Studies of transformations of sulphur in soils

Abstract

Different forms of S (inorganic sulphate-S and organic S forms) were measured in extracts from field moist, air-dried and conditioned soils (i.e. re-wetted air-dried samples) using a range of extractants. The amounts and proportions of different forms of S in the extracts varied with soil type, method of sample preparation and the extractant used. For the soils with pH values close to or above 6, water and phosphate extractants removed very similar amounts of inorganic and organic S. However, phosphate extractants removed greater amounts of S than water from soils with relatively low pH or with a large capacity for sulphate adsorption. CaCl₂ solution extracted the smallest amounts of sulphur from the soils studied. For strongly acid mineral soils (pH<5.3), S extraction with water and CaCl₂ solution may thus underestimate the pool of plant available S in soils.

Substantially less S was extracted from field-moist soil compared to air-dried and conditioned soil. This suggests that the analysis of air-dried or conditioned soil samples might overestimate the levels of plant available S.

In general, most S in extracts was present as inorganic SO₄²⁻, however there were also relatively large proportions of organic S in the phosphate extracts from air-dried soils. The concentrations of extractable SO₄²⁻-S, organic hydriodic acid reducible S (HI-S) and carbon bonded S (C-S) in field soils varied with season. Under field conditions, the short term release of SO₄²⁻-S in soils may be derived mainly from the mineralisation of extractable organic sulphur.

In order to examine the effects of soil preparation and incubation technique on S release in soils, S mineralisation was measured in field moist and conditioned soils which were incubated in both closed and open systems. Evidence obtained in this experiment indicates that conditioning of soils did not influence greatly the amounts of S mineralised in soils incubated in the open system, but did influence S mineralisation in some soils in the closed system. In the short term, soils incubated in the closed system released similar or greater amounts of S to those incubated in the open system. However, in the long term, more S was mineralised in the open compared to the closed system.

Dung and urine additions to soils not only increased the amount of S in the soil, but also affected S transformations. A substantial amount of extractable S in the dung-treated soil incubated in the closed system was immobilised during the early stage of the incubation. However, the immobilised S was released gradually thereafter. No net SO₄²⁻-S release occurred until after 8 weeks of incubation. In the open incubation, S release in the dung-treated soil was very slow once the original extractable S in the dung soil mixture was removed by leaching. Dung addition reduced the release of SO₄²⁻ from the native organic S in the soil. In contrast to dung, urine addition stimulated the decomposition of native soil organic S in both incubation systems, particularly in the early stages of the incubation. There was evidence that over 20% of S added in urine was incorporated into organic forms during the incubation and that the incorporated S seemed to be easily remineralised.

Experimental results show that K⁺ addition increased significantly the amount of mineralised SO₄²⁻ but had little effect on other forms of S in soils. Mg²⁺ application did not influence S release, but did influence the concentration of soil organic HI-S. The addition of SO₄²⁻ at relatively high rates appeared to encourage the mineralisation of native soil organic S. At the end of the incubation, most incorporated S was present in the form of HI-S.

A preliminary study using coarse textured soil indicated that it is possible to use undisturbed soil cores for S mineralisation or transformation studies and this simulates more accurately conditions in field soils. A qualitative comparison of results from these preliminary experiments showed that S was mineralised and that added ³⁵S was incorporated into soil organic matter in both repacked soil columns and undisturbed soil cores. It was also evident that in general, the distribution of added ³⁵S between HI-³⁵S and C-³⁵S in the undisturbed soil during the experiment was similar to that in the repacked soil.

The results obtained from this study also indicated that sulphate addition increased the concentration of phosphate-extractable SO₄²⁻-S in soils and changed the depth distribution of this form of S in soils. Most ³⁵S incorporated into organic sulphur was present as C-³⁵S in the top soil (0-25 mm), whilst HI-35S was dominant in the 25-100 mm depth. This suggests that conditions may be more favourable for the transformation of S to G-S in organic rich topsoils and to HI-S forms in lower horizons with less organic matter.