The research, published in the journal Plant and Soil, was led by Zhijie Dong of Ningbo University together with colleagues at institutions spanning several Chinese provinces. The team assembled 213 soil samples from legume-based agroecosystems distributed across China’s major climatic zones, from the frigid Qinghai-Tibetan Plateau to humid subtropical regions and the warm-temperate plains of the east. For each sample, they quantified eleven distinct soil functions, including processes central to nutrient cycling, organic matter decomposition, and microbial activity. The goal was to compute a single integrative metric, known as soil multifunctionality, that captures how many services a soil delivers simultaneously rather than judging it on yield alone.

To disentangle the drivers of multifunctionality, the researchers combined two complementary statistical approaches. Random forest modeling, a machine learning technique that ranks predictors by their contribution to explaining variation, was used to identify the most influential environmental variables. Structural equation modeling was then applied to trace the pathways through which those variables act, distinguishing direct effects from indirect ones mediated by other soil properties. This dual strategy matters because correlations alone cannot reveal whether, for example, climate influences soil functions directly through temperature effects on microbes, or indirectly by shaping the pools of carbon and nitrogen that microbes depend on.

The results were striking in their hierarchy. Among the predictors retained in the best-supported model, soil properties accounted for 54.8 percent of the relative predictor support, followed by climate at 31.0 percent and microbial diversity at just 14.2 percent. In other words, the abiotic stage on which soil microbes perform mattered far more than the diversity of the performers themselves. Soil pH emerged as the single strongest abiotic predictor of multifunctionality, and notably its influence was negative: as pH rose, multifunctionality declined. This finding aligns with a growing body of evidence that alkaline conditions constrain the solubility of essential nutrients and alter the kinetics of organic matter decomposition.

One of the most intriguing results concerns micronutrients. The study found that soil multifunctionality exhibited nonlinear, threshold-dependent associations with the availability of metallic trace elements. Rather than rising or falling smoothly with micronutrient supply, multifunctionality appeared to respond within specific windows: below certain thresholds, functions were constrained, while beyond them additional micronutrients brought diminishing returns. This makes mechanistic sense. Metals such as iron, zinc, manganese, and copper serve as cofactors for a wide array of extracellular enzymes that microbes use to break down organic matter, and they are also essential for the symbiotic nitrogen fixation process that defines legume biology. Yet the same metals become toxic at high concentrations, producing a genuine optimum rather than a simple dose-response relationship.

The stoichiometry of soil resources told a similar story of thresholds. Multifunctionality showed nonlinear associations with the soil nitrogen-to-phosphorus ratio, suggesting that legume-based systems perform best within particular stoichiometric windows. Microbial communities, like all living things, maintain a relatively fixed elemental composition in their biomass. When the ratio of available nitrogen to phosphorus in the soil drifts too far from microbial demands, one nutrient or the other becomes limiting, forcing microbes to invest energy in acquiring the scarce element at the expense of other functions. The study’s findings imply that managing these ratios, rather than maximizing any single nutrient, may be the key to sustaining multiple soil services simultaneously.

Structural equation modeling added a crucial nuance about climate. Mean annual temperature exerted a negative direct pathway on multifunctionality, consistent with the idea that warmer conditions accelerate the enzymatic breakdown of organic matter and can deplete the carbon reserves that sustain soil food webs. Yet this direct harm was partially counterbalanced by indirect pathways operating through soil resource pools, presumably because warmer regions also accumulate or retain different quantities of carbon and nitrogen depending on vegetation productivity and management. The net effect of warming on soil multifunctionality in these farmlands therefore depends on whether resource gains can keep pace with the temperature-driven acceleration of decomposition, a balance that will likely shift as the climate continues to change.

The regional patterns revealed by the survey were unexpected in several respects. The Qinghai-Tibetan Plateau, despite its harsh cold climate, exhibited relatively high multifunctionality, with a median value of 0.12, a result the authors attribute to cold conditions that slow decomposition and allow relatively carbon-rich soils to accumulate. Subtropical soils also showed elevated multifunctionality, in this case associated with abundant micronutrient availability. The warm-temperate zone, by contrast, lagged behind, with lower multifunctionality linked to high soil pH and carbon limitation. This inversion of expectations, in which the frigid plateau outperforms the temperate heartland, underscores how strongly edaphic constraints can override climatic intuition in managed systems.

For the authors, the practical implications point toward precision soil management. Legume rotations are promoted worldwide as a sustainability strategy, but this study suggests that their success is contingent on regional soil conditions that farmers and policymakers rarely measure in an integrated way. In alkaline warm-temperate soils, interventions that lower pH or replenish carbon, such as returning crop residues or applying organic amendments, may be prerequisites for realizing the full benefit of legumes. In soils where micronutrients fall below threshold levels, targeted supplementation of trace metals could unlock enzymatic processes that currently operate below capacity. Conversely, in already high-performing subtropical and plateau soils, the priority may simply be to protect existing carbon stocks and micronutrient supplies from degradation.

The study also reframes the role of microbial diversity in agricultural soils. Although microbial diversity contributed only 14.2 percent of predictor support in the best model, the structural equation analysis indicated that integrated soil carbon and nitrogen stocks were associated with multifunctionality through microbial pathways. This suggests that microbial communities act less as independent engines of soil function and more as intermediaries whose activity is governed by the chemical resources available to them. Building soil carbon, in this view, is not merely a climate strategy but a way of feeding the microbial networks that carry out the work of nutrient cycling. As legume-based farming expands to meet demands for lower-emission agriculture, this nationwide analysis provides a diagnostic framework: measure the pH, the carbon and nitrogen stocks, and the micronutrient thresholds first, and only then expect the biology to deliver its promise.

Subject of Research: Environmental drivers of soil multifunctionality in legume-based agroecosystems across China

Article Title: Soil pH and micronutrient thresholds regulate multifunctionality in legume-based agroecosystems across China

Article References: Dong, Z., Xi, J., Liu, Y., Chen, Y., Liu, T., Wu, J., Wang, X., Liu, H., Zhu, Z., Chen, J., Ge, T., & Liu, Q. (2026). Soil pH and micronutrient thresholds regulate multifunctionality in legume-based agroecosystems across China. Plant and Soil. https://doi.org/10.1007/s11104-026-09121-2

Image Credits: AI Generated

DOI: 10.1007/s11104-026-09121-2

Keywords: soil multifunctionality, legume rotation, soil pH, micronutrients, nitrogen fixation, random forest, structural equation modeling, soil organic carbon, Qinghai-Tibetan Plateau, nutrient stoichiometry, microbial diversity, precision agriculture

The logic behind the experiment rests on a fundamental constraint of microbial life: stoichiometry. Heterotrophic bacteria—the workhorses of nutrient removal in streams—need carbon, nitrogen, and phosphorus in roughly balanced proportions to build biomass and fuel respiration. Wastewater treatment plant effluents disrupt that balance. They deliver heavy loads of dissolved inorganic nitrogen and soluble reactive phosphorus, but comparatively little bioavailable dissolved organic carbon. The result is a stream in which microbes are, in effect, starving for carbon while awash in nutrients, a condition that constrains their capacity to take up nitrogen and phosphorus and allows those nutrients to travel unimpeded downstream, where they can fuel algal blooms, oxygen depletion, and coastal eutrophication far from the source.

Pineda-Morante and colleagues—including Miquel Ribot, Susana Bernal, Anna Lupon, Sara Castelar, Esperança Gacia, Stephanie N. Merbt, Helena Guasch, and Eugènia Martí at CEAB-CSIC, together with Francesc Sabater at the University of Barcelona—began with a careful stoichiometric diagnosis of the receiving stream. Their analysis suggested that heterotrophic nutrient uptake in this system was constrained by both dissolved organic carbon and dissolved inorganic nitrogen, relative to an abundant supply of soluble reactive phosphorus. In other words, phosphorus was not the limiting resource here; carbon and nitrogen were. That diagnosis shaped the team’s hypothesis: enriching the stream with bioavailable dissolved organic carbon should partially relieve the carbon constraint, stimulate heterotrophic activity, and thereby enhance the uptake of both nitrogen and phosphorus within the stream reach itself.

What followed was a reach-scale enrichment of the kind that has become a gold standard in stream biogeochemistry. The researchers continuously dosed the stream with the brewery-derived dissolved organic matter solution for four days, while collecting hourly water samples from one station upstream of the injection point and two stations downstream. This high-frequency sampling design was essential for capturing a phenomenon that shorter, coarser studies routinely miss: the pronounced day–night rhythm, or diel variability, in stream nutrient processing. Streams are not chemically static. Photosynthesis by benthic algae during daylight hours and respiration by microbes around the clock generate cycles in oxygen, pH, and nutrient demand that can shift the apparent behavior of a reach from hour to hour.

The response to carbon enrichment was rapid and, in places, dramatic. Within the first meters downstream of the injection point, the team observed strong net removal of dissolved organic carbon, coinciding with pronounced oxygen depletion and the visible development of benthic microbial mats. Those mats—dense communities of bacteria and algae colonizing the streambed—were, in effect, the physical signature of a microbial economy suddenly flush with its limiting resource. The influx of carbon triggered a surge in heterotrophic demand, and the microbial community consumed the added carbon quickly enough to draw down oxygen concentrations in the water flowing over the mats, a classic indication of intense aerobic respiration.

Most consequentially for water quality, the stream reach flipped its functional identity. Before enrichment, the reach acted as a net source of dissolved inorganic nitrogen, meaning more nitrogen exited the reach than entered it—a counterintuitive but well-documented behavior in effluent-dominated streams, where internal processes such as mineralization can release more nitrogen than the reach retains. During the enrichment, that pattern reversed: the reach became a net sink for dissolved inorganic nitrogen, driven mainly by enhanced nitrate uptake. Nitrate, the most abundant and mobile form of nitrogen in most effluents, is notoriously difficult for streams to retain, so a carbon-triggered boost in nitrate uptake is precisely the outcome a nutrient manager would hope to see.

Not every nutrient responded equally, however, and the nuances matter. Ammonium uptake showed a weaker, more spatially localized, and more clearly diel-dependent response, waxing and waning with the daily light cycle rather than increasing uniformly. Soluble reactive phosphorus uptake, meanwhile, showed no clear effect of the carbon addition at all. That outcome is consistent with the team’s initial stoichiometric diagnosis: because phosphorus was already abundant relative to carbon and nitrogen in this stream, relieving the carbon limitation did not create additional phosphorus demand. Microbes with ample phosphorus and newly available carbon directed their extra assimilatory capacity toward nitrogen, particularly nitrate, rather than toward phosphorus.

The study also delivers a sobering reality check. Despite the enhanced local uptake within the dosed reach, the whole-reach reduction in nitrogen and phosphorus export remained below fifteen percent relative to the total nutrient loads delivered by the wastewater treatment plant. In other words, even a four-day, deliberately generous carbon subsidy that visibly transformed the streambed could only trim a modest fraction of the nutrient export leaving the system. Effluent-dominated streams, the findings imply, cannot simply metabolize their way out of the nutrient loads they receive; the loads are simply too large relative to the residence time and processing capacity of the water moving through the channel. Any carbon-based attenuation strategy would therefore need to be one component of a broader management approach that also addresses nutrient loads at the treatment plant itself.

Yet the experiment’s most durable contribution may be methodological rather than outcome-based. The authors frame their results as support for using system-specific carbon–nitrogen–phosphorus stoichiometry to guide dissolved-organic-carbon-based nutrient attenuation strategies. Rather than applying a generic carbon-dosing prescription, managers could first diagnose which element limits heterotrophic nutrient processing in a given stream—carbon, nitrogen, or phosphorus—and target the addition accordingly. In the Catalan study stream, the stoichiometric framework correctly anticipated that carbon enrichment would boost nitrate uptake but leave phosphorus uptake largely untouched, a predictive success that suggests the approach could be exported to other effluent-receiving streams with confidence.

The research also highlights how much fine-grained temporal sampling matters for understanding stream function. Because the team sampled hourly across both upstream and downstream stations for four straight days, they could separate the sustained effects of carbon enrichment from the background diel oscillations in uptake that would have confounded a single grab-sample campaign. Ammonium’s light-dependent response, in particular, would likely have been invisible without that cadence. As freshwater systems worldwide face intensifying pressure from urban effluents, seasonal drought, and warming, studies of this kind—integrating ecological stoichiometry, reach-scale manipulation, and high-frequency sensing—offer a template for diagnosing and, where feasible, engineering the microbial machinery that stands between wastewater and the downstream ecosystems those nutrients would otherwise reach. The work was supported by the Spanish Ministry of Science, Innovation, and Universities through multiple national research projects, and the article is published open access, making the full dataset and methods available to researchers and water managers confronting the same carbon–nutrient imbalances elsewhere.

Subject of Research: How dissolved organic carbon enrichment alters microbial nitrogen and phosphorus uptake in a stream receiving wastewater treatment plant effluent

Article Title: Dissolved organic matter addition alters diel nitrogen and phosphorus uptake downstream of a wastewater treatment plant effluent

Article References: Pineda-Morante, D., Ribot, M., Bernal, S., Lupon, A., Castelar, S., Gacia, E., Merbt, S. N., Sabater, F., Guasch, H., & Martí, E. (2026). Dissolved organic matter addition alters diel nitrogen and phosphorus uptake downstream of a wastewater treatment plant effluent. Biogeochemistry. https://doi.org/10.1007/s10533-026-01363-7

Image Credits: AI Generated

DOI: 10.1007/s10533-026-01363-7

Keywords: biogeochemistry, wastewater effluent, dissolved organic carbon, nitrogen uptake, phosphorus uptake, nutrient stoichiometry, diel variability, stream ecology, heterotrophic microbes, nitrate, eutrophication, Mediterranean stream