Two Weeks in a Jar: How Long Soils Need to Stabilize Before Experiments

When dried soil is rewetted in the laboratory, it does not behave quietly. Microbes surge into activity, carbon dioxide pours out, and pools of ammonium and dissolved organic matter swing wildly before settling down. A new study of twelve contrasting Australian soils has now put numbers on that settling process, concluding that a pre-incubation period of approximately 10 to 14 days after rewetting offers a robust compromise between letting soils recover from disturbance and avoiding the artificial changes that come with keeping them too long in a jar.

The research, published in Frontiers in Soil Science, was carried out by Chelsea K. Janke of the Commonwealth Scientific and Industrial Research Organisation in Adelaide and colleagues, with co-authors affiliated also with the University of Queensland. The team set out to answer a deceptively simple question that has long been answered by convention rather than measurement: how long should soils be pre-incubated after sampling, drying, and storage before process-based experiments begin? The answer matters because nearly every controlled soil study depends on this step, yet the optimal duration has remained poorly defined across soil types.

The disturbance in question is real and well documented. During sampling, transport, and storage, soils experience physical and chemical disruptions that can alter microbial communities and nutrient availability. Drying is particularly disruptive. When dried soils are rewetted, they emit a pulse of carbon dioxide known as the Birch Effect, driven primarily by increased microbial respiration. Proposed mechanisms include the accumulation of soluble organic carbon during drought that is rapidly mineralized on rewetting, desiccation-induced cell lysis that releases cellular metabolites, disruption of macroaggregates that exposes previously protected carbon, and shifts in microbial communities driven by moisture fluxes. Which mechanism dominates appears to depend on the soil, with metabolite accumulation controlling the pulse in sandy soils and aggregate disruption driving it in loamy and clayey soils.

To quantify stabilization dynamics, the researchers collected twelve soils from key Australian agricultural regions, spanning Vertisols from Queensland, New South Wales, and Victoria; Luvisols from Victoria and Western Australia; Solonetz from South Australia and Western Australia; a Lixisol, a Calcisol, and a Regosol. Soils were air-dried, sieved to less than 2 millimeters, dried again at 40 degrees Celsius to ensure uniform status, and packed into cores at 40 percent of plant available water capacity. The cores were incubated at 22 degrees Celsius for 35 days, with destructive sampling at 0, 7, 14, 23, and 35 days, duplicated at each time point, yielding 120 experimental units across the twelve soils.

The measurements captured a consistent story of disturbance and recovery. Ammonium concentrations were elevated at the start of incubation, reaching 1 to 4 milligrams per kilogram in most soils, then declined rapidly within the first 7 days to between 0.05 and 0.7 milligrams per kilogram. This decline is consistent with rapid microbial immobilization and nitrification triggered by rewetting, an interpretation supported by concomitant increases in nitrate over the first week. Importantly, ammonium never returned to its initial values over the 35-day incubation, suggesting that the elevated starting concentrations were artefacts of drying and storage rather than representative of field conditions. From day 7 onward, ammonium levels stabilized at values more comparable to fresh soil measurements.

Nitrate told a more cautionary tale. Unlike ammonium, nitrate concentrations continued to increase over time in most soils, even where temporary stabilization occurred between 7 and 14 days. This ongoing accumulation highlights a risk of longer pre-incubation: without plant uptake, extended incubation can progressively shift soils away from field-relevant nitrogen conditions. The researchers argue that this reinforces the need to balance microbial recovery against the avoidance of progressive nutrient transformation during the pre-incubation window.

Dissolved organic nitrogen and carbon pools followed broadly similar trajectories, decreasing between days 0 and 7, rising to day 14, and then stabilizing, generally between 7 and 14 days, although some soils required up to 21 days. Free amino acids, which represent the fast-cycling low molecular weight fraction of dissolved organic nitrogen, stabilized earlier, within 7 to 14 days across all soils, making them a sensitive indicator of microbial recovery. Microbial biomass carbon and nitrogen showed an initial growth phase over the first week, stimulated by increased substrate availability, followed by a moderate decline or stabilization between 7 and 14 days. Stabilization generally occurred slightly earlier for microbial biomass nitrogen than for microbial biomass carbon. Notably, the stabilized phase was often transient, with many soils showing declines in biomass nitrogen and, in some cases, increases in biomass carbon beyond that point, suggesting ongoing shifts in community structure or resource availability.

Respiration presented the trickiest signal to interpret. Carbon dioxide emission rates were high immediately after rewetting, between 9 and 26 micrograms of CO2-carbon per hour, and declined rapidly within the first few days, after which rates remained relatively stable at 1 to 9 micrograms per hour. Yet when the researchers formally estimated stabilization timing, 58 percent of soils reached their most stable respiration phase only at day 23. The authors attribute this to the method’s sensitivity: soils reporting later stabilization did so only because changes in emission between days 23 and 28 were marginally smaller than at other intervals, even though all differences from day 7 onward were below 2.5 micrograms per hour. Relying on respiration alone, they conclude, may overestimate the required pre-incubation duration.

When all measured variables were considered together, stabilization times converged within a remarkably narrow window across all twelve soils, with a mean of 12.7 days and a median of 14 days. Clustering analysis identified four groups of soils, but differences between clusters were modest, with median stabilization times ranging only from 10.5 to 14 days. More strikingly, the study found no significant relationships between stabilization behavior and measured soil properties, including pH, texture, or organic carbon content, nor did geographic or bioregional factors explain the clustering patterns. This outcome contradicted the researchers’ own hypothesis that soil characteristics such as clay content would influence stabilization duration through their effects on the magnitude of the Birch response. The authors suggest that the lack of clear drivers likely reflects limited statistical power from the relatively small number of soils, as well as the overall similarity in stabilization times observed.

Based on this convergence of multiple indicators, the authors recommend a pre-incubation period of approximately 10 to 14 days as sufficient for most soils to reach a stable and representative state while minimizing artefacts of both under- and over-incubation. They are candid about the limitations of the evidence: the study did not include direct comparisons with fresh field-moist soils, so it cannot fully quantify how closely post-incubation conditions approximate true field states, and individual soil-level stabilization estimates rest on two replicates per timepoint, making the recommendation a consistent cross-soil convergence rather than a highly resolved estimate for any single soil. Future work comparing dried-rewetted soils with fresh counterparts, increasing replication, and testing a broader range of soils would strengthen confidence in the guideline. Even so, the finding offers laboratory researchers something they have long lacked: a measured, defensible window for letting disturbed soils settle before the real experiment begins.

Subject of Research: Determining the optimal pre-incubation duration for dried and rewetted soils in laboratory experiments

Article Title: Optimizing pre-incubation periods for process-based studies in diverse soils

Article References: Janke, C. K., Carter, T., Szarvas, S., Maxson, C., & Farrell, M. (2026). Optimizing pre-incubation periods for process-based studies in diverse soils. Frontiers in Soil Science, 6, 1923658. https://doi.org/10.3389/fsoil.2026.1923658

Image Credits: AI Generated

DOI: 10.3389/fsoil.2026.1923658

Keywords: soil science, pre-incubation, rewetting, Birch Effect, microbial biomass, respiration, nitrogen, dissolved organic carbon, nitrate, incubation experiments, soil incubation, Australia