Functional regimes define soil microbiome response to environmental change

Abstract

The metabolic activity of soil microbiomes has a central role in global nutrient cycles¹. Understanding how soil metabolic activity responds to climate-driven environmental perturbations is a key challenge^2,3. However, the ecological, spatial and chemical complexity of soils^4,5,6 impedes understanding how these communities respond to perturbations. Here we address this complexity by combining dynamic measurements of respiratory nitrate metabolism⁷ with modelling to reveal functional regimes that define soil responses to environmental change. Measurements across more than 1,500 soil microcosms subjected to pH perturbations^8,9 reveal regimes in which distinct mechanisms govern metabolite dynamics. A minimal model with two parameters, biomass activity and growth-limiting nutrient availability, predicts nitrate utilization dynamics across soils and pH perturbations. Parameter shifts under perturbation reveal three functional regimes, each linked to distinct mechanisms: (1) an acidic regime marked by cell death and suppressed metabolism; (2) a nutrient-limited regime in which dominant taxa exploit matrix-released nutrients; and (3) a resurgent growth regime driven by exponential growth of rare taxa in nutrient-rich conditions. We validated these model-derived mechanisms with nutrient measurements, amendment experiments, sequencing and isolate studies. Additional experiments and meta-analyses suggest that functional regimes are widespread in pH-perturbed soils.

Main

The metabolic activity of soil, marine and freshwater microbiomes drives carbon and nitrogen transformations that sustain biogeochemical cycles and life in the biosphere². These microbiomes are also subjected to environmental perturbations including changes in temperature, pH, moisture, oxygen and nutrients stemming from natural and anthropogenic events^10,11. To predict the effect of climate change on global nutrient cycles, it is necessary to understand how microbiome metabolism responds to environmental change in nature.

Determining how environmental change affects community metabolism has proved difficult owing to the complexity of natural microbiomes. This complexity is perhaps most apparent in soils, which possess immense taxonomic diversity¹², spatial heterogeneity⁵ and chemically diverse environments¹³. As a result, environmental perturbations can modify collective metabolic activity in many ways, from direct changes in microbial composition, physiology¹⁴ and ecological interactions⁸ to indirect modification of nutrient availability^15,16 and spatial organization¹⁷. Thus, a key question is which mechanisms determine the metabolic response of complex microbiomes to environmental change.

Large-scale surveys approach this question by quantifying correlations between environmental variation, community composition and metabolic processes in the wild^{4,18,19,20,21}. Although surveys have revealed robust correlations, they face two challenges in uncovering the mechanisms determining community response to environmental change. First and most importantly, surveys cannot control for confounding factors such as correlated environmental variables, rendering any causal inference infeasible. Second, it is difficult to quantify metabolic dynamics in situ on a large scale in the wild. As a result, surveys have limited power for determining the mechanisms that govern the metabolic response to environmental change in natural communities.

To control for confounding factors and gain mechanistic insights, we use soil microcosms—this removes correlated environmental fluctuations and permits controlled perturbations in the laboratory. To further control for confounding factors, these soils are sourced from a single site (Cook Agronomy Farm, WA, USA)²² that exhibits large natural pH variation but minimal variability in other environmental factors. Crucially, soil microcosms enable high-throughput quantification of metabolite dynamics in response to environmental perturbations. Leveraging insights from global surveys, we focus on pH, the environmental variable that is most strongly correlated with soil microbiome composition and metabolism^4,8,23.

Our metabolic measurements target anaerobic nitrate respiration, a central process in nitrogen cycling that is widely carried out by soil bacteria across pH gradients^24,25. Nitrate (${{\rm{NO}}}_{3}^{-}$), which has critical implications for agriculture and climate, is reduced in soils when bacteria use it as an electron acceptor during anaerobic respiration in the absence of oxygen. Both denitrification (${{\rm{N}}{\rm{O}}}_{3}^{-}\to {{\rm{N}}{\rm{O}}}_{2}^{-}\to ...\to {{\rm{N}}}_{2}$) and dissimilatory nitrate reduction to ammonia (DNRA; ${{\rm{NO}}}_{3}^{-}\to {{\rm{NO}}}_{2}^{-}\to {{\rm{NH}}}_{4}^{+}$) reduce nitrate to nitrite (${{\rm{NO}}}_{2}^{-}$) while consuming organic carbon.

We measure nitrate utilization dynamics in more than 1,500 microcosms across a wide range of natural and laboratory-induced pH changes. A judicious dynamic model describes nitrate utilization across microcosms in terms of three variables: the quantity of a single functional biomass, nitrate, and a limiting nutrient. Changes in nitrate dynamics in response to perturbations arise from differences in two model parameters that vary with pH: the initial quantity of biomass activity utilizing nitrate and the initial quantity of limiting nutrients. These two parameters emerge naturally from our mathematical model using only the community-level nitrate uptake data. The model predicts that changes in pH influence nitrate utilization dynamics through mechanisms that affect biomass activity and nutrient availability, and these predictions are validated experimentally. We demonstrate the generality of these findings through experiments on soils from other sampling sites and a quantitative meta-analysis of past studies.

Despite the ecological, chemical and spatial complexity of soils, we find that the functional response of the soil microbiome to changes in pH can be categorized into three mechanistically distinct regimes demarcated by the levels of these two parameters. Each functional regime is defined by which of the two parameters exerts greater control over nitrate utilization rates. During moderate pH perturbations, metabolic rates are set by the pH-mediated release of nutrients from soil particles that limit the growth of biomass (nutrient-limiting regime, Regime II). When soils are subjected to large basic perturbations, massive nutrient release relieves the nutrient limitation, but the dominant taxa are no longer metabolically active, and metabolism is set by the rapid growth of rare taxa (resurgent growth regime, Regime III). During large acidic perturbations, functional responses are limited by the pervasive death of the functional biomass in the community (acidic death regime, Regime I). The transition between functional regimes can be abrupt (Regime II to III) or smooth (Regime I to II) as pH is varied. Although the presence of functional regimes is consistent across soils, the pH at which regime transitions occur varies according to the long-term pH of the soil. Our study presents a generalizable approach in which high-throughput soil microcosm experiments and mathematical models reveal the microscopic processes driving microbiome responses to environmental change.

Soil metabolite dynamics change with pH

We measured nitrate utilization dynamics in soil microcosms across a range of native and perturbed pH levels. We sampled 20 topsoils with native pH from 4.7 to 8.3 (Fig. 1a and Supplementary Table 1) at the Long-term Agricultural Research Cook Agronomy Farm (CAF) (Pullman, WA, USA). At this site, long-term variation in soil pH arises from agricultural practices and differential erosion. Although 20 CAF sites had similar soil texture (silty clay loam) (Supplementary Table 1), their variation in soil pH correlated well with cation exchange capacity, sulfur and phosphorous levels (Extended Data Fig. 8q, Supplementary Information and Supplementary Table 3).

Fig. 1: Soil microbiome metabolite dynamics under short-term and long-term pH variation.

a, Schematic of field sampling across a long-term pH gradient (n = 20 soils, pH 4.7 to 8.3) subjected to short-term pH perturbations (n = 13) in laboratory conditions (Methods). Soil microcosms were created by making slurries (1:2, soil:water) amended with 2 mM nitrate, adjusted to 13 different pH levels, and treated with (no growth) or without chloramphenicol (growth) in triplicate (n = 1, 704 microcosms including no-nitrate (n = 120) and cycloheximide controls (n = 120)). Microcosms were incubated anaerobically for four days, and nitrate was quantified colorimetrically via sampling. Communities were quantified by 16S rRNA amplicon sequencing before and after incubation. Chl, chloramphenicol. b, Nitrate concentration over time for 3 of 20 soils with different native pH, perturbed to either acidic or basic pH. Results are shown with or without chloramphenicol treatment (n = 126 microcosms). Endpoint pH is indicated in each graph (Methods). c, Schematic of functional regimes. For moderate pH perturbations (middle), the growth of dominant taxa (blue) is limited by available nutrients (purple) and nitrate dynamics are linear (nutrient-limiting, Regime II). During strong basic perturbations (right), growth-limiting nutrients released from the soil matrix are in excess and rare taxa (green) dominate growth (resurgent growth, Regime III). Acidic perturbations show minimal activity, partly owing to cell death (acidic death, Regime I). Lines depict dominant (blue) and rare (green) biomass. See Extended Data Fig. 1 for nitrate dynamics across all slurry experiments.

Source data

For each soil sample, we created mixtures of soil and water (slurries) with 2 mM nitrate and varying levels of strong acid or base to perturb the pH of each soil to 13 values between 3 and 9 (Fig. 1a). By applying pH perturbations to soils sampled across a gradient of native pH values, our experiment probed short-term and long-term responses to perturbations. We used slurries to make amendments easier, limit the effects of differential water content and mimic rain events, when most anaerobic respiratory nitrate utilization occurs²⁶. Soil slurries retained much, but not all, of the complexity of the natural context, including the diversity of the communities, the soil nutrient composition and the spatial structure due to intact soil grains.

To separate the activity of pre-existing nitrate utilizers from growth in each condition²⁷, we included controls treated with chloramphenicol, which inhibits protein synthesis (Fig. 1a,b). In each microcosm, we focused on the dynamics of nitrate during the four-day incubation in anaerobic conditions (Fig. 1a). Focusing on a non-gaseous metabolite enabled us to perform measurements of metabolite dynamics with high temporal resolution across the approximately 1,500 microcosms. In addition, we selected ten soils spread evenly across the native pH gradient and performed 16S ribosomal RNA (rRNA) amplicon sequencing before and after incubation.

Nitrate dynamics for a subset of soils and pH perturbations are shown in Fig. 1b. All chloramphenicol-treated conditions, regardless of soil or pH, exhibited linear nitrate dynamics (Fig. 1b and Extended Data Fig. 1). Chloramphenicol is bacteriostatic and inhibits protein synthesis, arresting growth while leaving existing enzymes intact; as a result, nitrate is reduced at a constant rate (linear dynamics). Thus, in chloramphenicol-treated conditions, the slope of nitrate decline quantifies the activity of the pre-existing nitrate-reducing biomass at each pH²⁶. By contrast, non-chloramphenicol conditions reflect metabolite dynamics influenced by both pre-existing activity and growth.

Across pH perturbations and soils, we observed three distinct nitrate utilization regimes (Fig. 1b). First, under strong acidic perturbations, nitrate reduction was minimal in both chloramphenicol-treated and untreated conditions (Fig. 1b, left columns), indicating little pre-existing nitrate-reducing biomass and no growth. Second, at pH levels near the native pH, nitrate declined linearly even in non-chloramphenicol samples (Fig. 1b and Extended Data Fig. 1), at faster rates than chloramphenicol-treated controls. This suggests that some growth occurred in non-chloramphenicol conditions but was inhibited, potentially by a lack of nutrients other than nitrate (Fig. 1c, schematic). Third, under strongly basic conditions (pH > 8), non-chloramphenicol samples showed accelerating nitrate reduction, whereas chloramphenicol-treated samples showed little activity (Fig. 1b, right columns). The lack of nitrate reduction in chloramphenicol-treated suggests that the indigenous nitrate-utilizing population is small, but this rare population expands rapidly, exhausting nitrate in non-chloramphenicol conditions.

Model captures metabolite dynamics

To describe the nitrate dynamics, we used the consumer-resource model presented in Fig. 2. The model subsumes the ecological complexity of the soil microbiome into a single effective biomass rather than explicitly considering the multitude of taxa and their interactions. The model includes three variables: functional nitrate-utilizing biomass (x), nitrate concentration (A) and a growth-limiting nutrient (C), along with five parameters: consumption rates (r_A and r_C), affinities (K_A and K_C) and a biomass growth rate (γ).

Fig. 2: Consumer-resource model describes metabolite dynamics.

Left column, a consumer-resource model describes the community-level metabolism via a single functional biomass (x), nitrate concentration (A (mM)) and a growth-limiting nutrient concentration (C (mM)). Nitrate consumption rate ($\mathop{A}\limits^{.}(t)$) takes a Monod form with a reduction rate parameter (r_A (mM per g biomass per day)) and an affinity parameter (K_A (mM)). A non-substitutable nutrient (C) is consumed at a rate r_C (mM per g biomass per day) with affinity K_C (mM). Growth of functional biomass ($\mathop{x}\limits^{.}(t)$) is determined by the product of Monod uptake terms for non-substitutable nutrients (A and C) and the growth rate (γ (day⁻¹)). If either nutrient is exhausted, growth halts. The middle and right columns show dynamics of x(t), A(t) and C(t) for two regimes. Chloramphenicol-treated conditions always show constant A and C utilization rates. Middle column, small C(0), large x(0). With low initial C(0) and large x(0), C is rapidly exhausted (at t*), arresting biomass growth (bottom) and resulting in constant A utilization (top). Right column, large C(0), small x(0). Excess C(0) and small x(0) lead to exponential growth in functional biomass, driving exponential depletion of A (black line, top). Growth arrests when A is exhausted. We assume K_A ≪ A(0) and K_C ≪ C(0) (Methods). Tables define variables and parameters. See Extended Data Fig. 3a–c.

The model has two key properties. First, the nitrate utilization rate ($\dot{A}$) is proportional to the amount of functional biomass (x), even in the absence of growth. Thus, in chloramphenicol-treated conditions where biomass does not grow ($\dot{x}=0$), the consumption rate of A is determined by r_Ax and is constant over time, as observed experimentally (Fig. 1b). Second, the uptake of A and C by functional biomass follows a co-limiting Monod form, such that depletion of either resource halts growth. Together, these properties imply that when C is exhausted, growth ceases and nitrate (A) is consumed at a constant rate proportional to x.

To build intuition for how the model captures the observed nitrate dynamics, the right two columns of Fig. 2 illustrate two regimes. When the initial concentration of the limiting nutrient C(0) is low (middle column), C is depleted early (at time t*), halting growth and leading to constant nitrate (A) consumption thereafter (grey dashed line). This recapitulates the late-time linear dynamics in non-chloramphenicol conditions for moderate pH perturbations (Fig. 1b). By contrast, when the initial nutrient concentration C(0) is large (Fig. 2, right column), it is nitrate (A) that runs out first. Therefore, a small x(0) and a large C(0) recapitulate the initially slow but accelerating rate of nitrate utilization observed for basic perturbations (Fig. 2, right column).

We fit this model to nitrate utilization dynamics from all 20 soils and 13 pH perturbations, with and without chloramphenicol treatment. We fixed the growth rate γ and the affinity parameters (K_A and K_C), and varied two rescaled parameters: $\widetilde{x}(0)=x(0){r}_{A}$ and $\gamma \widetilde{C}(0)=\gamma C(0){r}_{A}/{r}_{C}$ (Extended Data Fig. 3c and Methods). These parameters retain the same interpretation as x and C: $\widetilde{x}(0)$ reflects the indigenous metabolic activity of all taxa that can perform nitrate reduction in a given condition, and $\gamma \widetilde{C}(0)$ reflects the available limiting nutrient. The rescaling corresponds to measuring these quantities in terms of nitrate utilization rates, and their values determine the metabolite dynamics as shown in Fig. 2. The model provided a good fit in all soils (less than 10% error per data point, Extended Data Fig. 3a,b).

Model reveals functional regimes

We plotted indigenous biomass activity ($\widetilde{x}(0)$) against available limiting nutrient ($\gamma \widetilde{C}(0)$) (Fig. 3a) and identified three regimes of nitrate utilization dynamics (Extended Data Fig. 3d–g and Methods). Regime I—the acidic death regime—in which both $\widetilde{x}(0)$ and $\gamma \widetilde{C}(0)$ are low, is observed for pH ≤ 4 and shows little to no nitrate reduction (Fig. 3b(i),(iv)). Regime II—the nutrient-limiting regime—in which $\widetilde{x}(0)$ is large and $\gamma \widetilde{C}(0)$ is small, is observed for 4 ≤ pH ≤ 8 and exhibits constant nitrate reduction rates with and without chloramphenicol treatment, with higher rates in the latter (Fig. 3b(ii),(v)). Regime III—the resurgent growth regime—in which $\widetilde{x}(0)$ is small and $\gamma \widetilde{C}(0)$ is large, is observed for pH ≥ 8 and displays a near-zero initial utilization rate, followed by an exponential speed-up that continues until nitrate is depleted (Fig. 3b(iii),(vi)). We observe the three functional regimes in all soils, but the exact pH at which a given soil transitions between regimes depends on the native pH. This is illustrated in Fig. 3c,d, in which the indigenous biomass activity and the quantity of the limiting nutrient are shown as heatmaps for soils at different perturbed and native pH; the same data are presented in Fig. 3e,f as lines with the x axis corresponding to perturbed pH.

Fig. 3: Functional regimes capture soil response to pH perturbations.

a, Scatter plot of two model parameters—indigenous biomass activity ($\widetilde{x}(0)$) and limiting nutrient concentration ($\gamma \widetilde{C}(0)$)—inferred from nitrate dynamics across all samples (n = 244 conditions, representing median fitted parameter values from 3 biological replicates, n = 732 dynamics). Data are log₁₀(x + 0.01)-transformed for visualization. Point colour indicates perturbed pH. Dashed lines separate three functional regimes: acidic death (Regime I), nutrient-limiting (Regime II) and resurgent growth (Regime III). Regime boundaries are set by thresholds in $\widetilde{x}(0)$ and $\gamma \widetilde{C}(0)$ (Extended Data Fig. 3d,e and Methods). b, Example nitrate dynamics in each regime for neutral and acidic soils. Panel labels correspond to points in c,d. c,d, Indigenous biomass (c) and limiting nutrient (d) for different native pH and perturbed pH values (n = 244 conditions). Colour indicates fitted values. Black lines indicate regime boundaries. e, $\widetilde{x}(0)$ trends across perturbed pH values for soils with different native pH, showing regime transitions and a plateau of high activity in Regime II. f, Same as e, but for $\gamma \widetilde{C}(0)$, showing increased limiting nutrients under basic pH. Points in e,f, show the median of three replicates (n = 244). Horizontal dashed lines mark thresholds from a (Extended Data Fig. 3d,e).

Source data

The pH-dependent changes in $\widetilde{x}(0)$ (Fig. 3c,e) reflect shifts in indigenous biomass activity. These changes may result from variations in the abundance of nitrate-reducing taxa, differential expression of relevant enzymes, or changes in enzymatic activity, all influenced by pH. To begin to determine the dominant mechanisms that govern changes in metabolism across regimes, we quantified compositional changes through sequencing, which enabled us to test model predictions.

Taxonomic patterns across regimes

We measured absolute abundances through 16S amplicon sequencing after the four-day incubation in both chloramphenicol-treated and untreated conditions in half of the soils shown in Fig. 3. We computed phylum-level growth folds as the ratio of endpoint absolute abundance between untreated and chloramphenicol-treated conditions (Extended Data Fig. 5b), where phylum-level abundance refers to the aggregate of all amplicon sequence variants (ASVs) within that phylum. A growth fold greater than 1 indicates increased abundance in the absence of the drug. We used non-negative matrix factorization (NMF) to decompose the variation in growth at the phylum level across all soils and pH perturbations (Methods). The analysis showed that most of the growth could be captured with just two axes of variation (Extended Data Fig. 5c). One axis comprised Pseudomonadota combined with Bacteroidota, and the other comprised Bacillota alone.

Figure 4a,b shows growth folds for the two groups of phyla identified by NMF that dominate growth across all soils and pH conditions. In Regime II, Pseudomonadota and Bacteroidota showed increased growth with increasing pH, followed by a decline near the onset of Regime III. This matches the behaviour of the indigenous biomass activity ($\widetilde{x}(0)$) revealed by the model in Regime II (Fig. 3c). Bacillota did not grow until a pH threshold between 7 and 8.5, which matches the onset of exponential nitrate utilization dynamics in Regime III (Fig. 3f). Notably, the boundary between Regime II and Regime III derived from the functional dynamics data (Fig. 3c,d) aligns with the increased growth of Bacillota (Fig. 4b) and declining growth of Pseudomonadota and Bacteroidota (Fig. 4a). These patterns suggest that changes in the identity of the phyla responsible for nitrate reduction reflect functional regimes.

Fig. 4: Taxonomic signatures of functional regimes.

16S rRNA amplicon sequencing at the end of each incubation was used to identify ASVs in chloramphenicol-treated and untreated conditions (Methods). ASVs were aggregated at the phylum level. For each phylum, a growth fold was computed as the ratio of abundance with growth to that without growth. A statistical decomposition across all conditions identified two groups of phyla that dominated abundance changes due to growth: with Pseudomonadota and Bacteroidota in one group, and Bacillota in another (Extended Data Fig. 5a–c). a, Top, growth folds for the aggregated abundances of Pseudomonadota and Bacteroidota (combined abundance) are indicated by colour for each native and perturbed pH condition (n = 130 conditions). Regime boundary lines are those determined in Fig. 3. Bottom, growth folds are plotted on a log scale, with colour indicating native pH given in the colour bar. b, Identical to a, but showing growth folds of Bacillota increasing during the transition from Regime II to Regime III after strong basic pH perturbations (n = 130 conditions). c, Scatter plot of growth folds of Pseudomonadota and Bacteroidota versus Bacillota (n = 130 conditions). Points outlined in red are associated with Regime II and exhibit high growth for all three phyla. d, As in Fig. 3a, plot of $\widetilde{x}(0)$ versus $\gamma \widetilde{C}(0)$ with points marked by regime, with red outlines indicating points in Regime II conditions near the boundary with Regime III; these points are in Regime II, but they exhibit high Bacillota abundance (c). Data in a,b,d are from a subset of samples from those in Fig. 3 because only half of the soils were sequenced (n = 10 soils). See also Extended Data Fig. 6.

Source data

An analysis of the likely metabolic traits of these taxa²⁸ suggests that the transition from Regime II to Regime III is accompanied by a metabolic shift from denitrification to DNRA, which agrees with the fact that excess carbon favours DNRA²⁹ (Extended Data Fig. 6g and Supplementary Information). Consistent with the metabolic shift to DNRA, we observed an accumulation of ammonia in Regime III (Extended Data Fig. 5g)

Increase in Bacilliota signals regime change

We observed that Bacillota began to increase in abundance at pH levels just below the transition from Regime II to Regime III, thereby acting as ‘early warning indicators’ for the impending transition (Fig. 4a–d, red circles). When we plot the growth folds of the Bacillota versus Pseudomonadota and Bacteroidota, we find that Bacillota abundances begin to increase before the system enters Regime III as defined by nitrate utilization dynamics (Fig. 4c,d). This finding indicates that compositional data can be used to anticipate impending regime transitions during environmental perturbations.

Having quantified pH-induced shifts in community composition, we next explored whether these data, combined with changes in model parameters, could reveal mechanisms underlying community metabolism across regimes.

Rare taxa grow rapidly in Regime III

Under strong basic pH perturbations, all soils enter Regime III, which is marked by a sharp shift from linear to exponential nitrate consumption dynamics in chloramphenicol-free conditions. Our model suggests that these metabolite dynamics arise from a small indigenous biomass activity ($\widetilde{x}(0)$), indicative of rare taxa undergoing exponential growth in response to excess nutrient availability ($\gamma \widetilde{C}(0)$) (Fig. 3c–f). Sequencing measurements show that Bacilliota dominate growth in Regime III (Fig. 4b,d). Together, these observations suggest that initially rare members of the Bacilliota phylum are responsible for growth in Regime III (Fig. 4b). Consistent with this prediction, we found that the taxa that drive growth in Regime III start from very low abundance and are enriched several hundred fold (Extended Data Fig. 6d).

Carbon release fuels Regime II growth

In the nutrient-limiting Regime II, nitrate reduction rates remain constant with or without chloramphenicol, but are higher in the absence of the drug. The model attributes the increased rate in the absence of the drug to the rapid initial utilization of nutrient (C(0)) (Fig. 2, middle column). Further, as the perturbed pH increases, the nitrate reduction rate in chloramphenicol-free conditions increases (Fig. 1b). Our model proposes that the increasing availability of the growth-limiting nutrient with pH ($\gamma \widetilde{C}(0)$) (Fig. 3a,d,f) drives growth that increases nitrate reduction rate. Here, we test this hypothesis.

Increasing pH can enhance the availability of organic carbon in soils^30,31 through substitution reactions at ion exchange sites on clay particles^32,33 (Fig. 5b). Since nutrients are released via exchange reactions, we hypothesized that the increase in biomass, and therefore nitrate reduction rate, should be proportional to the amount of acid or base added to the system. In Fig. 5a, we observed precisely this trend across all soils, as evidenced by a consistent increase in nitrate reduction rate with NaOH (light blue region). The trend is specific to Regime II (Extended Data Fig. 4c), and if the data are plotted against pH, the correlation is weaker (Extended Data Fig. 4a). If the increasing rate of nitrate utilization with NaOH is due to increased nutrient availability ($\gamma \widetilde{C}(0)$) driving the growth of nitrate reducers, we expect their abundances to rise with increasing NaOH. As expected, we observe a linear relationship between absolute abundances, measured via 16S rRNA amplicon sequencing, and the amount of NaOH added to the system (Extended Data Fig. 4c–h and Methods). This increase is reflected at coarse (total biomass; Extended Data Fig. 4d,e) and fine taxonomic levels (putative nitrate-reducing ASVs; Extended Data Fig. 4f–h and Supplementary Information).

Fig. 5: Carbon limits growth in Regime II and is released from the soil matrix.

a, Fold change in nitrate reduction rate (without chloramphenicol/chloramphenicol-treated) as a function of the amount of NaOH added to the soil (n = 470 samples of <30 mM OH⁻, <25 mM H⁺; 28 data points with fold change >5.5 were excluded for clarity of visualization). A linear increase in fold change from 0 mM to 25 mM NaOH corresponds to the nutrient-limiting regime (Regime II, light blue region), and suggests that $\widetilde{C}(0)$ increases with increasing NaOH. b, Cartoon illustrating the hypothesized mechanism of nutrient release. Adding NaOH results in the release of anionic nutrients (magenta spheres, Case 2) from soil particles (brown region), whereas the addition of HCl would release cationic nutrients (grey spheres, Case 1). Microorganisms access nutrients in pore water but not those adsorbed to soil particles. Added OH⁻ ions decrease the number of anion exchange sites, releasing anionic nutrients. Na⁺ ions stabilize the released anions (Extended Data Fig. 4j and Supplementary Information). Data in a and model in b suggest that growth-limiting nutrients are anionic (negatively charged). c, Scatter plot of model-inferred available limiting nutrient ($\widetilde{C}(0)$) with measured WSOC (endpoint WSOC, Methods) (n = 222 samples of 0–25 mM OH⁻ in soil samples 1–12). d,e, Amendment experiments for soil in the nutrient-limiting regime at an unperturbed pH (5.4). d, Nitrate dynamics with different glucose amendments. Dots represent data and lines show model predictions. e, Nitrate reduction rates after amending soils with different concentrations of nutrients. Dots represent mean rates, estimated by linear regression on triplicates, and error bars indicate s.d. Carbon amendments (succinate, acetate and glucose) increased nitrate reduction rates, but ammonium, sulfate and phosphate did not. Model predictions are shown for 1 < r_A/r_C < 3 (shaded region) with a line for r_A/r_C = 2 (best fit). See also Extended Data Figs. 2g,h and 4.

Source data

The charge of the limiting nutrient is suggested by the asymmetric response to NaOH and HCl (Fig. 5a), in which NaOH releases anions and HCl releases cations (Extended Data Fig. 4j and Supplementary Information). Since NaOH increased nitrate reduction rates, we infer that the limiting nutrient is anionic. Water-soluble organic carbon (WSOC) also increased linearly with NaOH (Fig. 5c), suggesting that it, or co-released anionic N, S or P, could be the limiting nutrient.

To identify the limiting nutrient, we performed an amendment experiment on a representative soil (Soil 6, pH 5.4; Methods). We amended a soil slurry with glucose (neutral), succinate (anion when pH > pK_a, pK_a = 4.2), acetate (anion when pH > pK_a, pK_a = 4.75), phosphate (anion), ammonium (cation) and sulfate (anion) added in varying concentrations without perturbing pH (Extended Data Fig. 2h and Methods). We found that the amendment of carbon, but not other resources (N, S or P), increased the nitrate reduction rate, changing the linear dynamics to exponential (Fig. 5d,e), indicating that carbon was the limiting nutrient. With a single free parameter (r_C/r_A), our model fit the nitrate utilization dynamics in soil amended with carbon (Fig. 5d,e). The ratio r_C/r_A can be interpreted as the stoichiometry of carbon to nitrate utilization (Fig. 2). This ratio is highest for glucose (2.5) and lowest for acetate (1), suggesting that glucose amendments support faster carbon utilization. This may reflect the fact that glucose is fermentable, whereas acetate is not.

The amendment experiment confirms the model prediction that a nutrient other than nitrate limits reduction dynamics and provides strong evidence against other mechanisms limiting growth, such as predation by phage or eukaryotes (Extended Data Fig. 2a,b). Critically, this insight emerged from our mathematical description of the nitrate utilization dynamics across pH perturbations.

Biomass is reduced by death in Regime I

In response to acidic pH perturbations, the model indicates a reduction in indigenous biomass activity ($\widetilde{x}(0)$) and a decrease in the availability of limiting nutrients (Fig. 3). To test whether the decline in indigenous biomass activity ($\widetilde{x}(0)$) with pH is associated with bacterial abundance, we computed the fold change in endpoint absolute abundance for each phylum in chloramphenicol-treated conditions relative to abundance at the initial time point T₀ (‘survival fold’; Extended Data Fig. 7a). Survival folds reflect the change in abundance in the absence of growth—thus, we regard this as a proxy for death. For all phyla except Bacillota, we observed a consistent decrease in survival folds during acidic perturbations (Extended Data Fig. 7a). Survival folds decline approximately linearly with the $\widetilde{x}(0)$, suggesting that a reduction in biomass activity might be associated with cell death.

However, changes in survival folds might also arise from pH-dependent degradation of relic DNA³⁴. To test the hypothesis that cell death reduces biomass activity in Regime I, we used isolates from our soils or strain collections representing the three phyla identified in Fig. 4. Using these isolates, we measured DNA degradation and cell death rates across a range of pH values from 3 to 7. Combining these rates with a model enabled us to conclude that the change in abundances measured via sequencing must arise at least in part from cell death (Extended Data Fig. 7c–j and Supplementary Information). Bacillota isolates exhibited lower death rates at low pH, consistent with their high survival fold in acidic conditions (Extended Data Fig. 7a). We conclude that acidic perturbations must induce some cell death, but that other physiological or ecological mechanisms are also likely to contribute (Extended Data Fig. 7a).

Functional regimes generalize across soils

Having characterized functional regimes in one set of soil samples (CAF; Fig. 3), we investigated whether the finding of regimes could be generalized to other soils. First, we performed pH perturbation experiments and sequencing for four additional soils from LaBagh (IL, USA), Pinhook (IN, USA), CLG13 and ELG13 (Sedgwick Reserve, CA, USA) (Methods and Supplementary Table 1). For each soil, we found that our model accurately described nitrate dynamics, and the same qualitative functional regimes observed in CAF soils (Fig. 3) were also present (Fig. 6b). Namely, indigenous biomass activity $\widetilde{x}(0)$ decreased, and the available limiting nutrient $\gamma \widetilde{C}(0)$ increased during basic perturbations (Regime III), large $\widetilde{x}(0)$ and modest $\gamma \widetilde{C}(0)$ were observed in Regime II, and both $\widetilde{x}(0)$ and $\gamma \widetilde{C}(0)$ diminished during acidic perturbations (Regime I; Fig. 6b). Sequencing analysis identical to that presented in Fig. 4 revealed qualitatively similar taxonomic patterns in these additional soils. First, an NMF analysis at the phylum level again showed two axes of variation with Bacilliota on one axis and Pseudomonodota and Bacteriodota on the other (compare Extended Data Fig. 5a–c with Extended Data Fig. 5d–f). Second, we observed a qualitatively consistent increase in Bacillota abundance in basic conditions (Fig. 6d). Corroborating this result, we re-analysed a pH perturbation experiment from another study and again found increasing Bacteroidota abundances for modest pH perturbations and substantial Bacillota growth under basic perturbations³⁰ (Fig. 6e).

Fig. 6: Generality of functional regimes.

a, Ternary plot of soil texture (fractions of sand, clay and silt). Soil types are indicated by the colour of the shaded regions. Points (n = 43) represent soils from this study (coloured markers) or our meta-analysis (black markers; Supplementary Tables 1 and 2). Studies in the meta-analysis: Anderson et al.³⁰, Šimek et al.³⁵, Parkin et al.²⁷, Nömmik³⁸, Bremner & Shaw³⁹, Šimek & Hopkins³⁶ and Khalifah & Folz³⁷. b, Scatter plots of fitted model parameters ($\widetilde{x}(0)$ and $\gamma \widetilde{C}(0)$) for soil samples from LaBagh (IL, USA), Pinhook (IN, USA) and Sedgwick CLG13 and ELG13 (CA, USA), each with n = 11 perturbed conditions (median values from triplicates). Colours indicate perturbed soil pH, and dashed lines connect slurries of adjacent perturbed pH levels. c, For three soils from the Czech Republic (Šimek et al.³⁵), scatter plots of early-time denitrification rate (denitrification enzyme activity (DEA, same as $\widetilde{x}(0)$)) and the difference between DEA and long-term denitrification rate (denitrification potential (DP), same as $\gamma \widetilde{C}(0)$; see Supplementary Information). Colours indicate the perturbed soil pH and dashed lines connect soil samples of adjacent perturbed pH. Compare b,c with Fig. 3a. d, Growth folds of Bacillota and the combined abundances of Pseudomonadota and Bacteroidota for soil samples (n = 132 samples) from LaBagh (IL), Pinhook (IN) and Sedgwick CLG13 and ELG13 (CA) across different pH levels (Extended Data Fig. 5d–f and Methods). The y axis was log(x + 0.01)-transformed, as in Fig. 4a,b. e, Relative abundance of phyla over time after pH perturbations, from Anderson et al.³⁰. Bacillota abundance increases under basic perturbations (right two panels) and Bacteroidota abundance increases under moderate pH perturbations. f, Soils from four studies (Parkin et al.²⁷, Šimek & Hopkins³⁶, Šimek et al.³⁵ and Khalifah & Folz³⁷) show a decline in indigenous biomass activity ($\widetilde{x}(0)$) under acidic or basic pH perturbations (n = 10 soils). The y axis is equivalent to $\widetilde{x}(0)$. g, Constant rates of nitrogen utilization near the native pH are observed in soils from previous studies (Bremner & Shaw³⁹ and Nömmik³⁸), as in our Regime II. Amending carbon (wheat straw or glucose (g per g soil)) increased the rate of denitrification (Fig. 5). The legend indicates the type and amount of carbon amended^38,39.

Source data

Next, we conducted a meta-analysis of nine denitrification studies covering soils from diverse locations and soil types (Fig. 6a and Supplementary Table 2). These studies confirmed the fundamental findings of regimes. We were able to infer $\widetilde{x}(0)$ and $\gamma \widetilde{C}(0)$ from denitrification rates measured for three soils from the Czech Republic³⁵, confirming the existence of functional regimes during pH perturbations (Fig. 6c). Second, we observed the same decline in indigenous biomass activity, $\widetilde{x}(0)$ (Fig. 3c,e), under extreme acidic or basic pH perturbations inferred from short-timescale measurements of denitrification rates in ten other soils across four studies^27,35,36,37 (Fig. 6f). Third, consistent with our Regime II, other soils showed linear utilization of electron acceptors near the native pH^38,39 (Fig. 6g). Finally, an increase in reduction rates was observed when the soils were amended with carbon (Fig. 6g), confirming our nutrient-limiting mechanism in Regime II. Together, these results demonstrate the generality of our functional regimes and mechanisms across diverse soil types and environmental conditions (Supplementary Information).

Long-term pH sets regime boundaries

Finally, we observed that the native (long-term) pH of a soil determined the pH threshold for transitioning between functional regimes, with more acidic soils shifting from Regime II to Regime I after smaller acidic perturbations than neutral soils (Fig. 3c,d and Extended Data Fig. 8e). This observation suggests that soil communities are adapted to their long-term pH conditions^27,40.

We suspected that this behaviour was shaped by the community’s past exposure to pH fluctuations. Although we lack direct records of historical pH fluctuations, we can characterize soil response to pH perturbations using titration curves (Extended Data Fig. 8n). Across soils, these curves show a consistent shape: plateaus at high (pH 9) and low (pH 3) pH values connected by a steep, nonlinear transition. The native pH of a soil determines its position along the curve in the absence of perturbations. Acidic soils lie near the lower plateau and are therefore strongly buffered against pH changes (Extended Data Fig. 8n). By contrast, more neutral soils fall along the steep region, where small acid or base additions lead to large pH shifts. We speculate that this difference in buffering makes communities from acidic soils less tolerant to pH fluctuations, potentially explaining their transition from Regime II to Regime I after smaller perturbations (Extended Data Fig. 8e). Quantitative analysis of the transition point supports this interpretation (Extended Data Fig. 8a–e and Supplementary Information).

In addition, our sequencing data support the idea that microbiome composition reflects specific adaptation to native pH, helping to explain variation in regime boundaries across soils. In more acidic soils, Pseudomonadota and Bacteroidota exhibit greater survival at lower pH (Extended Data Fig. 8f). Conversely, the pH at which Bacillota begin to grow in Regime III increases with native soil pH (Fig. 4b and Extended Data Fig. 8f,g). Moreover, the identity of strains that grow in Regime III predicts the native pH of the soil (Extended Data Fig. 8h–l). These findings suggest that prolonged exposure to specific pH conditions selects for particular taxa, potentially shaping the pH thresholds at which communities shift between functional regimes (Methods).

Discussion

Our study has important implications for theoretical microbial ecology and empirical studies of microbiome metabolism. First, the success of our model that abstracts the entire community into a single effective biomass is in stark contrast to complex models of ecosystems that capture many interacting species^41,42. This contrast suggests that natural communities might be better understood through coarse-grained descriptions⁴³ that capture the handful of metabolically relevant groups or guilds in a consortium^44,45. This low-dimensional picture of ecosystems respects underlying mechanisms and is tightly connected to observable functional properties of the community. Extending this success to more complex metabolic processes and connecting effective biomass variables such as $\widetilde{x}$ to underlying abundance dynamics are key directions for future work.

Second, understanding how microbial community metabolism responds to environmental perturbations remains a central problem in applied microbiome science. This challenge arises from the complexity of communities comprising many species with diverse metabolic traits and ecological interactions taking place in chemically complex environments. The complexity of communities has motivated increasingly sophisticated measurements, from metagenomics and transcriptomics^46,47 to single-cell metabolomics⁴⁸ and quantitative stable isotope probes^49,50. Yet, connecting these data to community metabolism and its response to environmental perturbations has remained difficult. Our findings suggest that instead of focusing on microscopic processes individually, we first make quantitative, system-level observations such as nitrate fluxes and describe these fluxes using simple models. The model then proposes mechanisms, such as nutrient limitation or shifts in biomass activity, that organize community metabolism. In turn, the physiological, chemical or ecological origins of these proposed mechanisms can be investigated.

For example, in Regime II, our model and amendment experiments demonstrated that carbon limitation gave rise to constant rates of nitrate utilization (Figs. 3 and 4). The constant rate of electron acceptor utilization is likely to reflect bacterial physiology under starvation, where intracellular components are degraded to support maintenance respiration^51,52. We performed experiments on denitrifying isolates that corroborated this conclusion (Extended Data Fig. 2e,f). This supports previous results linking carbon limitation to nitrate dynamics^15,39,53,54 and highlights how our approach can begin to connect physiological and ecosystem-level processes^26,53,55.

Our analysis revealed conserved phylum-level associations with functional regimes (Fig. 4). Rare Bacillota expanded under strong basic perturbations, whereas dominant Pseudomonadota and Bacteroidota thrived near native pH. These patterns suggest that rare taxa adapted to transient stress and dominant taxa adapted to stable conditions, highlighting the role of fluctuations in maintaining diversity⁴¹. Understanding these taxonomic patterns requires linking dynamics to physiology. Although recent work has developed quantitative insights into the physiology of balanced growth⁵⁶, our understanding of the role of fluctuations in determining microbial traits remains limited⁵⁷.

Even in the absence of physiological insights, our approach enables the prediction of nitrate dynamics from sequencing and nutrient measurements. Using abundance of nitrate reductase genes inferred from 16S data²⁸ and WSOC measurements, we trained regressions to predict the model parameters $\widetilde{x}(0)$ and $\gamma \widetilde{C}(0)$ (Extended Data Fig. 10a–g). This enabled us to predict nitrate fluxes from pH, WSOC and amplicon data alone, demonstrating that functional regimes can link structure and function even in complex communities⁵⁸.

Our study has several limitations. First, soil slurries lack the physical structure and natural environmental fluctuations of intact soils; in particular, in situ nitrate utilization occurs under dynamically changing conditions²⁴. Second, natural pH variation is typically small (less than 1 unit)^59,60, meaning that most soils are likely to remain in Regime II, except under large exogenous perturbations such as fertilizer or urine inputs³⁰. Third, although our soils span global taxonomic diversity reasonably well, they under-represent highly basic (pH > 8) or strongly acidic (pH < 4) soils (Extended Data Fig. 8m). Highly buffered basic soils may resist pH-induced regime transitions altogether.

This work provides a framework for linking community composition and physiology to ecosystem metabolism, offering a route to understanding how complex microbiomes respond to environmental change.

Methods

Sample collection, site description and soil characterization

Twenty topsoils were sampled across a range of pH values (4.7–8.32) from the Cook Agronomy Farm (Supplementary Table 1). The CAF (46.78° N, 117.09° W, 800 m above sea level) is a long-term agricultural research site located in Pullman, WA, USA. CAF was established in 1998 as part of the Long-Term Agroecosystem Research (LTAR) network supported by the US Department of Agriculture. Before being converted to an agricultural field, the site was zonal xeric grassland or steppe. CAF operates on a continuous dryland-crop rotation system comprising winter wheat and spring crops. CAF is located in the high rainfall zone of the Pacific Northwest region, and the soil type is classified as Mollisol (Naff, Thatuna and Palouse Series)⁶¹. Sampling occurred from 8 to 12 September 2022, post-harvest of spring crops, to reduce the impact of the plant on soil microbial communities. This period was during the dry season preceding the concentrated autumn rainfall.

Topsoils were collected from the eastern region of the CAF at a depth of 10–20 cm, other than Soil 1 and Soil 2 (depth of 0–10 cm). Eastern CAF practices ‘no tillage’, which eliminates soil inversion and mixing of the soil surface to 20 cm. The N fertilizer in this field has been primarily deep banded to depths of approximately 7–10 cm during the time of application, which creates a spike of nitrate resource in the soil depth we sampled. Each soil sample was obtained by cutting down through the hardened dry soil with a spade in a circular motion to create a cylindrical cake of soil of radius 10–20 cm until the desired depth. Each soil sample was not merged from sampling multiple replicates due to differences in pH in different locations. Samples were collected within a diameter of 500 m within the CAF to minimize the variation of edaphic factors other than pH. The large variation in soil pH comes from the long-term use of ammoniacal fertilizers and associated N transformations, which may undergo nitrification, resulting in the release of H ions. In combination, spatial pH variation increases with field-scale hydrologic processes that occur under continuous no tillage superimposed over a landscape that has experienced long-term soil erosion.

To maximize the coverage of sampled native pH, we used a portable pH meter (HI99121, Hanna Instruments) to directly measure and estimate the soil pH without having to make slurry on site to determine whether to sample the soil before sampling. For accurate pH values, pH was measured in the laboratory using a glass electrode in a 1:5 (soil to water w/w) suspension of soil in water (protocol of International organization for standardization (ISO) 10390:2021), where 7 g of soil was vortexed with 35 ml of Milli-Q filtered water, spun down, and filtered with 0.22 μm pore size. With these pH values, we selected 20 topsoil samples that are spread across a range of pH from 4.7–8.32 with intervals of 0.1–0.6. Twenty soil samples were sieved (<2 mm), removed of apparent roots and stones, and gravimetric water content was determined (105 °C, 24 h). The sieved samples were stored in the fridge for 0–3 months until the incubation experiment. For sequencing the initial community, subsamples were stored in −80 °C until the DNA extraction. The twenty soils were sent to the Research Analytical Laboratory (University of Minnesota, USA) to measure soil texture (soil particle analysis; sand, silt and clay composition), total carbon and nitrogen and cation exchange capacity. The soils were also sent to Brookside Laboratories for a standard soil analysis package (Standard Soil with Bray I phosphorus). Twenty soils had relatively similar edaphic properties: 5–9% gravimetric water content (g per g dry soil), soil texture of silty clay loam with 0% sand and 32–43% clay, and C:N ratio of 12–16 with 10–20 total carbon (mgC per g soil) (Supplementary Table 1).

To demonstrate the generality of our results from 20 CAF soils in other soils, we collected topsoil samples from 8 additional sites spanning natural preserves in IL (LaBagh Woods, Orland Grasslands, Moraine), IN (Pinhook Bog, Ambler Flatwoods) and the Sedgwick Reserve in CA (CL, EL and SCL grassland samples, managed by the University of California, Santa Barbara) in August 2024. Owing to the strong pH buffering capacity of the Orland, Moraine and SCL samples, we proceeded with five soils for which pH perturbations were feasible. Using the same methods as in the experiments conducted on the CAF soils (see below), we performed pH perturbation experiments, dynamic measurements of metabolites, and 16S rRNA sequencing. Of these, we analysed the metabolite dynamics of four soils (LaBagh, Pinhook, CLG13 and ELG13), as the Ambler soil had issues with nitrate retrieval under basic conditions during the Griess assay (for example, we could not reliably recover added nitrate using our extraction methods). Details of these soils and their characteristics are provided in Supplementary Table 1.

Soil rewetting, constructing soil pH titration curves and pH perturbation experiments

To mimic the autumn rainfall in the Pacific Northwest region and minimize the effect of spiking microbial activity by rewetting dry soils⁶², we rewetted the sieved soil for 2 weeks before the perturbation experiments at room temperature with sterile Milli-Q water at 40% of each soil’s water holding capacity. After resetting, a soil slurry was made by adding 2 mM sodium nitrate solution to the soil (2:1 w/w ratio of water to soil). The slurry was then transferred to 48-deep-well plates (2.35 ml of slurry per well) for incubation under anaerobic conditions (950 rpm, 30 °C) for 4 days. Anaerobic incubation was performed in a vinyl glove box (Coy Laboratory Products 7601-110/220) purged of oxygen with a 99%:1% N₂:CO₂ gas mixture, where the gaseous oxygen concentration was maintained below 50 ppm to prevent aerobic respiration⁵⁸.

To perturb the soil pH to desired levels, we constructed each soil’s pH titration curve for the 20 soils with varying native pH to know exactly how much acid or base to add to each soil sample. To do so, separate from the main pH perturbation experiment, we added 23 different levels of HCl (acid) or NaOH (base) in the slurry, final concentrations ranging from 0–100 mM HCl or NaOH. We colorimetrically measured the pH (see section below) immediately after and 4 days after adding each well’s designated amount of acid/base. Owing to natural soil’s buffering capacity, it takes 1–2 days to stabilize its pH level. Thus, we used the endpoint (after 4 days) pH measurements for all pH perturbations. We did a spline interpolation on the titration data points, plotting endpoint pH (y axis) against acid or base input (x axis), to compute how much HCl and NaOH needs to be added to the soil to obtain 13 different levels of pH with ~0.4 intervals ranging from pH 3 to 9, including the pH level without the addition of any acid or base. For Soil 19 and Soil 20, we had only 7 and 3 perturbed pH levels respectively, because the strong buffering capacity of these soils (native pH over 8) limited the range of pH perturbation. We additionally tested whether the anion of acid (Cl⁻) or the cation of base (Na⁺) had a distinctive effect on the nitrate reduction dynamics, which was not the case (for results, see Extended Data Fig. 2g and Supplementary Information, ‘Effect of base cation on nutrient release in soils’).

For the main pH perturbation experiment, the computed levels of concentrated HCl or NaOH were added to the slurry in the 48-deep-well plate with and without chloramphenicol treatment for each perturbed pH level in triplicate. The plates were immediately transferred to the shaking incubator (950 rpm in Fisherbrand Incubating Microplate Shakers 02-217-759, 3 mm orbital radius, 30 °C) inside the anaerobic glove box and incubated for 4 days. For chloramphenicol-treated samples, we added concentrated chloramphenicol solution to the slurry to obtain a final concentration of 1 g l⁻¹, after testing different doses to rule out the possibilities of unwanted growth due to chlormaphenicol resistance or degradation (Extended Data Fig. 2c and see Supplementary Information, ‘Effectiveness of chloramphenicol in preventing microbial growth in the soil’). To gauge the effect of the 2 mM nitrate, we had no-nitrate controls (0 mM nitrate) for both chloramphenicol-treated and untreated treatments in the unperturbed pH conditions. With antifungal cycloheximide controls (200 ppm) for all 20 soils, we confirmed that fungal activity minimally affects nitrate utilization dynamics (Extended Data Fig. 2a,b). We also confirmed that abiotic nitrate or nitrite reduction does not occur by measuring metabolic dynamics of autoclaved soil (120 °C, 99 min, autoclaved 5 times every 2 days; Extended Data Fig. 2d). To offset the effect of increasing metabolite concentration due to evaporation throughout the 4-day incubation period, we used the wells with just 2 mM nitrate, nitrite and ammonium solutions to correct for evaporation in the slurry samples for every time point. The values of the gravimetric water content of each soil were taken into account to correct for the dilution of 2 mM nitrate due to moisture in the soil. After the incubation, the plates were stored in −80 °C for sequencing endpoint communities.

Time-series slurry sampling, extraction, and colorimetric assays to measure nitrate, nitrite, ammonium, WSOC and pH

To obtain the metabolic dynamics, we subsampled 60 μl of the slurry into 96-well plates 10 times throughout 4 days (0, 4, 8, 19, 25, 31, 43, 55, 67 and 91 h), where the initial time point (T₀) is the time of pH perturbation and the start of anaerobic incubation. To measure nitrate and nitrite dynamics, extracts were prepared from the sampled slurries by adding and vortexing for 2 min with 90 μl of 3.33 M KCl solution (final concentration of 2 M KCl) and centrifuging at 4,000 rpm for 5 min. The supernatant was filtered at 0.22 μm with a vacuum manifold to remove soil particles that could interfere with colorimetric assays. Concentrations of nitrate and nitrite in the extracts were determined colorimetrically using the Griess assay⁶³ and vanadium (iii) chloride reduction method, following the protocol outlined previously⁵⁸. We confirmed that 95%–99% of the nitrate in the soil can be accurately retrieved and detected using this method, as verified by nitrate spike-in and extraction experiments in the soil. For all 20 CAF soils, the ammonium dynamics were measured colorimetrically using the salicylate–hypochlorite assay from the soil extracts⁶⁴. Chloramphenicol treatments in the samples led to consistent detection of 0.5 mM ${{\rm{NH}}}_{4}^{+}$ due to its N-H moiety. The salicylate–hypochlorite assay is also affected by the amount of base (NaOH) in the samples, resulting in slightly lower detection of chloramphenicol in the chloramphenicol-treated samples (0.45 mM ${{\rm{NH}}}_{4}^{+}$ in 100 mM NaOH perturbations). Taking advantage of these control measurements, we used the constant ${{\rm{NH}}}_{4}^{+}$ levels in the controls without 2 mM ${{\rm{NO}}}_{3}^{-}$ (no-nitrate controls) in the chloramphenicol-treated samples for each soil to offset the NaOH effect in the non-chloramphenicol samples and subtracted ${{\rm{NH}}}_{4}^{+}$ levels caused by chloramphenicol in chloramphenicol-treated samples.

For WSOC measurements, we subsampled 60 μl of the slurry into 96-well plates at T₀ and endpoint (4 days). Then, soil extracts were prepared by adding, vortexing with 90 μl Milli-Q water, centrifuging at 4,000 rpm for 5 min, and 0.22 μm filtering the supernatant. Concentrations of the organic carbon in the supernatant were measured colorimetrically by the Walkley–Black assay, which uses dichromate in concentrated sulfuring acid for oxidative digestion⁶⁵. We subtracted 0.4 mgC ml⁻¹ from the chloramphenicol-treated samples because chloramphenicol gave rise to a measured value of 0.4 mgWSOC ml⁻¹ without additional carbon. For pH measurements, we subsampled 100 μl of the slurry into 96-well plates at T₀ and the endpoint. Then, soil extracts were prepared by adding, vortexing with 150 μl KCl solution (final concentration of 1 M KCl), centrifuging at 4,000 rpm for 5 min, and 0.22 μm filtering the supernatant. The pH of the 120 μl supernatant was determined colorimetrically by adding 4 ul of the multiple indicator dye mixture via the protocol described previously⁶⁶. The reason we used 1 M KCl method for pH measurement (ISO 10390:2021) was that, contrary to the KCl method, the H₂O method (using water instead of 1M KCl) resulted in a highly yellow colouration of the supernatants in strong basic perturbed samples, which interfered with the wavelength of the colorimetric pH assay. For samples of pH outside the range of the assay (below pH 3 and above pH 9), we used a pH micro-electrode (Orion 8220BNWP, Thermo Scientific). We calculated the endpoint perturbed pH as the average pH of the three biological replicate endpoint samples.

DNA extraction with internal standards, library preparation and 16S rRNA amplicon sequencing

We performed 16S amplicon sequencing on half of all samples: 10 (soils 3, 5, 6, 9, 11, 12, 14, 15, 16, 17; Supplementary Table 1) out of 20 soils were sequenced before perturbation and at the endpoint in both chloramphenicol-treated and untreated conditions, totalling 1,085 amplicon sequencing measurements. Genomic DNA was extracted from 500 μl aliquots in a combined chemical and mechanical procedure using the DNeasy 96 PowerSoil Pro Kit (Qiagen). Extraction was performed following the manufacturer’s protocol, and extracted DNA was stored at −20 °C. To estimate the absolute abundance of bacterial 16S rRNA amplicons, we added known quantities of genomic DNA (gDNA) extracted from Escherichia coli K-12 and Parabacteroides sp. TM425 (samples sourced from the Duchossois Family Institute Commensal Isolate Library, Chicago, IL, USA) to the slurry subsamples before DNA extraction. Equal concentrations of gDNA from these two strains were added. Both strains have identical rRNA copy numbers of 7 and comparable genome sizes of approximately 5,000 kb. DNA library preparation was performed using the 16S Metagenomic Sequencing Library Preparation protocol with a 2-stage PCR workflow (Illumina). The V3–V4 region was amplified using forward primer 341-b-S-17 (CCTACGGGNGGCWGCAG) and reverse primer 785-a-A-21 (GACTACHVGGGTATCTAATCC)⁶⁷. We confirmed using gel electrophoresis that the negative samples containing all reagents did not show visible bands after PCR amplification. Sequences were obtained on the Illumina MiSeq platform in a 2× 300-bp paired-end run using the MiSeq Reagent Kit v3 (Illumina) with 25% PhiX spike-ins. A standardized 10-strain gDNA mixture (MSA-1000, ATCC) was sequenced as well to serve as a positive control, which was confirmed to have relatively uniform read counts after assigning taxa.

Model and fitting

Consumer-resource model

Consider a consumer-resource model with one consumer variable (functional biomass x(t), biomass) and two resource variables (nitrate A(t) and carbon-nutrient C(t), mM), which evolves in time (t, day). The ordinary differential equations of the consumer-resource model can be expressed as:

$$\begin{array}{c}\mathop{A}\limits^{.}(t)=-{r}_{A}x(t)\frac{A(t)}{A(t)+{K}_{A}},\\ \mathop{C}\limits^{.}(t)=-{r}_{C}x(t)\frac{C(t)}{C(t)+{K}_{C}},\\ \mathop{x}\limits^{.}(t)=\gamma x(t)\frac{A(t)}{A(t)+{K}_{A}}\frac{C(t)}{C(t)+{K}_{C}}.\end{array}$$

(1)

The first two equations represent the resource consumption rates, which are determined by the functional biomass (x (g biomass)), the maximum consuming rates per unit biomass (r_A and r_C (mM per g biomass per day)), and the Monod functions (A/(A + K_A) and C/(C + K_C) (dimensionless)). Here we assume the affinities (K_A and K_C (mM)) to be fixed and small. So the Monod functions are 1 when A ≫ K_A or C ≫ K_C, and 0 when A→0 or C→0. The third equation represents the growth of functional biomass, which is determined by the maximum growth rate per biomass (γ (day⁻¹)) and the multiplication of two Monod terms indicating the fact that nitrate and carbon are non-substitutable (electron acceptor and donor, respectively). The growth is exponential (x(t) = x(0)e^γt) when both A ≫ K_A or C ≫ K_C, but growth stops when either A→0 or C→0. Therefore, in this model, the growth of biomass is limited by both resources, but the consumption of one resource can continue when the other resource runs out and the biomass growth stops. For example, we believe this happens when C→0 in Regime II and the consumption of A continues (Fig. 2). For a discussion of models that consider multiple biomasses and carbon sources, see Supplementary Information ‘Justifying the effective 1-biomass model despite the diversity of denitrifying taxa’.

Solution for nitrate dynamics

To find the solution for nitrate dynamics, we rescale the equations by combining parameters: $\widetilde{x}={r}_{A}\,x$, $\widetilde{C}=C{r}_{A}/{r}_{C}$, $\widetilde{{K}_{C}}={K}_{C}{r}_{A}/{r}_{C}$. Therefore, the equations become:

$$\begin{array}{c}\mathop{A}\limits^{.}(t)=-\widetilde{x}(t)\frac{A(t)}{A(t)+{K}_{A}}\\ \mathop{\widetilde{C}}\limits^{.}(t)=-\widetilde{x}(t)\frac{\widetilde{C}(t)}{\widetilde{C}(t)+{\widetilde{K}}_{C}}\\ \mathop{\widetilde{x}}\limits^{.}(t)=\gamma \widetilde{x}(t)\frac{A(t)}{A(t)+{K}_{A}}\frac{\widetilde{C}(t)}{\widetilde{C}(t)+{\widetilde{K}}_{C}}\end{array}$$

(2)

In the rescaled equations (2), the units of parameters and variables are: $[\widetilde{x}]$ (mM day⁻¹) and $[\widetilde{C}]=[\widetilde{{K}_{C}}]$ (mM). Therefore, the solution of nitrate dynamics only depends on three parameters (γ, K_A and ${\widetilde{K}}_{C}$) and three initial conditions (A₀, $\widetilde{C}(0)$ and $\widetilde{x}(0)$). Since the affinities are very small (K_A ≈ 0.01 mM, ${\widetilde{K}}_{C}\approx 0.01\,{\rm{mM}}$), the solution of biomass activity approximately equals $\widetilde{x}=\widetilde{x}(0){{\rm{e}}}^{\gamma t}$ before the time at which growth stops t* (Fig. 2). So the resource dynamics before t* are approximately $A\,=$ ${A}_{0}-\widetilde{x}(0)({{\rm{e}}}^{\gamma t}-1)/\gamma$ and $\widetilde{C}=\widetilde{C}(0)-\widetilde{x}(0)({{\rm{e}}}^{\gamma t}-1)/\gamma$. Accordingly, the time at which growth stops is given by ${t}^{* }=\log (\min ({A}_{0},\widetilde{C}(0))\gamma /\widetilde{x}(0)+1)/\gamma$. If $\widetilde{C}(0) < {A}_{0}$, the nitrate dynamics after t* and before running out are given by $A=A({t}^{* })\,-\,(\gamma \widetilde{C}(0)\,+$ $\widetilde{x}(0))(t-{t}^{* })$. As a result, the nitrate consumption rate after t* is $\gamma \widetilde{C}(0)$ larger than the initial rate $\widetilde{x}(0)$. Therefore, the two key parameters of the model are $\widetilde{x}(0)$ and $\gamma \widetilde{C}(0)$ which are both rates (in mM day⁻¹).

Least-squares fitting scheme

To infer the model parameters from the metabolite measurement, we use the least-squares fitting scheme to find the closest dynamic curves to the time-series data. Our metabolite measurement including the time points (${\underline{t}}^{-}=[{t}_{1}^{-},{t}_{2}^{-},...,{t}_{N}^{-}]$) and nitrate amount (${\underline{a}}^{-}=[{a}_{1}^{-},{a}_{2}^{-},...,{a}_{N}^{-}]$) for each CHL- sample, and the measurements of ${\underline{t}}^{+}$ and ${\underline{a}}^{+}$ for a corresponding chloramphenicol-treated sample. We set up the loss function as the mean-squared error:

$$L=\frac{1}{2N}\left(\mathop{\sum }\limits_{k=1}^{N}{(A({t}_{k}^{-})-{a}_{k}^{-})}^{2}+\mathop{\sum }\limits_{k=1}^{N}{({A}^{c}({t}_{k}^{+})-{a}_{k}^{+})}^{2}\right).$$

(3)

Here, the functions A(t) and A^c(t) are theoretical solutions of the consumer-resource model (2) for non-chloramphenicol and chloramphenicol conditions, respectively. Because the nitrate dynamics A(t) and A^c(t) are determined by the parameter set ${\Theta }=\{\widetilde{x}(0),\widetilde{C}(0),{A}_{0},{A}_{0}^{c},\gamma ,{K}_{A},{\widetilde{K}}_{C}\}$, we minimize the loss function L(Θ) to get the optimal model parameters Θ*. We note to the readers that three parameters are fixed (γ = 4.8 day⁻¹, ${K}_{A}={\widetilde{K}}_{C}=0.01\,{\rm{mM}}$) as justified by the sensitivity analysis in the following paragraph. Note that these parameters were globally fixed across all the data for CAF soils. For other soils (IL, IN, CA), γ is fixed within each site but may vary for soils from different sites, the value of which is chosen in the regime where fit quality is insensitive to γ (see next section). The optimization algorithm is the interior-point method, which is built in the MATLAB fmincon function. The codes and data are available at the Open Science Framework (https://doi.org/10.17605/OSF.IO/CTF8K). The fitting errors over all samples are shown in Extended Data Fig. 3a,b, in which the root-mean-squared error (RMSE, $\sqrt{L({{\Theta }}^{* })}$) and the error per data point ($| A({t}_{k}^{-})-{a}_{k}^{-}|$ or $| {A}^{c}({t}_{k}^{+})-{a}_{k}^{+}|$) are normalized by the input value of nitrate (2 mM).

Sensitivity analysis on model parameters

Here we justify the decision to globally fix γ, K_A and ${\widetilde{K}}_{C}$. We analysed the sensitivity of γ, K_A and ${\widetilde{K}}_{C}$ on simulated dynamic data. To reflect the three typical dynamics (regimes) observed from the measurement, we simulated three nitrate curves by setting up the initial conditions to be $\widetilde{x}(0)=0.01,0.1\,$${\rm{and}}\,0.001\,{\rm{mM}}\,{{\rm{day}}}^{-1}$ and $\widetilde{C}(0)=0.005,0.05\,{\rm{and}}\,2\,{\rm{mM}}$, respectively. Other parameters are given by ${A}_{0}={A}_{0}^{c}=2\,{\rm{mM}}$, ${K}_{A}={\widetilde{K}}_{C}=0.01\,{\rm{mM}}$, γ = 4 day⁻¹. We then used different fixed parameter values to fit the three examples. In the first row of Extended Data Fig. 3c, we used different fixed γ values—from γ = 2 day⁻¹ to γ = 6 day⁻¹—to fit three simulations. We demonstrate very small mismatches (RMSE <5%) from these variations of parameter values, which are almost invisible in Regime I and Regime II fittings. In the second and the third row of Extended Data Fig. 3c, we use different fixed K_A and ${\widetilde{K}}_{C}$ values—from 10⁻⁴ mM to 1 mM—to fit three simulations. When K_A < 0.1 mM or ${\widetilde{K}}_{C} < 0.1\,{\rm{mM}}$, the mismatches were again very small (RMSE <1%) and invisible. These results indicate that the fixed values of γ, K_A and ${\widetilde{K}}_{C}$ are insensitive in large ranges.

Determination of regime boundary with model parameters

To define the regime boundaries, we examined the distributions of each parameter’s value. $\widetilde{x}(0)$ had a bimodal distribution (Extended Data Fig. 3d). This bi-modality becomes more evident when we separately observe its distribution from the left half (perturbed pH < 4) and right half (perturbed pH > 6) of the parameter space displayed in the perturbed pH versus native pH grid in Fig. 3c (Extended Data Fig. 3e). Therefore, we set the threshold for the $\widetilde{x}(0)$ boundary where these two modes are separated ($\widetilde{x}(0)$ = 0.05). The distribution of $\gamma \widetilde{C}(0)$ exhibited a significant mode around 0, prompting us to set the threshold ($\gamma \widetilde{C}(0)$ = 1.5) at the tail region, where the $\gamma \widetilde{C}(0)$ threshold also separated the Regime III samples in the top left quadrant of the $\widetilde{x}(0)$ versus $\gamma \widetilde{C}(0)$ scatter plot (Fig. 3a). The separation of Regime I and Regime II data points may not be clear cut in the $\widetilde{x}(0)$ versus $\gamma \widetilde{C}(0)$ scatter plot (Fig. 3a).

Sequence data analysis

Sequencing data preprocessing and assigning taxonomy to ASVs with DADA2

Raw Illumina sequencing reads were stripped of primers, truncated of Phred quality score below 2, trimmed to length 263 for forward reads and 189 for reverse reads (ensuring a 25-nucleotide overlap for most reads), and filtered to a maximum expected error of 4 based on Phred scores; this preprocessing was performed with USEARCH v.11.0⁶⁸. The filtered reads were then processed with DADA2 v.1.18 following the developers’ recommended pipeline⁶⁹. In brief, forward and reverse reads were denoised separately, then merged and filtered for chimeras. For greater sensitivity, ASV inference was performed using the DADA2 pseudo-pooling mode, pooling samples by soil. After processing, the sequencing depth of denoised samples was 10⁴–10⁶ reads per sample. Low-abundance ASVs were dropped (≤10 total reads across all 1,085 samples), retaining 34,696 ASVs for further analysis. Taxonomy was assigned by DADA2 using the SILVA database v.138.1, typically at the genus level, but with species-level attribution recorded in cases of a 100% sequence match. R scripts used for DADA2 sequencing data preprocessing were deposited at the Open Science Framework (https://doi.org/10.17605/OSF.IO/CTF8K).

Computing absolute abundance with internal standards of each ASV per sample

As an internal control, we verified that the ASVs corresponding to the two internal standard genera Escherichia–Shigella and Parabacteroides were highly correlated with each other as expected (Pearson correlation (ρ) = 0.94). These ASVs were removed from the table and combined into a single reference vector of ‘spike-in counts’. The spike-in counts constituted 8.9 ± 8.8% of the total reads in each sample. For downstream analysis, the raw ASV counts in a sample were divided by the spike-in counts of the internal standard per sample to obtain the absolute abundance of the ASV in a sample. Total biomass per sample was obtained by dividing the total raw read counts with the spike-in counts of the sample.

Differential abundance analysis to identify enriched ASVs

We conducted differential abundance analysis to statistically determine which ASVs were significantly enriched in the nutrient-limiting regime (Regime II) or the resurgent growth regime (Regime III), respectively. To do so, we identified enriched ASVs for each perturbed pH condition in each native soil comparing endpoint chloramphenicol-untreated samples with endpoint chloramphenicol-treated samples. For each native soil, we then compiled a list of enriched ASVs by aggregating a union set of enriched ASVs across perturbed conditions that belong to Regime II (or Regime III). To remove ASVs that could be false-positive nitrate reducers, we similarly performed differential abundance analysis to identify ASVs that are enriched in no-nitrate controls (nitrate⁻) by comparing endpoint chloramphenicol-untreated (−Chl, nitrate⁻) samples with endpoint chloramphenicol-treated (+Chl, nitrate⁻) samples. This filtering was done when inferring nitrate reducer biomass (Extended Data Fig. 4e,f) and inferring the Regime III strains (Extended Data Fig. 6c–e). For each native soil, we only had nitrate⁻ controls for the condition without acidic/basic perturbation. We assumed that these enriched ASVs in no-nitrate conditions (NNresponders) without acid/base perturbation would also be false-positive nitrate reducers in other acidic or basic perturbation levels. For each native soil, we filtered out these false-positive NNresponders from the aggregated list of Regime II (or Regime III) enriched ASVs.

To identify the ASVs enriched for each perturbed pH level, it was necessary to determine what change in recorded abundance constitutes a significant change, relative to what might be expected for purely stochastic reasons. The relevant null model would combine sampling and sequencing noise with the stochasticity of ecological dynamics over a four-day incubation, and cannot be derived from first principles. However, since all measurements were performed in triplicate with independent incubations, the relevant null model can be determined empirically. The deviations of replicate–replicate comparisons from 1:1 line were well-described by an effective model combining two independent contributions, a Gaussian noise of fractional magnitude c_frac and a constant Gaussian noise of magnitude c₀ reads, such that repeated measurements (over biological replicates) of an ASV with mean abundance n counts are approximately Gaussian-distributed with a standard deviation of $\sigma ({c}_{0},{c}_{\text{frac}})=\sqrt{{({c}_{\text{frac}}n)}^{2}+{c}_{0}^{2}}$ counts. In this expression, c_frac was estimated from moderate-abundance ASVs (>50 counts) for which the other noise term is negligible; and c₀ was then determined as the value for which 67% of replicate–replicate comparisons are within ±σ(c₀, c_frac) of each other, as expected for 1-sigma deviations. This noise model was inferred separately for each soil and each perturbed pH level, as the corresponding samples were processed independently in different sequencing runs. For example, the parameters in Soil 11 were c_frac = 0.21 ± 0.04 and c₀ = 4.5 ± 0.7 counts (Extended Data Fig. 9i).

The model was used to compute the z-scores for the enrichments of absolute ASV abundances in non-chloramphenicol treatments against chloramphenicol-treated controls (three independent z-scores from three replicate pairs; rep1–rep1, rep2–rep2 and rep3–rep3). The median z-score was assigned to each ASV for each perturbed condition. In consideration of ASVs with 0 read count in samples with or without chloramnphenicol treatment, all raw ASV counts were augmented by a pseudocount of 0.5 and divided by the per-sample spike-in counts, yielding values that can be interpreted as the absolute biomass of each taxon (up to a factor corresponding to the copy number of the 16S operon), measured in units where 1 means as many 16S fragments as the number of DNA molecules in the spike-in. Significantly enriched ASVs were identified in each perturbed condition as those with z-scores greater than z = Φ⁻¹(1 − α/2n_ASV), where Φ⁻¹(x) is the inverse cumulative distribution function of the standard normal distribution, α = 0.05, and n_ASV is the number of non-zero ASVs in a given sample. This critical z-score (z = 4.2, when n_ASV = 2,000 for enriched ASVs and z = 4.3, when n_ASV = 2,500 for filtering no-nitrate responders (NNresponders)) corresponds to a two-tailed Bonferroni-corrected hypothesis test at significance level α under the null hypothesis that counts in the chloramphenicol-treated and untreated conditions are drawn from the same distribution. These analyses were performed using custom MATLAB (Mathworks) and R scripts, which are deposited at the Open Science Framework (https://doi.org/10.17605/OSF.IO/CTF8K); for additional technical details, the reader is referred to the detailed comments in these scripts.

NMF analysis on phylum-level growth folds

To analyse the abundance change at the phylum level, we compute the growth fold of each phylum at each condition. For each phylum, we compute the absolute abundance by aggregating the abundances of all ASVs within that phylum. Taking chloramphenicol-treated abundance (Abs⁺) as the reference abundance and non-chloramphenicol abundance (Abs⁻) as the endpoint abundance (where Abs denotes taxon abundance normalized to internal standard), the logarithm of the growth fold for phylum i and condition j is given by ${g}_{ij}=\log ({{\rm{Abs}}}_{ij}^{-}+1{0}^{-3})-\log ({{\rm{Abs}}}_{ij}^{+}+1{0}^{-3})$. Note that we use chloramphenicol-treated abundance as reference instead of the initial abundance (at T₀), to account for any effects on read counts unrelated to growth which would be common between chloramphenicol-treated and untreated conditions (for example, direct effect of acid/base addition), allowing us to focus only on growth-mediated abundance changes. We also set all negative g_ij to 0 since we are focusing on growth. For all 130 conditions (10 soils × 13 perturbations) and 40 phyla, the phylum-level growth folds G is a 40 × 130 matrix. For each phylum, the row vector ${\overrightarrow{g}}_{i}$ represents how it grows under different conditions (see Extended Data Fig. 5b for the growth vectors of the first 10 phyla). In order to reduce the dimensionality of the growth matrix and extract the main features of the growth vectors, we use NMF to decompose the growth matrix G = W × H into 2 factors, which retain 93.36% of the original G matrix variation. Here, the feature matrix H is of size 2 × 130, and the 2 rows ${\overrightarrow{h}}_{1}$ and ${\overrightarrow{h}}_{2}$ are 2 modes of growth vectors (shown in Extended Data Fig. 5c). Therefore, the growth vector of phylum i is thus decomposed as ${\overrightarrow{g}}_{i}\approx {w}_{i1}{\overrightarrow{h}}_{1}+{w}_{i2}{\overrightarrow{h}}_{2}$, while the weights w_i1 and w_i2 are from the 40 × 2 weight matrix W. The weights of all 40 phyla are plotted in Extended Data Fig. 5c, showing that Bacillota is mostly composed of the second mode ${\overrightarrow{h}}_{2}$ and other phyla are mostly composed of the first mode ${\overrightarrow{h}}_{1}$. Additionally, Bacteroidota and Pseudomonadota show the most significance in the first mode.

Genotyping enriched ASVs with PICRUSt2

To understand what traits make resurgent growth strains unique, we used PICRUSt2 v.2.5.2²⁸ to infer putative genotypes of the enriched ASVs in the resurgent growth regime (Regime III) (Extended Data Fig. 6g). Using the script place_seqs.py from PICRUSt2, we matched the representative 16S rRNA sequences of each ASV to PICRUSt2’s curated reference genome database (multiple sequence alignment). Then, using the hsp.py script from PICRUSt2 with default parameters, we predicted KEGG orthologues (KO) abundance of each ASV with the matched reference genome (hidden-state prediction). To narrow down to KOs or genes related to denitrification and DNRA, we focused on nitrate reductase in denitrification (narG/K00370, narH/K00371, narI/K00374, napA/K02567 and napB/K02567) and nitrite reductase to ammonium (nirB/K00362, nirD/K00363, nrfA/K03385 and nrfH/K15876). To track which KOs were enriched at which pH in the 89 families used in the peak pH analysis (see Supplementary Information, ‘Determination of peak pH for each family’), we summed the relative abundance (reads/total reads of each perturbed pH level in non-chloramphenicol condition) of the ASVs belonging to each family that possessed at least 1 predicted gene respectively for narGHI, napAB, nirBD and nrfAH. Then, we plotted their relative abundance values across pH for all soils, indicated by the intensity of the point’s colours (Extended Data Fig. 6g).

Taxonomy of growing strains in Regime III varies with soil native pH

To further investigate whether the taxonomic identity of resurgent growth (Regime III) strains varies across natural pH environments, we performed a regularized regression analysis to see if we can predict the native pH level of the source soil from the presence or absence of taxa that grow in Regime III at the ASV, species, genus, family, or higher taxonomic levels. The resurgent growth strains were determined by the differential abundance analysis as described previously. We used the sequencing data to build a matrix where the rows are samples (including three biological replicates) belonging to the resurgent growth regime (Regime III), where each row has a corresponding native pH value of the original soil. There are ten source soils with different native pH levels, and each soil has 3–6 pH-perturbed samples (replicates) of which metabolite dynamics are classified as Regime III. The matrix’s columns are different taxa belonging to the identified Resurgent growth strains, either in ASV, species, genus, family or higher levels. Each element of the matrix is 0 if absent and 1 if present in the sequencing data of the sample. Because the presence and absence of taxa can randomly depend on the random sampling depth of each sample, we test varying threshold values (0, 0.001 or 0.005 relative abundance) to call the taxa present if their relative abundance is greater than the threshold (effects shown in Extended Data Fig. 8k).

The regularized regression was performed to predict the native pH of the source soil of the samples from the presence and absence of taxa using only additive terms and LASSO regularization to avoid overfitting⁷⁰. To estimate the regularization hyperparameter, tenfold cross-validation was performed on the samples from ten different soils with different native pH levels. All models were fit using the package glmnet in R v.4.1.4. To make predictions of the native pH, we used two strategies. First, ‘in-sample’ predictions used all available data points to fit the regression coefficients and predicted native pH using those coefficients. Second, to ask whether we can still predict the native pH without the model seeing the samples belonging to that specific native pH level, we implemented a leave-one-soil-out (LOSO) procedure, where all the perturbed samples from one native soil were left out as a test set, and the model was trained on the remaining data to fit the regression coefficients. Then, we used the model to predict the native pH of the left-out samples (out-of-sample prediction). The observed versus predicted pH values are shown in the scatter plots (Extended Data Fig. 8h). The prediction quality (R²) was computed using the mean predicted and mean observed native pH levels for each soil (for different taxonomic levels and prediction strategies, see Extended Data Fig. 8j,k; negative R² values indicate the predictions are worse than just predicting the pH as the mean predicted pH). To ascertain that our high prediction quality was not a random artefact, we randomly permuted the native pH values of our soils 1,000 times and then predicted in-sample the native pH to obtain 1,000 R² values under permutation and used this distribution to compute a P value (0.012) for the observed R² obtained with our true native pH values (black arrow, Extended Data Fig. 8l).

Testing the effect of different bases and salts on nutrient release

To see the effects of different bases (NaOH and KOH) on nitrate reduction dynamics, we added different concentrations of NaOH and KOH (final concentration of 0, 8, 16 and 24 mM in the slurry), following the same protocol previously described (without chloramphenicol), to measure the nitrate and nitrite dynamics (Extended Data Fig. 2g, using Soil 6; Supplementary Table 1). In addition, to test the effects of Na⁺, K⁺ and Cl⁻ separately, we added different concentrations of salts (NaCl and KCl) (without chloramphenicol and without adding any acid or base) and measured the metabolite dynamics (Extended Data Fig. 2g).

Nutrient amendment experiments with slurries

To experimentally determine what nutrient was limiting growth in the nutrient-limiting regime, we conducted nutrient amendment experiments respectively with glucose, succinate, sodium acetate, ammonium chloride (NH₄Cl), monosodium phosphate (NaH₂PO₄) and sodium sulfate (Na₂SO₄) (for results, see Fig. 5d and Extended Data Fig. 2h). Among them, succinate (pK_a = 4.21 and 5.64, 25 °C), acetate (pK_a = 4.76, 25 °C), and phosphate (pK_a = 2.2, 7.2 and 12.4, 25 °C) were strong candidates for the limiting nutrient according to our soil nutrient release hypothesis, due to their anionic nature in mid-range pH (5–7). The incubation was conducted following the same protocol using Soil 6 (Supplementary Table 1) without chloramphenicol and without adding any acid/base. Concentrations were either in mM C, mM N, mM S or mM P with final concentrations in slurry varying from 0 to 5 mM, each in triplicate. Because we have previously tested the effect of Na⁺ and Cl⁻ to be negligible in nitrate dynamics, the effect of these amendments can be attributed solely to C, N, S or P nutrients other than Na⁺ and Cl⁻.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.