The spatial variability of soil nitrates in arable and pasture landscapes: implications for the development of geographical information system models of nitrate leaching

Abstract. In response to the European Community Nitrate Directive (91/676) a catchment scale Geographical Information System (GIS) model of nitrate leaching has been developed to map nitrate vulnerability and predict average weekly fluxes of nitrate from agricultural land units to surface water. This paper presents a pilot study which investigated the spatial variability of soil nitrates in order to: (1) define an appropriate pixel size for modelling N leaching; (2) quantify the within-unit variability of soil nitrate concentrations for pasture and arable fields; and (3) assist in the design of an efficient sampling strategy for estimating mean nitrate concentrations. Soil samples, taken from two 800 m transects in early September 1994, were analysed for water soluble nitrate. The arable soils had a mean nitrate-nitrogen concentration of 0.693 μg/g (S.E. 0.054 μg/g) and the pasture soils had a higher mean nitrate-nitrogen concentration of 0.86 μg/g (S.E. 0.085 μg/g). Spatial variability was investigated using variograms. The pasture data had a weak spatial relationship, whereas the arable data exhibited a strong spatial relationship which fitted a spherical variogram model (r² 0.87), with a range of 40 m. A pixel size of 40 m is suggested for nitrate modelling within the GIS based on the arable variogram and an improved sampling strategy for model validation is suggested, involving bulking sub-samples over a 40 m grid for estimating mean nitrate concentrations in combined land use and soil units.     

Nitrate leaching losses under Miscanthus grass planted on a silty clay loam soil

Abstract. Nitrate leaching under newly planted Miscanthus grass was measured for three years. The crop received either no fertilizer-N or an annual spring application of 60 kg or 120 kg N ha^-1. During three winters soil water was collected from porous cup probes installed 90 cm deep. Nitrate leaching was calculated from the mean drain flow recorded in two drain gauges multiplied by the mean nitrate-N concentration in the soil water solutions collected. In the first year soil water nitrate concentrations were high on all treatments and N losses were 154, 187 and 228 kg ha^-1 respectively on the unfertilized treatment and those that received 60 or 120 kg N ha^-1. Leaching losses in the second and third years were, in turn, 8, 24 and 87 kg ha^-1 and 3, 11 and 30 kg ha^-1 for the unfertilized treatment and for the 60 and 120 kg N ha^-1 treatments respectively. Leaching losses were closer to those recorded under extensively managed grassland than arable land. The large losses in the first year were probably due to the previous agricultural management at the site and excessive inputs of N on the fertilized plots. In the second and third year, lower drainage volumes may also have influenced losses. The results show that Miscanthus , once established, can lead to low levels of nitrate leaching and improved groundwater quality compared with growing arable crops.     

Nitrate leaching from a small catchment in central England

Abstract. Flow and nitrate concentrations were measured weekly for four years at twelve stream-water monitoring sites in a catchment in the English Midlands designated as a Nitrate Advisory Area. Farm surveys and satellite images have provided soil and land use information. Measurements show the nitrate load to be dominated by discharge, with large variability due to differing weather conditions from year to year. Within-year variability in nitrate concentrations is also related to weather conditions, with high concentrations when field capacity is reached if this occurs late in the year. There is also clear evidence of dilution of nitrate during intense storms. The effect of changing weather conditions makes it impossible to identify catchment-scale changes in leaching due to changes in agricultural practice over a period as short as four years. Measurements from a major spring in the catchment show an increasing trend in nitrate concentrations through the period. There is some evidence that the greatest N leaching to streams in the catchment is associated with intensive grassland on soils which are naturally poorly drained.     

Simulating nitrate concentrations in water draining from a small arable catchment in England

Abstract. The catchment simulated comprises 57 hectares of heavy clay soil managed as six arable fields. Mole and pipe drains carry surplus water into two ditches, one feeding into the other. Their combined flow was passed through a flume with an automatic water sampler, samples from which were analysed for nitrate. Measurements of nitrate concentration made during periods of water flow from 1990 to 1993 were simulated using a model comprising sub-models for leaching, mineralization, nitrogen uptake by crops and subsoil denitrification. The simulations were plotted against the measurements. For statistical evaluation, the correlation coefficient was used to assess the degree of association between the measurements and the simulations and the mean difference to assess the agreement. The correlation between the simulations and the measurements was significant in two of the three seasons, but the mean difference was significant in all three. However, taking all three seasons together gave a very highly significant correlation and a non-significant mean difference.     

Investigating preferential flow in a large intact soil block under pasture

Abstract. A large soil block was constructed to determine the importance of preferential flow routes compared with matric flow pathways at a pasture site in mid-Devon. The sandy loam soil was well structured and uniform. The soil block measured 5 m×3 m×1 m and was instrumented with an array of 54 tensiometers, TDR wave guides and suction samplers connected to an in situ chloride analysis system. Four steady state irrigation experiments were conducted with a range of rainfall intensities. During each experiment chloride and nitrate tracers were applied and the patterns of movement were observed. Although the application of tracer was uniform and the soil was relatively homogeneous, there was large variability across the block in terms of time taken to reach the peak concentration (TPC) and the peak concentration itself. About 44 samplers operated at the greatest intensities (10–2 mm h⁻¹) and only 35 at the smallest (1 mm h⁻¹). No relationship was found between TPC and depth. The fastest TPC and largest concentrations were associated with the greatest rainfall intensities. Relative importance of the individual water pathways was a function of soil heterogeneity: parts of the soil block were highly active with several pathways having short TPCs and conductivities in excess of 4 m day⁻¹ whereas other areas had longer TPCs and conductivities of 1–2 m day⁻¹. The pattern was also dynamic, with conductivities of the pathways changing through time, though most of the faster pathways maintained their greater conductivities for more than one year.     

Development of improved advice for farmers and advisers

Abstract. The MAFF Nitrate Programme has provided policy makers and the agricultural industry with a much improved understanding of soil nitrogen cycling, and the cost-effectiveness of a range of nitrate reduction strategies. This understanding has been disseminated as improved economically-based advice on fertilizer N inputs to arable and grass crops, and information on the N value of livestock manures as affected by timing and method of application. In addition, strategies for reducing nitrate losses have been developed, including the use of over-winter cover crops, the management of cultivations and crop residues, and the conversion of arable land to low-input grassland.
A wide variety of information dissemination methods have been used, including face-to-face discussions, demonstrations, articles in the farming press, booklets aimed at farmers, technical information for advisers and consultants, recommendation reference books and computer-based fertilizer recommendation systems.     

The effects of crop husbandry and nitrogen fertilizer on nitrate leaching from a shallow limestone soil growing a five course combinable crop rotation

Abstract. Nitrogen (N) leaching losses from a shallow limestone soil growing a five course combinable croprotation (oilseed rape, wheat, peas, wheat, barley) were measured from 1990 until 1995 using porous ceramic cups, at 60 cm depth, and drainage estimates. The crops were grown with three husbandry systems and two levels of N fertilizer. The husbandry systems were designed to reflect local practice (Standard), the best possible techniques to reduce N loss (Protective) and an Intermediate system which was a compromise between the two. Nitrogen was applied at full and half recommended rates. Drainage started during September in four years and November in one year, with above average drainage in three years. Losses of N were largest after peas (58 kg/ha) and oilseed rape (42 kg/ha), and least (17 kg/ha) before peas sown in spring after a cover crop. Over five years, the Protective management system, which used early sowing and shallow cultivation wherever possible, lost least N (31 kg/ha/y) and the Standard system, with conventional drilling dates and ploughing as the primary cultivation, lost most (49 kg/ha/y). Halving the N fertilizer decreased N loss by 11 kg/ha/y, averaged over the rotation. None of the treatments gave mean drainage water nitrate concentrations of less than 50 mg/l, averaged over the five years. Changes to arable cropping alone will not eliminate the need for other measures to control nitrate concen-trations in public drinking water supplies.     

Effects of overwinter cover on nitrate loss and drainage from a sandy soil: consequences for water management?

Abstract. A lysimeter study from April 1993 to June 1997 assessed the effects of winter cover crops and unfertilized grass on both the volume of water draining over winter and the amounts of nitrate leached. There were three to five replicates of each treatment in a fully randomized design. The lysimeters were undisturbed monoliths of loamy medium sand, 1.2 m deep and 0.8 m diameter. There were six treatments: sown cover before spring-sown crops (SC), natural regeneration (‘tumbledown’) before spring-sown crops (T), unfertilized grass (UG), bare soil permanent fallow, (PF), winter barley (WB) and conventional overwinter fallow before spring-sown crops (WF). Sugarbeet replaced cereals in 1996 as a disease break, and in consequence no cover was established in SC and T in autumn 1996. Of the four years of the study, two were above-average rainfall, while two were of less than average rainfall. Results are only quoted if statistically significantly different from WB (P=0.10). Over the first winter, NO₃―N losses were similar under UG (26 kg ha^?1) and PF (29 kg ha^?1), due to the slow establishment and growth of the grass. In the following three winters NO₃―N losses under UG were small (c. 6 kg ha^?1), giving an overall mean of c. 11 kg ha^?1. Sown cover crops and T gave means of c. 16 and 22 kg ha^?1 respectively, compared with c. 27–31 kg ha^?1 under PF, WB and WF. Mean NO₃―N concentrations were smallest under UG (4.4 mg l^?1) and SC (10.6 mg l^?1), although both T (13.7 mg l^?1) and PF (12.4 mg l^?1) were less than under WB and WF (15.8–18.7 mg l^?1). Overwinter drainage was greatest from UG and PF, at 239 and 247 mm respectively. In the three winters that cover crops were grown, drainage was decreased by, on average, 30 mm year^?1 compared with WF. However, there were large differences in effects between years, with significant decreases in only one year. We conclude that the widespread adoption of cover crops before spring-sown crops will reduce overwinter drainage in UK Nitrate Vulnerable Zones by no more than c. 2%, compared with no cover before spring-sown crops.