Liming effects on legume production and phosphorus availability in acid South Island hill and high country soils : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

There is large potential for widespread development and increased production in South Island hill and high country by increasing legume production. Despite regular past applications of superphosphate fertiliser, current low P and S fertility levels, along with soil acidity and Al toxicity, limit legume production and persistence in these environments. In a survey of 19 farmed South Island hill and high country soils pH H₂O ranged from 4.3 to 5.5, and soil exchangeable Al ranged from 0.5 to 24.5 mg/kg in 0-7.5 cm deep soil. Sulphur (S) concentrations were generally very low, and S deficiency is therefore likely to be a major limitation to legume production and biological nitrogen (N) fixation across the hill and high country. In contrast, a considerable amount of P has accumulated in these soils, with total P ranging from 587 mg P/kg up to 1570 mg P/kg. However, plant available P only accounted for 7.2±0.45% of total soil P, or 78±7.9 mg P/kg across all of the soils. The greatest amount of P has accumulated in moderately labile organic (39.1±1.50% of total P) and inorganic (13.3±0.91% of total P) P fractions. To investigate if liming would invoke a ‘P-sparing effect’, four soils were treated with rates of 0, 1, 2, and 4 t/ha of lime, with and without additional P, and exhaustively cropped in a glasshouse experiment. Russell lupins (Lupinus polyphyllus), and lotus (Lotus pedunculatus) were used as bio-indicators of soil P availability in four high country soils. Liming increased plant growth (P<0.001), shoot P concentrations (P<0.001), plant P uptake (P<0.001), and the utilisation of available soil P (P<0.001) for both species up to certain optimal pH ranges. For Russell lupins this optimal pH H₂O range was 5.5-6.5, and for lotus it was 6.2-6.8. When soil pH rose above these levels plant growth, shoot P concentration and P uptake rapidly declined. As the optimal pH levels for each species was different, the results indicate that liming effects up to optimal pH levels were due to plant physiological responses and increased plant ability to access and utilise existing soil P, rather than increased soil P availability due to a ‘P-sparing effect’. Liming effects on soil P availability were also investigated in a field experiment at Mt Grand Station. The annual application of 1 and 3 t lime/ha alone had no effect on pasture growth (P=0.583) and clover content (P=0.187), and did not increase soil P availability. Instead, liming was shown to promote the accumulation of plant unavailable Ca-P secondary minerals in the soil, especially in plots treated with additional P fertiliser (P=0.016). The initial exchangeable Al concentration at the site was 1.9 mg/kg, so it is suspected that greater clover growth responses to liming would have occurred if the initial exchangeable Al concentration had exceeded the general 3 mg/kg toxicity threshold. Instead, clover growth (P<0.001) and content in the sward (P<0.001) responded greatest to the application of S, and the suppression of grass competition by the application of a selective grass herbicide (P<0.001). Subsoil acidity and Al toxicity can also restrict the growth of deep rooting legumes, such as lucerne (Medicago sativa). Another aspect of this project was to implement and test a prototype machine for directly injecting lime deep into acid subsoils to alleviate acidity and Al toxicity. This was successfully achieved at Omarama Station, where soil pH H₂O was increased to >5.5 (P=0.010) from 5.0-5.2, and exchangeable Al was reduced to <3 mg/kg (P<0.005) from 4.6-7.5 mg/kg, at 20-30 cm depth by the deep application of lime. This resulted in lucerne yield being increased from 980±140 kg DM/ha to a maximum of 2890±330 kg DM/ha (P<0.001) in plots treated with 2 t/ha of deep placed lime in the fourth spring of the experiment. However, lucerne was out-yielded by Russell lupins, which were not affected by deep liming at any of the four sites. From this project it can be concluded that liming acid South Island hill and high country soils will not cause ‘P-sparing effects’ and increase the quantity of plant available P in soil. But by alleviating Al toxicity, improving micronutrient availability, and many other beneficial effects, legume production and the ability of plants to utilise existing soil available P will be increased by liming.