Few studies have examined the effects of winter soil temperatures typical of temperate regions (0-15 degreesC) on the release of nitrogen (N) from plant residues. Similarly, few have studied gross N transformation rates (mineralization, nitrification and immobilization) as an interactive unit. N cycling was examined in clover amended or unamended soil incubated under constant laboratory temperatures of 2, 5, 10 or 15 degreesC for 161 days. Under laboratory conditions we also examined the impact of a sudden change in soil temperature whereby amended soil previously incubated for 98 days at 2, 5 or 10 degreesC was subsequently incubated at 15 degreesC, while amended soil previously incubated at 15 degreesC was incubated at 2 degreesC for a further 63 days. The effect of fluctuating winter temperatures was studied using intact soil cores under winter field conditions for 35 days. The kinetics of N transformations were determined in the laboratory incubation and field experiment by measuring soil ammonium (NH4+-N) and nitrate (NO3--N) concentrations and gross rates of mineralization, nitrification and immobilization. The fate of N-15 labelled clover residue was also measured in the field experiment.In the laboratory incubation and field experiments, soil mineral-N concentration was significantly (P < 0.001) higher in amended, compared with unamended soil. Under laboratory conditions mineral-N concentration significantly (P < 0.05) increased with increasing incubation temperature in amended soil. In unamended soil, mineral-N concentration was significantly (P < 0.05) greater when incubated at 15 degreesC than at 2, 5 or 10 degreesC alone. Under winter field conditions all mineral-N released from clover residues was at risk of leaching during winter rainfall.Gross nitrification was initially (7-56 days) inhibited in amended soil incubated at 2 or 5 degreesC, causing an accumulation of NH4+-N. However, after 77 days at 2 or 5 degreesC, gross nitrification rates increased, such that NO3--N increased to concentrations which were greater than those of NH4+-N. This suggests that nitrifying bacteria took longer to acclimatize to the cold conditions than ammonifying microorganisms. Nitrate-N was the dominant form of mineral-N throughout the incubation experiment in amended soil incubated at 10 or 15 T. In unamended soil, gross immobilization rates generally followed the same pattern as gross mineralization rates throughout the incubation. Unamended soil incubated at 10 degreesC and below produced negligible NO3--N, indicating that N or carbon limited nitrification at these temperatures. Increasing incubation temperature from 2, 5 or 10 degreesC to 15 degreesC caused a rapid increase in soil NO3--N concentration and gross mineralization and nitrification rates, but significantly (P < 0.05) less mineral-N was released than if incubated at a constant 15 T. This suggests that intermediate substrates may have been depleted during the initial incubation period at 2, 5 or 10 degreesC, hence limiting mineralizable-N. Decreasing soil temperature from 15 to 2 degreesC caused an initial increase in mineral-N, which was quickly followed by rapid immobilization of mineral-N; gross immobilization rates were up to 2.8 fold greater than gross mineralization rates. Similarly, under field conditions, microbial biomass N and gross immobilization increased with decreasing soil temperature suggesting there was population growth of adapting micro-flora. The release of mineral-N from clover residues in the incubation experiment also seemed to occur in two-phases, interpreted as first the mineralization of the labile and then the more recalcitrant fractions of the residues.This research has shown that significant mineral-N is released from soil amended with clover residues at temperatures as low as 2 degreesC. Therefore, the incorporation of N-rich plant material should be delayed until spring to avoid winter N leaching. (C) 2002 Elsevier Science Ltd. All rights reserved.

The effects of amending an acid Andisol with lime and gypsum on soil pH, aluminum (Al) saturation,the plant-availability of sulfur (S) and phosphorus (P), and the growth, botanical, and chemical composition of a ryegrass/white clover pasture were measured in a field trial in southern Chile.The combined amendment of limestone, dolomite and gypsum raised soil pH slightly, decreased Al saturation from 20 to less than 1% and increased the concentration of plant available P, S, calcium (Ca), magnesium (Mg), and potassium (K) in the soil. The growth of a recently-sown rye grass-white clover pasture increased by 50% and the proportions of ryegrass, white clover and weed species changed from 12.6, 0.2, and 87% to 75, 7.7, and 17%, respectively. Plant chemical analysis suggested that, while N, P, and S concentrations in mixed pasture were less than critical values in spring, S was the most limiting nutrient in pastures from both the amended and control areas.

Decreasing the production of nitrate-N (NO3−-N) in cattle urine patches may reduce the environmental impacts of pastoral agriculture. This paper reports the effects of the nitrification inhibitor dicyandiamide (DIDIN) on NO3−-N and ammonium-N (NH4+-N) production, soil pH and pasture yield in urine-amended and control soil under field conditions. In control plots, DIDIN application did not affect NO3−-N, NH4+-N, pH or pasture yields. In urine-amended plots, DIDIN application significantly (P<0.05) reduced peak NO3−-N concentrations, the amount of NO3−-N leached, hence decreasing the potential for denitrification losses. Conversely, soil NH4+-N concentration decreased more gradually when urine was amended DIDIN which increased the opportunity for greater immobilization to occur. Differences in NO3−-N and NH4+-N concentrations between urine-amended treatments were also reflected by increases in soil pH but not pasture yields.

Mixed pastures of white clover and ryegrass are profitable and widely used. However, loss of clover is common with time due to N fertiliser applications in late winter. The factors that cause this loss have yet to be determined. This work investigated reasons for the competitive disadvantage suffered by white clover relative to ryegrass in autumn and winter. White clover (Trifolium repens cv. 'Grasslands Huia') and ryegrass (Lolium perenne cv. 'Grasslands Nui') were sown in March in Canterbury, New Zealand, and grown as single plants in silica sand. Nitrogen (N; 0.5 mol m(-3) N); was applied daily simulating an unfertilised soil; solution N concentration was increased to 5.0 mol m(-3) N for treatments in autumn (May) or winter (August), simulating fertiliser treatment. Net photosynthesis was measured before each harvest and over a diurnal period at 123 days after sowing. Plants were destructively harvested six times (33-185 days after sowing) and plant dry weight and N concentration (%) measured. White clover had a significantly (P < 0.05) lower dry weight than ryegrass. Extra N applied from May significantly increased (P < 0.05) dry weight of ryegrass. No significant changes in shoot N concentration were found. Except at the last harvest, net photosynthesis was significantly lower (P < 0.05) in white clover than in ryegrass; this was also apparent for diurnal net photosynthetic rates 123 days after sowing. Results indicated that white clover growth was limited by temperature, whereas ryegrass growth was limited by N supply.

The release of SO42–-S, K+, Ca2+ and Mg2+ from soil amended with spent mushroom compost (SMC), a byproduct of mushroom production, was measured in leachate from field lysimeters for 30 weeks. Rates of application were 0 and 80 t ha–1 moist SMC. The SMC contained 1.7% K, 6.5% Ca, 0.4% Mg and 1.2% S (of which 87% is SO42–-S), and has a C : S ratio of 26. The break-through curves of ion leaching were polymodal indicating the preservation of soil structure in the lysimeters and its influence on leaching. SO42–-S release from SMC was rapid (first-order exponential) and was very similar to the release from a laboratory incubation. The release of K+, Ca2+ and Mg2+ was described using first/zero-order models which were also used to describe their release in the laboratory. The rate and amount of Ca2+ release was similar in the field and laboratory, but the amount of K+ (and to a lesser extent Mg2+) release was less in the field than in the laboratory. Recoveries of SMC applied nutrients in leachate were 80% of S (263 kg ha–1), 3% of K (14 kg ha–1), 16% of Ca (284 kg ha–1) and 37% of Mg (40 kg ha–1). Little if any S was mineralised. Using SMC could provide plants with S, K, Ca and Mg but there is potential for SO42–-S losses via leaching.

While increasing amounts of nitrogen (N) are being applied to white clover and perennial ryegrass in autumn and winter when low temperatures prevail, the uptake of fertiliser N at low temperatures, and consequent possible effects on leaching, have not been quantified. A lysimeter study to investigate the fate of N-15 applied in May or August to white clover (Trifolium repens L.) or perennial ryegrass (Lolium perenne L.) was established in the autumn of 1998. The objective was to determine the extent to which differences in agronomic features between the two species would affect the uptake and losses of N. The composition of the leachate collected during winter was determined. Plant uptake and the distribution of N-15 in soil were analysed after destructive harvesting in September 1998. Leaching losses of mineral N were greater under white clover than perennial ryegrass (P<0.05). White clover took up less N-15 than perennial ryegrass (P<0.05). This was due to the slower growth of white clover, which may have been a consequence of biological N fixation. Sixty five percent of fertilizer N remaining in the soil in white clover was present in the top 100 mm; 65% of both white clover and perennial ryegrass roots were in the top 100 mm suggesting that N was available in the rooting zone for subsequent uptake and growth.

While the importance of the concept of the sustainability of land management practices is now widely accepted, there remains considerable debate on methods of identifying sustainability. This paper proposes a series of criteria which can be used to select indicators for assessing the sustainability of land management systems. «Sustainable land management» is defined using the five objectives of productivity, security, protection, viability and acceptability. The paper introduces the concept of «areas of concern», which are defined as any factor able to influence the ability of a production system to meet the five objectives of sustainable land management. Environmental indicators used to monitor areas of concern must be sensitive to management actions and must be related in a functional way with those parts of the system which may be at risk. Indicators must have identified critical values beyond which a particular system of land management is no longer sustainable. Critical values for indicators often depend on an understanding of the mechanisms which control the relationships between management and the final arbiter of sustainability. Critical values may vary depending on the characteristics of the system but will be independent of management, although management will influence the rate at which an indicator approaches its critical value. It is argued that the criteria used to select biophysical indicators can also be used for selecting indicators of the economic, social and commercial aspects of sustainability.

Soil tests suitable for estimating the phosphorus (P) status of soils fertilised with soluble or sparingly soluble P fertilisers (reactive phosphate rock) were evaluated using the New Zealand Ministry of Agriculture Technology (NZMAFTech) ‘National Series forms of phosphate trials’ on permanent pastures located throughout NZ. This included a common core of treatments comparing Sechura phosphate rock (SPR) with triple superphosphate (TSP). At each site, a re-application of twice maintenance TSP was superimposed on one-half plots that previously had received six annual applications of increasing amounts of P (0, 0.5, 0.75, 1.0 and 2.0 times the maintenance rate) in the form of TSP or SPR. Before the re-application of TSP, soil samples (0–30 and 0–75 mm depths) were collected from each plot. All the trials were run for 1 year during which seven to ten harvests were taken. Pasture response was expressed as percent increase in yield obtained with re-application over the previous treatment.

A 2-year field trial determined the influence of applying spent mushroom substrate (SMS) on soil physical properties and the growth of 4 consecutive vegetable crops (sweetcorn, cabbage, potato, cabbage). Treatments comprised 0, 20, 40, and 80 t/ha of moist SMS, both with and without inorganic fertiliser, applied to each crop, giving a range of SMS rates up to 320 t/ha. SMS improved the environment for plant root growth by decreasing soil bulk density (by 0.05-0.25 g/cm(3) at 100 mm depth), increasing aggregate stability (by 13-16%), reducing clod and surface crust formation (by 16-31 and 18-94%, respectively), increasing the infiltration rate (by 130-207 mm/h), increasing the water content of the soil (by 0-7% w/w), and reducing diurnal temperature changes. Some of these changes were not evident until repeated applications of 80 t/ha SMS had been made. Soil physical properties were related to crop yield, and soil physical properties' principal components were related to crop principal components using regression analysis (r(2) of 0.20-0.60 and 0.16-0.54, respectively). The soil physical properties that had the most influence on plant growth were specific to each crop and included bulk density, water content, surface crust cover, infiltration rate, and aggregate size distribution. Soil physical properties had a large influence on the potato yield irrespective of fertiliser use and on both cabbage crop yields when fertiliser was not used, but not on the sweetcorn yield (the first crop to be grown). The effect of changing soil physical properties on plant growth was most apparent when fertiliser was not used. This was because the improved physical properties increased plant yield (at least in part) because of increased plant nutrient uptake.

Between November 1991 and 1993, 4 consecutive vegetable crops (sweetcorn, cabbage, potato, and cabbage) were grown in Lincoln, New Zealand. The treatments included spent mushroom substrate (SMS, a by-product of the mushroom industry) applications before each crop at rates of 0, 20, 40, or 80 t/ha (moist), both with and without 1 rate of inorganic fertiliser for each crop (120-338, 40-100, 53-100, and 60-114 kg/ha, respectively, of nitrogen, phosphorus, potassium, and sulfur). SMS applications caused a rapid increase in soil inorganic N concentration, but after this it had a variable effect. There was some evidence of N immobilisation following initial SMS applications of 20 t/ha. SMS applications increased both soil pH and CEC, whereas inorganic fertiliser decreased both. Sweetcorn and cabbage yields were increased by SMS when inorganic fertiliser was not used, and potato yield was increased irrespective of fertiliser use (i.e. yield increases of 38%, 82-96%, and 26-46%, respectively, for sweetcorn cob, cabbage head, and potato tuber fresh yields). Inorganic fertiliser increased crop yields by a greater amount than SMS. A lack of soil inorganic N was the major limitation to crop growth following SMS applications, so crops may require additional N with SMS.

Inorganic-N release from soil amended with spent mushroom compost (SMC), a by-product of mushroom production, was measured in three open laboratory incubations (25–30°C) and in field lysimeters. Rates of SMC application to the soil were up to 80tha−1 equivalent (0.84% dry weight in the laboratory). SMC contained 1.8% N of which 94% was organic, and had a C-to-N ratio of 17. Small amounts of inorganic-N were leached from SMC in the first incubation (3–18% of that applied). Trends in the data suggested that N in the SMC was initially immobilized in the 20 and 40tha−1 treatments, as shown by modelling using a negative first order exponential term; it was then slowly mineralized according to zero order kinetics. The laboratory optimized model of inorganic-N loss, when modified to account for field soil temperatures, estimated a similar amount of inorganic-N loss as was observed in the field. The century model overestimated inorganic-N leaching from SMC in the laboratory and underestimated inorganic-N leaching in the field. Fertilizer, containing N, P, K and S, reduced the net amount of inorganic-N recovered from SMC–soil mixtures. The rate of inorganic-N leaching from mushroom compost was considerably slower than from glycine or chicken litter applied at the same N rate. The sterilants applied to mushroom compost during mushroom production and compost sterilization had little effect on the rate of inorganic-N leached from the compost; however, hypochlorite and formaldehyde caused a small increase and decrease respectively in the cumulative amount of inorganic-N leached from mushroom-compost-amended soil. The slow rate of release of inorganic-N from SMC-amended soil is predominantly the result of the slow mineralization of recalcitrant organic-N in SMC.

The release of sulphate-sulphur (SO42–-S), potassium (K), calcium (Ca) and magnesium (Mg) from soil amended with spent mushroom compost (SMC), a by-product of mushroom production, was measured for 16 weeks in an open laboratory incubation at 25°C. Rates of application were up to 80 t ha–1 moist SMC (0.84% SMC dry weight) both with and without inorganic fertilizer. The rates of nutrient application in the inorganic fertilizer were: 338 kg ha–1 N, 100 kg ha–1 of both phosphorus and K, and 114 kg ha–1 S. SMC contains 1.7% K, 6.5% Ca, 0.4% Mg and 1.2% S (of which 87% is inorganic), and has a carbon:sulphur ratio of 26. The release of SO42–-S was rapid, and was described using either a first or mixed order exponential equation, or (underestimated) by the CENTURY model. The release of K, Ca and Mg was initially rapid (first order) and then declined to a constant rate (zero order). Their release was also described using first/first order or first order/parabolic diffusion equations. Model parameters indicated the relative sizes of both readily releasable and recalcitrant nutrient pools. The recovery of SMC-supplied nutrients in the absence of fertilizer was 75–83% of the S, 40–45% of the K, 14–20% of the Ca and 43–66% of the Mg. When fertilizer was applied 33–45% of the S, 22–36% of the K, 12–24% of the Ca and –4 to 20% of the Mg that were supplied by the SMC and fertilizer were recovered in the leachate. The generally lower nutrient recovery when fertilizer was applied could have resulted from the incomplete recovery of fertilizer S and K, from soil fixation of applied nutrients, and from the lower pH following fertilizer application.

We examined the effects of conifers on the forms of P in low-fertility tussock grassland soils using 31P nuclear magnetic resonance (NMR) and soil P fractionation. Results from field and glasshouse experiments clearly demonstrated that conifers enhanced the mineralization of labile (and to a lesser extent more resistant) forms of soil organic P which, in turn, increased amounts of labile inorganic P in the soil. These findings have important implications for P availability and long-term sustainable management of grassland soils in New Zealand.

A range of soil tests, (moderately alkaline extractions of Olsen and Colwell, acid extractions of Truog and Bray and an anion exchange resin, with and without a cation exchange resin) were assessed for their ability to extract phosphorus (P) from soils incubated with monocalcium phosphate (MCP) and phosphate rocks (PRs) of varying reactivity. The mixed anion-cation resin test (Resin) extracted part of residual PR corresponding to the amount present and its reactivity, as well as part of the dissolved P and therefore appeared more suitable than other tests which extracted either no residual PR (alkaline extractants and anion resin) or an amount independent of reactivity (acid extractants). Glasshouse studies measuring P response and uptake by ryegrass from six PRs of different reactivity were used to compare the Olsen and Resin tests. The Resin test gave better predictions of yield than the Olsen test when a calibration derived from MCP fertiliser was used. The Resin test has the potential to provide improved predictions of pasture growth response on soils fertilised with varying amounts of P sources of different solubility.

Models of nutrient cycles (PKS) in grazed pastures are used to estimate fertiliser requirements for New Zealand farmers. Work relevant to the development of these models is reviewed. Fertiliser recommendations should be based on the long-term economic maintenance rate. Short-term modifications based on current soil nutrient status (PS) estimated by either soil tests or dynamic cycling models are probably not justified unless current fertiliser use differs greatly from the estimated maintenance rate. The effect of maintenance recommendations should be monitored, and if necessary modified, by soil testing and plant analysis.

Six phosphate rocks (PRs) of varying reactivities were compared with monocalcium phosphate (MCP) in a glasshouse experiment growing perennial ryegrass (Lolium perenne cv. Nui) as the test plant on four soils of contrasting P sorption capacity and exchangeable Ca. The cumulative dry matter yield over 10 harvests showed a significant response to P application in all soils. Based on relative yield and P uptake, MCP was the most effective P fertilizer followed by the reactive phosphate rocks, which were superior to the unreactive rocks in all soils. The relative agronomic effectiveness (RAE) and substitution ratio (SR) of individual PR fertilizers, calculated with respect to MCP using the methods of ‘vertical’ and ‘horizontal’ comparison, respectively, were similar over a range of fertilizer rate. There was a decline or slight increase in the performance of PRs with time in the low P sorption soils but a consistent increase in the high P sorption soils. Some initial influence of exchangeable Ca content of the soils on the relative performance of PRs was also observed. Generally the PRs performed better in high P sorption soils than low P sorption soils and in low exchangeable Ca soils than high exchangeable Ca soils.

A glasshouse experiment was conducted on four soils contrasting in P sorption capacity and exchangeable Ca content with perennial ryegrass using six phosphate rock (PR) sources and a soluble P source applied at four rates (including a control). After three harvests (11 weeks) replicate pots of each treatment were destructively sampled and Olsen P and mixed cation-anion exchange resin (Resin P) extractions carried out. The remaining replicated treatments were harvested another seven times (during 41 weeks). Yields (for the last seven harvests) were expressed as percentages of the maximum yield attainable with MCP.

New Zealand topsoil samples from 108 soils were analysed for pH, phosphate retention, exchangeable cations (calcium (Ca), magnesium (Mg), potassium (K)), and phosphate-extractable sulphur (S) at MAF Technology, Ruakura Agricultural Centre, Hamilton, New Zealand and DSIR Land Resources (DLR), Lower Hutt, New Zealand and results compared. Ruakura determinations followed their standard soil test procedures for fertiliser advice, which meant that extraction times were significantly less than those of DLR for the determination of phosphate-extractable S and particularly for the cations Ca, Mg, and K. There were other specific method differences also. However, the methods used by both laboratories for pH and phosphate retention were closely comparable. Linear regressions (significant at 0.1%) best described the relationships between Ruakura and DLR values, for all determinations except phosphate-extractable S where an exponential curve (also significant at 0.1%) best described the relationship. The poorest relationships were obtained for Ca and phosphate-extractable S. The results obtained for pH and phosphate retention were closely comparable. Converting quick test values for Ca, Mg, and K to a mass basis, using bulk density values, improved the relationships with DLR values, particularly for Ca, although quick test values for these cations were usually markedly lower than equivalent DLR values. Ruakura S values were generally higher than DLR values at lower absolute levels, with the reverse the situation at higher levels. These differences could largely be explained in terms of differences in analytical methods. Although all significant at 0.1%, 95% confidence limits indicated that the Ca, Mg, K, and S relationships were of little use for predictive purposes.