Nonphytotoxic Chloride Loading Limits for Rice Paddy Basal Fertilizer

When potassium chloride is incorporated as a basal dressing in an alluvial paddy soil with a saturated paste electrical conductivity below 1.0 dS m⁻¹ and a clay content of 35%, the nonphytotoxic chloride loading limit is best expressed as a soil solution concentration term rather than a single broadcast application mass. A basal potassium recommendation of 40 kg K₂O ha⁻¹ supplied as potassium chloride requires 63.4 kg KCl ha⁻¹ because the K₂O equivalent of KCl is 63.2%. The corresponding chloride load is 30.2 kg Cl ha⁻¹ because chloride comprises 47.6% of KCl by mass. If this load is incorporated evenly into a puddled layer of 10 cm depth and a bulk density of 1.05 Mg m⁻³, the soil mass is 1,050,000 kg ha⁻¹. At a saturated water content of 0.45 L kg⁻¹, the water-filled pore volume is 472,500 L ha⁻¹. Uniform distribution would increase soil solution chloride by 63.9 mg Cl L⁻¹, or 1.80 mmol L⁻¹. This increment is additive to background chloride, which in percolating alluvial paddies ranges from 0.5 to 4.0 mmol L⁻¹ depending on canal water quality. The saturated-paste chloride threshold for transplanted rice seedlings is taken operationally as 20 mmol L⁻¹, derived from the Maas–Hoffman salinity threshold of 3.0 dS m⁻¹ for rice grain yield and the assumption that chloride contributes 50% of the total anion charge in the saturated extract. The bulk calculated increment of 1.80 mmol L⁻¹ therefore remains below the seedling exposure limit when the background chloride is less than approximately 18 mmol L⁻¹, but this result depends entirely on uniform mixing and does not describe banded or furrow-placed fertilizer zones.

Chloride does not participate in cation exchange on permanent negative clay surfaces, and it is not reduced to a gaseous form under flooded paddy redox conditions. Its movement is governed by the same water flux that drives puddling water retention and percolation. The hazard from basal application therefore depends strongly on hydraulic displacement before root elongation reaches the incorporated layer. At a percolation rate of 5 mm d⁻¹, the daily water throughput is 50 m³ ha⁻¹, and over 7 days the cumulative throughflow is 350 m³ ha⁻¹. This corresponds to 0.74 pore volume replacements of the 472,500 L ha⁻¹ soil solution, assuming complete mixing; in practice, preferential flow paths and dead-end pores reduce chloride displacement efficiency to 0.4–0.6 of the theoretical value. Under such conditions the bulk solution increment of 63.9 mg Cl L⁻¹ may decline to 25–38 mg Cl L⁻¹ before transplanting. Field monitoring with suction cup lysimeters installed at 10 cm depth and sampled at 24 h, 72 h, and 7 days after basal incorporation provides the only reliable verification of the actual soil solution chloride exposure under a given paddy water regime. The arithmetic gradient of bulk solution chloride across puddled layer bulk densities is set out in Table 1.

Can a Single Nonphytotoxic Loading Rate Be Derived Across Contrasting Paddy Soil Textures?

A single kg Cl ha⁻¹ loading limit cannot be derived across contrasting paddy soil textures because the same chloride mass partitions into widely different soil solution volumes and rooting depths. The controlling variables are the depth of the puddled layer, bulk density, saturated water content, and antecedent chloride in irrigation water. For a fixed basal potassium dose of 40 kg K₂O ha⁻¹, the chloride increment ranges from 61 mg L⁻¹ in a low bulk density paddy soil to 77 mg L⁻¹ in a compacted high bulk density soil, as shown in Table 1. The corresponding chloride molar concentrations are 1.72 mmol L⁻¹ and 2.17 mmol L⁻¹, respectively. These values are below the 20 mmol L⁻¹ seedling exposure limit, but local granule dissolution and incomplete mixing invalidate the assumption of uniform distribution. When a rotary puddler with a rotor speed of 180–220 rpm incorporates granules under 8–10 cm of standing water, the coefficient of variation of chloride in the upper 10 cm soil solution after 24 h may exceed 45%. Published data for this specific configuration are limited, but the nonuniformity demonstrates that a single field-averaged loading limit is insufficient without a spatial heterogeneity factor. Consequently, the nonphytotoxic loading limit should be expressed as a maximum saturated extract chloride concentration, not as a fixed mass per hectare.

The saturated paste extract is prepared by adding deionized water to air-dry soil until saturation paste criteria are met and extracting under vacuum. Chloride is then determined by silver nitrate titration according to ISO 9297:1989. The electrical conductivity of the same saturated extract is measured according to ISO 11265:1994. For paddy rice, the baseline screening limit is 3.0 dS m⁻¹ for ECe and 20 mmol Cl⁻ L⁻¹ for chloride in the seedling root zone. Depending on cultivar and growth stage, the operational limit can be lower because direct-seeded rice at the coleoptile and first-leaf stage is more sensitive to osmotic stress than transplanted rice at the five-leaf stage. The chloride concentration in the bulk saturated extract integrates the mass of chloride applied but does not capture the transient concentration peak around dissolving fertilizer granules. Therefore, risk assessment for basal potassium chloride must combine the bulk arithmetic loading with a placement-specific local concentration model and a measured saturated paste chloride value at 72 h after incorporation.

In direct-seeded basins with six-row walking-type seeders, fertilizer is often placed in a separate furrow 5 cm to the side and 3 cm below the seed furrow. The seedbed is frequently non-puddled or only lightly pressed, and chloride applied as basal KCl remains in a narrow band during the first 5–7 days after sowing. Under these conditions the local chloride concentration in soil solution around the fertilizer band is governed by granule mass, row spacing, furrow geometry, and soil water content. A band application of 63.4 kg KCl ha⁻¹ at rows spaced 20 cm apart corresponds to 1.27 kg KCl per 1,000 m of row. In a furrow with a cross-sectional area of 15 cm², a bulk density of 1.30 Mg m⁻³, and a soil water content of 0.25 L kg⁻¹, the initial solution concentration in the furrow can reach 1,200–1,500 mg Cl L⁻¹, equivalent to 34–42 mmol L⁻¹, before diffusive dilution and water uptake by the surrounding soil. This localized concentration exceeds the 20 mmol L⁻¹ seedling exposure limit by a factor of 1.7–2.1, explaining why furrow placement causes stand loss even when the field-averaged chloride load appears acceptable. Field observations on six-row seeding equipment indicate that separation below 4 cm between seed and fertilizer furrows increases seedling stand loss, although published data for the specific interaction of furrow cross-section, soil texture, and rice variety remain limited.

When Basal KCl Is Combined with Ammonium Sulfate in the Same Microzone

Blending potassium chloride with ammonium sulfate in basal fertilizer alters the nitrogen and sulfur transformation environment around the granule. Chloride remains chemically conservative, but ammonium sulfate acidifies the dissolution microzone and sulfate is reduced to sulfide after prolonged flooding. No direct precipitation of chloride occurs because potassium sulfate and ammonium chloride are both highly soluble, but the combined salt index of the mixture raises the ionic strength and osmotic potential in the seedbed. Application of 63.4 kg KCl ha⁻¹ together with 100 kg ammonium sulfate ha⁻¹ creates a dissolution zone in which the ionic strength can exceed 1.0 mol L⁻¹ during the first 24 h. Sulfate is partly retarded by sorption on variable-charge surfaces under acid conditions, while chloride moves ahead in the wetting front. In acid sulfate paddy soils with pH below 4.5 measured according to ISO 10390:2021, chloride mobility is enhanced because the anion exchange capacity of iron and aluminum oxides increases. The combined basal application should therefore be spatially separated or split, and the two materials should not be placed in the same furrow. Rotary blending of potassium chloride granules of 2–4 mm with ammonium sulfate granules of 1–3 mm can also produce size-guided segregation in twin-disc broadcast spreaders when the size guide number difference exceeds 5, causing field-scale chloride distribution to vary by more than 20% across the spreader width. The use of a twin-disc spreader calibrated at 400–600 kg ha⁻¹ total product is acceptable only when the blend is thoroughly mixed and applied immediately after blending, because storage and vibration increase segregation.

In saline-affected coastal paddy soils where saturated paste ECe already exceeds 2.0 dS m⁻¹ and soil solution chloride exceeds 10 mmol L⁻¹ before basal application, potassium chloride should not be used as basal fertilizer unless the paddy is leached with fresh water for 7–10 days before transplanting. The additional 30.2 kg Cl ha⁻¹ from 40 kg K₂O ha⁻¹ as KCl may elevate the puddled soil solution chloride by 1.8–2.2 mmol L⁻¹. If background chloride is 12 mmol L⁻¹, the total enters the 14 mmol L⁻¹ range, approaching the 20 mmol L⁻¹ transplanting safety limit before evapoconcentration. Published data for this specific configuration in tidally influenced paddy fields are limited, but the operational boundary is clear: when pre-flood ECe exceeds 2.5 dS m⁻¹ or chloride in the saturated extract exceeds 15 mmol L⁻¹, basal potassium should be supplied as potassium sulfate or magnesium potassium sulfate, which contain chloride at less than 0.5% by mass. Supplying 40 kg K₂O ha⁻¹ as potassium sulfate requires 73.9 kg K₂SO₄ ha⁻¹ and provides 13.6 kg S ha⁻¹, which may be agronomically useful in reduced alluvial soils where sulfate respiration contributes to early carbon mineralization. The substitution removes the dominant chloride source while maintaining basal potassium nutrition, but it does not correct background salinity and should be combined with leaching and surface drainage management.

Analytical Verification and Fertilizer Quality Clauses

Water-soluble chloride in basal fertilizer is measured by potentiometric titration in aqueous extract according to ISO 8157:2015. The result is expressed as percent Cl by mass. Solid fertilizer sampling follows ISO 14820-1:2016 to ensure that the analytical sample represents the bulk product. Soil chloride in saturated paste extracts is titrated with silver nitrate using potassium chromate indicator according to ISO 9297:1989. Electrical conductivity is measured with a conductivity cell at 25 °C according to ISO 11265:1994. Soil pH is determined in 1:5 soil-water suspension according to ISO 10390:2021. Particle size distribution is determined by sieving and sedimentation according to ISO 11277:2009. Fertilizer label chloride declarations in the European Union are governed by Regulation (EU) 2019/1009 and the predecessor Regulation (EC) No 2003/2003; water-soluble chloride is declared as grams of chloride per kilogram of product when present at or above the applicable regulatory threshold. Products containing potassium chloride must therefore show a chloride declaration reflecting the 47.6% chloride mass fraction of the ingredient, and this declaration provides the starting point for calculating chloride load at the farm gate. Compliance with these analytical and labeling clauses does not by itself establish nonphytotoxic application, but it creates a defensible chain from bulk fertilizer composition to soil solution exposure.

A chloride loading verification protocol on a production paddy can be implemented with porous cup vacuum samplers installed at 10 cm depth and sampled at 24 h, 72 h, and 7 days after basal incorporation. The sampler vacuum is set at 50 kPa for 30 min. Collected soil solution is filtered through 0.45 µm membrane filters and titrated within 24 h. If the 72 h chloride concentration exceeds 20 mmol L⁻¹ in the root zone, the basal chloride loading for that field should be reduced by replacing part of the potassium chloride with potassium sulfate or by shifting part of the potassium dose to a split application at tillering. Under high rainfall or continuous fresh water percolation above 10 mm d⁻¹, the leaching fraction reduces the observed chloride concentration by 30–50% relative to stagnant paddy conditions; the root zone monitor therefore identifies operational fields where basal KCl remains below the nonphytotoxic threshold despite a high calculated mass loading. The use of this protocol on a production scale has shown that fields with identical fertilizer invoices can diverge in root zone chloride concentration by more than 10 mmol L⁻¹ within 7 days because of differences in percolation, puddling depth, and antecedent chloride. Published data for the specific interaction of basal chloride loading with paddy microbial methane emissions and redox-sensitive nutrient transformations remain limited, and no agronomic directive should infer a single safe chloride value without measuring the saturated extract chloride under the target water management regime.