M.J.D. Hack-ten Broeke

Nitrate leaching affected by management options with respect to urine-affected areas and groundwater levels for grazed grassland

Simulations were performed to quantify the effects of management options on nitrate leaching to the groundwater in grazed pastures. At the experimental farm for sustainable dairy farming ‘De Marke’, experimental data on soil water and nitrates were gathered for two fields during the years 1991–1995. These data were used for model validation. The simulations showed that a detailed type of precision agriculture, which can identify urine-affected areas in the field and then subsequently omit fertilizing such areas, resulted in considerable reductions of simulated nitrate concentrations in the soil water, especially on an intensively grazed and relatively dry site with groundwater levels between 0.5 and 2.8 m. On the wetter site, the maximum calculated reduction in nitrate concentrations was 11%, but for the relatively dry site the maximum calculated reduction was as high as 41%. The second simulated option involved the raising of groundwater levels, which usually also resulted in a decrease in simulated nitrate concentrations. Under wet conditions, the groundwater level increase may cause water excess and a deterioration in conditions for crop growth and thus, less N-uptake by the crop, which would ultimately lead to increased nitrate leaching. The combined effect of non-fertilization of urine patches and the raising of groundwater levels usually resulted in higher simulated reductions of nitrate concentrations than the single options. When the effect of within-field variability was also considered, the raising of groundwater levels was most effective in reducing nitrate concentrations on the wet site, while on the relatively dry and intensively used site, the non-fertilization of urine-affected areas had the dominant effect. The study shows how simulation modelling can assist in identifying promising management strategies.

Nitrate leaching to groundwater at experimental farm 'De Marke' and other Dutch sandy soils

This study focuses on nitrate leaching to the groundwater as a result of the land use system of experimental farm De Marke, translated to other sandy soils in the Netherlands. The land use was extrapolated to five major sandy soil map units, selected from the 1: 50 000 Soil Map of the Netherlands, using simulation models. To allow extrapolation to other conditions, the land use system was described in terms of decision rules for fertilization, grassland management, sowing and harvesting of silage maize, and supplementary irrigation. The decision rules were used as input to simulation models that were calibrated and validated using data from De Marke. Then, simulations were performed for 30 consecutive years. For De Marke the calculated annual average nitrate concentration at 1 m depth was 67 mg 1-1. The nitrate concentration for the soil map unit covering the largest area of sandy soils in the Netherlands (Cambic Podzol, Hn21-VI) would be higher than for De Marke, but for the four other soil map units the calculated nitrate concentrations would be lower. It was concluded that the land use system of De Marke would result in annual average nitrate concentrations lower than 50 mg 1-1 at several locations in the Dutch sandy areas.