Determining critical nitrogen application rates to reduce nitrate leaching in dairy pastures

Over the past decade, there has been a rapid increase in dairy farming in New Zealand. Nitrate leaching from intensive agricultural systems, e.g. dairy farms, is suspected to cause increased nitrate concentrations in ground and surface waters. The objective of this research programme was to determine nitrate leaching losses on dairy pasture systems as affected by the application of dairy shed effluent (DSE), nitrogen (N) fertilizers and cow urine on free-draining soils. The data obtained were used to develop a simple, semi-empirical computer model to estimate nitrate leaching losses and establish critical N application rates to minimize nitrate leaching. Undisturbed soil monolith lysimeters (50-80 cm diameter and 70-120 cm depth) were collected from a free-draining Templeton fine sandy loam (Haplustepts) and were installed to a field lysimeter laboratory. N leaching losses were measured after the application of urea, ammonium chloride, DSE, and/or cow urine. The data obtained were then used to develop a simple, semi-empirical management model to estimate N leaching losses and establish critical N application rates The nitrogen leaching estimation (NLE) model is based on the concept that the annual N leaching loss is related to the annual average amount of potentially leachable N (mineral N and mineralizabe N) in the soil, and the annual drainage flux. The potentially leachable N (NPL) is determined by the sum of annual fluxes of N cycling processes including atmospheric N deposition (NAD); fertilizer or effluent N application rate (NF); biological N fixation (NB); net N mineralization (NM); animal N returns (mainly urine) to the pastures (NA); pasture N uptake (NP); volatilization losses after the application of N fertilizers or effluents (NV); and denitrification loss (ND) (Eqn 1) (kg N ha). NPL = NAD + NF + NB + NM + NA NP NV ND (1) The relationship between nitrate leaching and NPL was established on the basis of the data obtained from the lysimeter studies. The NLE model estimated that the critical N application rates (above which the NO3-N concentration in the drainage water will exceed drinking water standard of 11.3 mg NO3-N L) for grazed pasture systems (with urine input) were 160-200 kg N ha y if urea was applied, or 250-300 kg N ha if dairy shed effluent was applied. For cut pastures (i.e. no urine input), the critical application rates were about 400 for urea, and 600 for DSE.