The effects of short-term pasture-arable rotations on the accumulation, forms and subsequent mineralisation of organic sulphur and nitrogen in soils

In the short-term mixed cropping farming system, commonly used on the Canterbury Plains region of New Zealand, grazed ryegrass-white clover pastures are grown in rotation with arable crops. Arable crops are generally grown for 2-5 years and are then followed by 2-5 years of grazed grass/legume pasture to maintain soil productivity. Historically, ryegrass-white clover pastures have been used as restorative crops in New Zealand mixed cropping rotations. Recently mixed herb leys have received considerable attention in New Zealand as restorative crops, especially in organic farming systems, and many New Zealand farmers regard lupins as an annual restorative crop due to both its nitrogen fixing ability and its reported effect as a biological plough. Soil organic matter (OM) increases under improved pasture and decreases during arable cropping. Several aspects of organic sulphur (S) and nitrogen (N) accumulation during short-term pasture under different pasture types and management and changes during cropping were addressed in this thesis. The effects of restorative crops on the accumulation of organic S and N in soils and the effects of S and N accumulated under pasture on the subsequent arable crops were investigated in a field experiment established on the AgResearch farm in Canterbury. Treatments included perennial and annual pastures and arable crops during a 6-year pasture phase, and cultivation of wheat during a 3-year cropping phase. Soils from four other locations were also included in the study to investigate the OM build up in soils under a wider range of soil types and management. Results from the restorative crops trial (RCT) soils showed that organic S and N accumulation were significantly different between some of the treatments during the pasture phase and was related to the amount of residue return to the soil. The perennial treatments showed greater increase in soil organic S and N accumulation than the annual treatments. Organic matter accumulation was higher in the grazing treatment than the mowing treatment, and direct drilled cultivation of annual pasture accumulated higher amounts of S and N in soils than conventional cultivation. The grazed perennial ryegrass and perennial ryegrass/white clover treatments were the most beneficial in accumulating OM in soil during the pasture phase of the trial. Over 95% of soil S and over 99% of soil N accumulated under pasture was in organic form. The major proportion of the accumulated S was in the HI-reducible form (ester-S). Cultivation of wheat during the cropping phase reduced OM from soils under all the former restorative treatments. The beneficial effects of the pasture accumulated organic S and N disappeared very quickly during cropping of wheat in the cropping phase. The grazed perennial ryegrass and perennial ryegrass/white clover treatments showed the highest benefit on the subsequent wheat crops in terms of S and N uptake. An open incubation study was conducted with soils from selected treatments of the RCT trial for 18 weeks to investigate the mineralisation potentials of the pasture accumulated organic S and N in soils. Net mineralisation of S was significantly different between treatments up to eight weeks of incubation. Net N mineralisation, on the other hand, was significantly different between some of the treatments throughout the entire incubation period. During the incubation study rapid S release occurred until the eighth week of incubation, whereas, N release slowed down soon after the second week of incubation probably due to a decrease in readily mineralisable substrate in the system. However, the total amounts of S and N release from the soils were small. Studies on light fraction organic matter (LF-OM) in soils from the RCT, long-term pasture ecology (LET) and the Pukekohe soils indicated that the restorative treatments of the RCT trial significantly affected the accumulation of LF-OM in soils during the pasture phase. Amounts of LF-OM in soil were variable under different soil types and management practices. Residue return from the pastures during the pasture phase directly influenced the accumulation of LF-OM in soil. Arable cropping reduced the LF-OM in soil in the RCT trial but the contribution of S and N from LF-OM during the cropping phase was less than 10% of the requirements of S and N for wheat. Mineralisation of S and N from LF-OM in an aerobic incubation revealed that NaI solution used for isolation of LF-OM from soils reduced mineralisation potential of LF-OM. Overall S and N release from LF-OM was very small. Extraction of OM from soils was carried out with NaOH, aqueous acetylacetone and chelating resin. Results obtained from the extraction procedures revealed that all the three extractants extracted substantial and comparable amounts of organic S and N from soils. The proportions of organic S and N extracted were clearly different over the wide range of pasture management and soil types. In the RCT soils the major proportions of soil S, N and HI-S were present in the humic acid fractions, but in the LET and the Pukekohe soils the major proportions of N and HI-S were present in the fulvic acid fraction. Mineralisation of S and N from the humic and fulvic acid fractions was significantly different under the restorative treatments and both humic and fulvic acid fractions contributed to the S and N turnover during the cropping period. Greater N turnover occurred from the fulvic acid fraction except for the annual barley treatment whereas, S mineralisation occurred mainly from the humic acid fraction. Ultrasonic dispersion in the acetylacetone extraction did not increase OM extraction compared with the NaOH and chelating resin methods but sequential extraction of OM by acetylacetone indicated that quite high proportions of OM are more difficult to extract. Chelating resin, although a milder extractant, can extract substantial amounts of OM from soils under short-term pasture-arable rotations and it extracted potentially labile fractions of HI-S as shown by the significant correlation between chelating resin extractable HI-S and fulvic acid S after the pasture phase. Among the soil OM extraction methods studied, the NaOH extraction was the cheapest, quick and easy to use procedure