite soil cores across five stratified depth intervals: 0 to 20 centimeters, 20 to 40 centimeters, 40 to 60 centimeters, 60 to 80 centimeters, and 80 to 100 centimeters. Soil augers with an inner diameter of 5 centimeters were utilized to collect core samples from the central zones beneath panels, the equidistant midpoint between panel rows, and three unshaded bare land reference transects surrounding the facility. A total of 45 biological composite samples were generated across the three microhabitats and five depth layers, ensuring rigorous spatial representation while avoiding panel edge effects.
Laboratory analysis paired soil physicochemical measurements with high-throughput amplicon sequencing of soil genomic DNA extracted using magnetic bead kits. The researchers targeted the V4 hypervariable region of the bacterial 16S rRNA gene utilizing 515F and 806R primers, alongside the fungal internal transcribed spacer region amplified via ITS1F and ITS2 primers. Paired-end sequencing was executed on an Illumina NovaSeq 6000 platform, generating raw reads that were subsequently merged and quality-filtered before being clustered into operational taxonomic units at a 97 percent sequence identity threshold against SILVA reference databases. Potential bacterial metabolic functions were predicted in silico through the FAPROTAX database mapped to the Kyoto Encyclopedia of Genes and Genomes, while fungal ecological guilds and trophic classifications were evaluated using the FUNGuild analytical pipeline. Ecological assembly mechanisms were quantified using normalized stochasticity ratios, beta nearest taxon indices, and Bray-Curtis-based Raup-Crick null model distributions.
Taxonomic profiling revealed pronounced shifts in bacterial community composition between the photovoltaic zones and open bare land, with divergence being most conspicuous in surface soils. The relative abundance of bacterial Pseudomonadota was significantly higher in photovoltaic microhabitats than in bare ground, a divergence that expanded with increasing depth. In contrast, the relative abundances of Bacillota and Bacteroidota were markedly lower in bare land compared to photovoltaic zones, though this distinction diminished at greater depths. Actinomycetota exhibited elevated relative abundance in photovoltaic surface soils, but their enrichment attenuated rapidly down the vertical profile, eventually falling below bare land levels in the 80 to 100 centimeter layer. Fungi demonstrated a substantially more restricted response to panel installation: Ascomycota relative abundance surpassed bare land values exclusively in the 0 to 20 centimeter surface layer, while Basidiomycota displayed the opposite distribution pattern in that same upper layer.
Predictive functional profiling showed that bacterial metabolic capabilities were substantially reorganized under the photovoltaic arrays, with noticeable depth limitations. Microorganisms associated with methanol oxidation, methylotrophy, and nitrogen fixation were significantly enriched within the photovoltaic microhabitats, consistent with elevated soil water content and total nitrogen availability under the panels. Conversely, bacterial pathways linked to nitrate reduction and hydrocarbon degradation were significantly more prevalent in the undisturbed bare land. Functional predictions also highlighted significant enrichment of bacterial aerobic chemoheterotrophy and aromatic compound degradation beneath the arrays. All observed functional differences between photovoltaic microhabitats and bare land weakened steadily as soil depth increased down the 100-centimeter profile. Meanwhile, fungal trophic modes—including pathotrophic, symbiotrophic, and saprotrophic categories—exhibited no statistically significant differences among microhabitats across any of the surveyed soil horizons.
Measurements of alpha diversity revealed that bacterial and fungal richness, Shannon diversity, Simpson index, and Pielou evenness values were numerically lower beneath and between photovoltaic panels compared to bare ground; however, non-parametric Kruskal-Wallis testing confirmed these alpha-diversity differences were not statistically significant across any soil layer. In contrast, beta diversity and community structure differed significantly between the photovoltaic zones and bare land. Permutational multivariate analysis of variance demonstrated that bacterial community structure differed significantly between the open bare land and photovoltaic microhabitats across all five depth layers, with the strength of separation declining from an R-squared value of 0.626 at the 0 to 20 centimeter depth to 0.479 at the 80 to 100 centimeter depth. Fungal community structure, however, separated significantly between treatments only in the shallow 0 to 20 centimeter and 20 to 40 centimeter layers, showing no significant divergence in the deeper three layers.
Community assembly modeling revealed fundamental divergences in the ecological mechanisms governing bacterial and fungal responses to solar panel infrastructure. For soil bacteria, community assembly shifted from deterministic homogeneous selection in bare land toward predominantly stochastic assembly processes under the solar panels, accompanied by a marked increase in dispersal limitation. This assembly shift indicates that panels weaken environmental filtering while their physical presence restricts bacterial dispersal, though these assembly differences diminished in deeper soil layers. Fungal community assembly, in sharp contrast, remained overwhelmingly dominated by stochastic processes across all three microhabitats and depth layers, with undominated ecological mechanisms accounting for more than 80 percent of assembly variation. The greater structural complexity and environmental tolerance of fungal hyphae likely buffered fungal communities against panel-induced microclimatic variations.
Microbial co-occurrence networks constructed via Spearman rank correlations indicated that bacterial network scale and topological complexity decreased sharply with soil depth, with surface horizons exhibiting the largest networks and deep horizons the simplest. Bare land supported the largest, most interconnected bacterial network at the surface, comprising 124 nodes and 3,082 edges, whereas networks beneath panels were simplified to 113 nodes and 2,039 edges. Bacterial networks in photovoltaic zones exhibited significantly higher modularity but demonstrated greater topological vulnerability during simulated node and edge removal analyses. Fungal networks were considerably smaller overall and exhibited minimal structural disruption from panel installation, although surface fungal networks in photovoltaic zones displayed an increased proportion of positive correlations. The resilience of fungal networks suggests they may provide critical buffering that preserves functional stability in fragile desert soils during the early phases of photovoltaic power station operation.
The study authors emphasized several key interpretive caveats, noting that the three-year operational history of the station captures early-stage microbial responses that may reflect temporal lag effects rather than long-term ecological equilibria. Moderate shifts were recorded in soil moisture and total nitrogen, whereas soil organic carbon, pH, and electrical conductivity remained largely unchanged, potentially explaining the absence of significant alpha-diversity shifts. Furthermore, functional attributions derived from FAPROTAX 16S predictions reflect genetic potential rather than direct in situ metabolic rates or transcriptomic activity. The researchers concluded that assessing the environmental footprints of renewable energy infrastructure requires comprehensive vertical profiling across multiple microhabitats, and they called for long-term chronosequence studies to confirm how belowground desert ecosystems evolve as solar facilities age.
Subject of Research: Soil microbial community responses to solar photovoltaic installations in hyper-arid deserts
Article Title: Depth-dependent responses of soil microbial communities and co-occurrence networks to solar photovoltaic panel installation along a soil profile in a hyper-arid desert
Article References: Wang, T., Zhao, D., Li, M., Lin, Z., Song, Y., Fan, J., Li, L., & Zhang, H. (2026). Depth-dependent responses of soil microbial communities and co-occurrence networks to solar photovoltaic panel installation along a soil profile in a hyper-arid desert. Frontiers in Microbiology, 17, Article 2008754. https://doi.org/10.3389/fmicb.2026.2008754
Image Credits: AI Generated
DOI: 10.3389/fmicb.2026.2008754
Keywords: photovoltaic panels, soil microbiome, hyper-arid desert, bacterial community structure, fungal ecology, co-occurrence networks, soil depth profile, community assembly, dispersal limitation, homogeneous selection, amplicon sequencing, FAPROTAX, microbial modularity, network vulnerability, aeolian sandy soil, Taklimakan Desert, soil water content
News Source: Morgan Morrow. (November 13, 2026). Solar Panels Alter Desert Soil Microbial Networks Across Depth Profiles. Scienmag.
We bring you the latest biotechnology news from best research centers and universities around the world. Check our website.
Enter your email address to subscribe to this blog and receive notifications of new posts by email.
Email Address
Subscribe
Bioengineer.org © Copyright 2023 All Rights Reserved.
Login to your account below


Remember Me

Please enter your username or email address to reset your password.


Bioengineer.org © Copyright 2023 All Rights Reserved.

source