One field, one season: what the manure-depth trial actually shows

A single-season trial on saline silty loam in Jilin Province, China, found that 40 tonnes per hectare of animal manure lowered soil electrical conductivity from 6.81 to 4.63 dS/m and pH from 8.12 to 7.97, against control plots that received none [11]. Deeper manure, mixed into the soil rather than banded, pushed conductivity lower again: the trial's lowest value, 3.99 dS/m, came from 40 t/ha mixed at 30 cm [11].

That is the whole finding. One field, one season, three soil measurements after harvest [11]. The abstract reports no crop yield, no second year, and nothing on whether any treated plot crossed a defined salinity threshold. Promising first result, not a method to copy with confidence.

What the trial did

The experiment ran in 2025 at Bajiazi Village, Dongsanjiazi Township, Qian Gorlos Mongol Autonomous County, Songyuan City, Jilin Province, on silty loam [11]. It tested three manure rates (0, 20 and 40 t/ha), three depths (10, 20 and 30 cm) and two placement methods — mixed through the soil versus a subsurface band [11]. The design was a split-split plot inside a randomised complete block [11]. Corn went in straight after the manure [11].

Post-harvest samples were analysed for electrical conductivity, pH and bulk density. Nothing else [11].

Three effects:

• Rate. At 40 t/ha, EC fell to 4.63 dS/m and pH to 7.97, against 6.81 dS/m and 8.12 in the control [11].
• Depth. 10 cm to 30 cm dropped EC from 6.13 to 5.48 dS/m, with no significant difference to pH [11].
• Method. Mixing gave lower EC, 5.42 dS/m, than subsurface banding at 6.13 dS/m. pH did not differ [11].

Bulk density moved too, from 1.23 to 1.20 Mg/m3 as depth went from 10 to 30 cm, and mixing lowered it further, best at 40 t/ha mixed to 30 cm [11]. About 0.03 Mg/m3, roughly 2 percent. Real, but small.

One oddity: the abstract states depth had no significant effect on pH, then reports the lowest pH, 7.95, at 30 cm [11]. Both can be true — the best single plot can sit at a depth where the average effect is not statistically distinguishable — but it means the pH story is weaker than the headline number suggests.

Why depth and mixing might matter

The paper does not test a mechanism. It measures outcomes. Everything below is reasoning about why such a result might appear, not a finding of this study.

Salts accumulate near the surface when water evaporates faster than it moves down. Organic matter changes several things that bear on that. It raises cation exchange capacity, giving dissolved ions somewhere to sit rather than staying in the soil solution. It improves aggregation and porosity, so water moves through the profile instead of ponding and evaporating. And it feeds the microbial community, whose residues and exudates help hold aggregates together.

Mixing manure through the top 30 cm puts it in contact with a much larger soil volume than a band does, one reason mixing might beat banding on a measure like EC [11]. Burying it deeper also places the organic material below the most intense evaporation zone. But the trial measured none of this: no evaporation, no water movement, no salt distribution through the profile [11]. The abstract reports EC as a single value per plot. Whether salts moved down, were bound, or were diluted by some other route, this study cannot say.

The same caution applies to the machine. The work is framed around subsoil technology, but the abstract gives no detail on the implement, its power requirement, its cost, or how it behaves at scale [11].

What the rest of the evidence adds

Not much that transfers directly, and it is worth being blunt about that.

The broad case for organic amendments and reduced disturbance is well established. A 2022 review of conservation agriculture describes conventional practices as drivers of compaction, water erosion and salinisation, and concludes conservation agriculture can improve the physical, chemical and biological properties of soil [1]. That is a review of farming systems, not of manure depth in saline soil.

A 2020 review of rice establishment found non-puddled establishment plus residue retention raised soil organic carbon by 79 percent and total nitrogen by 62 percent relative to conventional puddling, with reduced bulk density and increased porosity and available water content [4]. Large numbers, but a different system, crop and set of practices — and the review notes no single establishment technology is superior across contexts.

On indicators, a 2025 systematic review that screened 954 publications and included 16 found soil organic carbon, available macronutrients, mean weight diameter and microbial biomass to be the most sensitive soil quality indicators, and reported organic amendments and conservation tillage support soil functioning [7]. Sixteen studies is a thin base, and the review concerns rainfed land uses, not saline soils.

Deep tillage does change soil biology. A 2020 study in Yunnan mountain soils found subsoiling to 30 and 40 cm in flue-cured tobacco increased urease, acid phosphatase and cellulase activity and raised bacterial diversity relative to 20 cm rotary tillage, while reducing catalase activity [10]. Different crop, different soil, no manure, no salinity measurement. It shows the tool does something to the rhizosphere. It does not show it does the same in saline soil.

On amendments generally, a 2026 review of engineered biochar argues amendment effects depend heavily on soil texture and pH, and that recommendations must be matched to the specific soil rather than applied uniformly [12]. A 2026 soil health review makes a related point: single inputs, manures included, often fail to sustain soil quality under intensive agriculture on their own [13]. Both are reviews, and both concern different amendments or broader systems.

Where the evidence runs out

Everything below is unknown from [11], and it matters.

No yield. Corn was planted immediately after application [11]. The abstract reports no grain or biomass figure. A salinity drop that does not show up in the crop is interesting but not yet useful.

No multi-year data. One growing season, one set of post-harvest samples [11]. Topsoil salinity effects can reverse with the next irrigation or dry season. Whether the reduction persisted is not addressed.

No baseline. The abstract reports post-harvest values only. Treated plots ended lower than controls, but we do not know how much of that gap existed at the start — though the randomised block design exists precisely to handle that.

No threshold. The paper says the treatments reduced EC and that this offers a strategy to mitigate salinity stress [11]. It does not say whether 4.63 or 3.99 dS/m is low enough for corn, or by whose standard. The control was 6.81 dS/m; the best treatment 3.99. Whether that crosses a line that matters to the crop is not answered.

No mechanism, no profile. EC was measured as a single value per plot [11]. No data on where the salts went, no water balance, no microbial or biological measurement, no organic carbon measurement. Three physical and chemical properties were measured, and that is where it stops.

No economics. 40 t/ha is a substantial quantity of manure [11]. The abstract gives no cost, no labour requirement, no machinery specification and no comparison against cheaper alternatives.

No generality. One site, silty loam, one season, saline conditions in Songyuan City [11]. Nothing here tells you what happens on sandy soil, at a different salinity level, in a wetter climate, or with a different manure source.

The honest summary: in one saline field, in one season, heavy manure applied deeper and mixed rather than banded ended the year with lower EC, slightly lower pH and marginally lower bulk density than control plots [11]. That is a reasonable basis for a replicated trial on your own ground. It is not a basis for changing how you farm.