Effects of a catch crop on leaching of nitrogen from a sandy soil: Simulations and measurements

The effects of an undersown catch crop on the dynamics and leaching of nitrogen in cropping systems with spring cereals were investigated in southern Sweden. Field measurements of soil mineral nitrogen and nitrogen concentrations in drainage water were made for 4 years in a sandy soil. The experiment was performed on four tile-drained field plots sown with spring cereals. On two of the plots, Italian rye grass was undersown and ploughed down the following spring during three of the years. The other two plots were treated in a conventional way and served as controls. Soil nitrate levels were substantially reduced in the catch-crop treatment, but increased during the fourth year when no catch crop was grown. The differences between the treatments in soil nitrate were reflected in the nitrate concentrations measured in the drainage water. A mathematical model was used to simulate nitrogen dynamics in corresponding treatments. There was good agreement between measurements and simulations with regard to patterns of change in soil mineral nitrogen and nitrate concentrations in drainage water for each treatment. Simulated leaching of nitrate in the conventional treatment was 1.9–3.9 g N m^–2 y^–1 during the first three years while calculated leaching based on the measurements was 2.7–4.4 g N m^–2 y^–1. In the catch-crop treatment leaching of nitrate was reduced by 1.4–2.6 g m^–2 y^–1 according to the simulations and by 2.2–4.1 g m^–2 y^–1 according to calculations based on the measurements. Measurements showed that leaching of nitrogen compounds other than nitrate was hardly affected by the catch crop. In the simulations the ploughed-down catch crop resulted in temporary increases of the litter pool, a net increase of the humus pool and a reduced C-N ratio of the litter pool. Simulated net mineralization from the litter pool was substantially higher in the catch-crop treatment compared with the conventional treatment. In the fourth year, the yield of the main crop was 20–25% higher in the catch-crop treatment, and leaching was higher than in the conventional treatment.     

Simulation of field scale denitrification losses from soils under grass ley and barley

Denitrification losses from soils under barley and grass ley crops were simulated. The model, which includes the major processes determining inputs, transformations and outputs of nitrogen in arable soils, represents a scale compatible with information generally available in agricultural field research. The denitrification part of the model includes a field potential denitrification rate and functions for the effect of soil aeration status, soil temperature and soil nitrate content. Easily metabolizable organic matter is assumed not to limit denitrification. Simulated values were compared with denitrification measurements made during two growing seasons in the barley and grass ley treatments of a field experiment in central Sweden.Calibration revealed that the optimal parameter values describing the effect of soil aeration on denitrification rates were similar for both treatments. The response function derived agreed well with two data sets found in the literature. The potential denitrification rate constant, derived in the simulations, was higher for grass ley than for barley, which was consistent with the differences in overall rates of carbon and nitrogen turnover found between treatments.The simulated mean denitrification rates for the two seasons were within 20% of the mean of the measured values. However, simulated denitrification showed less temporal variability and a less skewed frequency distribution than measured denitrification. Some of the measured denitrification events not explained by the model could have been due to the stimulating effects of soil drying/wetting and freezing/thawing on microbial activity.     

Simulated and measured nitrogen conditions in a manured and fertilised soil

The governing factors for soil nitrogen dynamics were identified with a simulation model. In addition, the model was used to interpret measurements from a plot fertilisation experiment in southwest Sweden.Simulated moisture and temperature conditions were the driving variables for the simulation of soil nitrogen dynamics and leaching during a 6-year period. The results of the simulation were compared with monthly observations on two plots with grain crops, one with liquid manure and commercial fertilisers applied and one with commercial fertilisers only.Simulated temporal variations of the nitrate and ammonium storages generally agreed with observations. The dominant role of the crops as a determinant of soil nitrogen conditions was demonstrated. A higher leaching loss from the plot with application of commerical fertilisers only occurred both in simulations and measurements compared to the plot with application of both commercial fertilisers and manure. The main reason was the higher N-application in the former treatment.The effect of water flows in macropores was interpreted as a delay of simulated leaching compared to observed leaching on some occasions in summer and early autumn. No direct effect of the macropores on the yearly rates of leaching could be seen.     

Changes in soil mineral nitrogen during and after 3-year and 5-year set-aside and nitrate leaching losses after ploughing out the 5-year plant covers in the UK

The management and effects of 3-year and 5-year set-aside covers on soil mineral nitrogen (SMN, 0.0–0.9 m) were studied at six sites in England. Soil mineral N was measured annually in autumn and spring during the period of set-aside cover, with more frequent SMN sampling over the first winter after ploughing out the covers. Spring SMN was measured in the second year after set-aside. Nitrate leaching losses were also measured at three sites in the first winter after destruction of the 5-year set-aside covers. Winter cereals were grown in both test years after each set-aside period.Amounts of both autumn and spring SMN in the perennial rye-grass (PRG), perennial rye-grass/white clover (PRG/WC) and natural regeneration (NR) covers were generally less than, or similar to those in the continuous arable treatment during each year of set-aside, indicating a slightly smaller nitrate leaching risk under set-aside management. Slight increases in autumn SMN, and hence leaching potential were, however, observed under PRG/WC in the fourth and fifth years, compared with continuous arable cropping.Ploughing out of both 3-year and 5-year covers increased soil N supply and potential nitrate leaching losses over winter, compared with continuous arable cropping. By the following spring, mean increases across all sites in amounts of SMN after 3-year covers of PRG, NR and PRG/WC were 14, 18 and 33 kg ha^–1 N, respectively, compared with the arable rotation. Equivalent increases in spring SMN following destruction of the 5-year set-aside covers were almost identical, at 17, 19 and 33 kg ha^–1, respectively, although only the ploughed-out PRG/WC covers increased SMN at the clay sites. Measured nitrate leaching losses in the first winter after 5-year set-aside were greatest after PRG/WC at two sites on shallow chalk but greatest after NR, which had a naturally large clover content, at the third site which was on a sandy soil. However, the leaching losses after set-aside were relatively small, relative to typical losses after ploughing out intensively managed grass or grass/clover swards, and would have been compensated for by potentially less leaching during set-aside.Spring SMN measurements in the second year after ploughing out the set-aside covers, showed negligible or, for PRG/WC, only slight increases (12 – 18 kg ha^–1) in residual soil N supply after both 3-year and 5-year covers, compared to continuous arable cropping. The extra N mineralisation after cover destruction justified small reductions in fertiliser N inputs for the first, but not second crop following either 3- or 5-year set-aside, unless the cover had contained a large clover content. Both 3-year and 5-year set-aside covers had minimal or no effect on either organic matter content, apart from a slight increase in the PRG/WC treatments, or extractable phosphorus, potassium and magnesium status in the topsoil.     

Soil denitrification in three cropping systems characterized by differences in nitrogen and carbon supply

Rates of denitrification measured over a growing season, both within and between rows of plants, in grass and lucerne leys and in barley plots were related to soil moisture and nitrate levels using bivariate, nonlinear regression models. Both within and between rows in all three crops, moisture explained a significant fraction of the variation in denitrification rates when the rates were regressed as increasing exponentially with increasing soil water content under moist conditions. Only soil moisture explained a significant fraction of the variation of the denitrification rates measured in the soil cores taken in (19.5%) and between (46.3%) plant rows at the barley field and in (42.4%) the plant rows in the lucerne ley. In some treatments, the rates appeared to be unrelated to moisture below a critical moisture threshold.Nitrate proved to be a useful variable in predicting denitrification rates in the grass ley and between lucerne rows. Water and nitrate could explained 36.6% of the within-row variation and 24.2% of the between-row variation in the grass ley and 65% of the between-row sum of squares.An attempt to use the regression model based on data from one year for the grass ley to predict the losses during another year for the same crop was not successful.     

Strategies to decrease nitrate leaching in the Brimstone Farm Experiment,Oxfordshire, UK, 1988–1993: the effects of winter cover crops and unfertilised grass leys

Nitrate losses in drainflow were measured over five years on eight hydrologically isolated field plots, pairs of which had the following cropping regimes: (a) a 3-yr unfertilised, ungrazed grass ley followed by winter and spring cereals, (b) mixed cropping including winter cover crops, spring cereals, winter cereals, winter fallow and spring beans, (c) a similar sequence to (b) but with a winter fallow replacing the cover crop in the first year and a winter cover crop replacing the fallow in the third year, and (d) continuous winter cereals (control plots). Less nitrate was lost in winter drainflow from winter cover crops than from the winter fallows, but over all five years less nitrate was leached from the continuous cereal plots than from those with mixed cropping. Most of the extra nitrate lost from the mixed cropping regimes probably resulted from mineralisation of the cover crop residues, which occurred at times when subsequent crops could not take advantage of the mineral nitrogen released. Crops grown after the grass ley and cover crops did not benefit from their residues, in terms of either grain yield or of total nitrogen uptake. We conclude that on heavy clay soils in UK a cropping regime of continuous winter cereals offers the best compromise between profitable crop production and minimised nitrate loss to surface waters.     

Significance of nitrate leaching and long term N immobilization after deepening the plough layers for the N regime of arable soils in N.W. Germany

Field studies were conducted to assess the turnover and the leaching of nitrogen in arable soils of Lower Saxony (NW Germany). The mean surplus N (difference between N inputs by fertilization and N export by the yield; 146 field plots) from 1985–1988 amounted to 38 kg ha^-1 yr^-1 in fine textured (clay, loam, silt) and to 98 kg ha^-1 yr^-1 in coarse (sandy) soils. Leaching of nitrate calculated by a simple functional model for simulation of the N regime over the winter period (i.e. mineralization and leaching) was 16 kg ha^-1 in the fine and 63 kg N ha^-1 in coarse soils (mean values of the winter periods 1985–1988 from 256 plots).Before the 1960s, the depth of the A_p horizons rarely exceeded 25 cm in arable soils of the former FRG. During the last three decades, ploughing depth has increased to at least 35 cm. The mass balance calculations for total N after ploughing to 35 cm in loess soils of southern Lower Saxony (105 farm plots) yielded a mean increase in total N by about 900 kg ha^-1 in 20 years. With respect to soil organic matter equilibria, N accumulation will continue for at least another 10 years on 67% of the examined farm plots. This study suggests that long term N immobilization is one of the most important sinks for nitrogen in arable soils of Germany. For simulation of the N dynamics over the growing season and for long time periods total nitrogen dynamics need to be considered.     

Effects of long-continued treatment on the mineral nitrogen content and mineralisable nitrogen of soil from selected plots of the Broadbalk experiment on continuous wheat,Rothamsted

Summary The changes in the ammonium-N and nitrate-N contents of bare fallow and soil under the first and third crops of winter wheat after fallow were followed on plots of Broadbalk Field, Rothamsted, which have received for each crop 14 tons farmyard manure (FYM) per acre, complete minerals (P, K, Na, Mg), or complete minerals + nitrogen fertilizers.More mineral N was produced during fallow on the plot receiving FYM than on the other plots. Soil under wheat also contained more mineral N on the FYM plots than elsewhere. Nitrogen fertilizers applied in the spring temporarily increased the mineral-N content of the soil, but were rapidly removed by the crop. Ammonium sulphate applied in the autumn was lost from the surface soil by the following March through nitrification and leaching.Twice as much mineral-N was produced when soil from the FYM plot was incubated as when soils from other plots were similarly treated. Nitrate formed during fallow was leached into the subsoil during the autumn and winter, and recovered by the wheat during the following spring and summer. Its existence is not detected by sampling the surface soil, nor by an incubation test. This source of nitrogen complicates the use of laboratory measurements to assess the fertilizer nitrogen required by winter wheat. Since the crop removed mineral N from the surface soil by March, estimation of the amount then present was also of no value for making fertilizer recommendations.     

Nitrate nutrition ofDeschampsia flexuosa (L.) Trin. in relation to nitrogen deposition in Sweden

Summary Current and maximally induced nitrate reductase activity (NRA), total-N, nitrate, K, P, Ca, Mg, Mo and sucrose in leaves ofDeschampsia flexuosa was measured three times during the vegetation period in forests along a deposition gradient (150 km) in south Sweden, in north Sweden where the nitrogen deposition is considerably lower, and at heavily N-fertilized plots. In addition, the interaction between nitrogen nutrition and light was studied along transects from clearings into forest in both south and north Sweden. Plants from sites with high nitrogen deposition had elevated current NRA compared to plants from less polluted sites, indicating high levels of available soil nitrate at the former. Current NRA and total N concentration in grass from sites with high deposition resembled those found at heavily N-fertilized plots. Under such circumstances, the ratio current NRA: maximally induced NRA as well as the concentration of nitrate was high, while the concentration of sucrose was low. This suggests that the grass at these sites was already utilizing a large portion of its capacity to assimilate nitrate. Light was found to play an important role in the assimilation of nitrate; leaf concentration of sucrose was found to be negatively correlated with both nitrate and total N. Consequently, grass growing under dense canopies in south Sweden is not able to dilute N by increasing growth. The diminished capacity of the grass to assimilate nitrate will increase leaching losses of N from forests approaching N saturation.     

Observations on take-all and eyespot diseases of wheat in Yorkshire

Wheat crops were surveyed in Yorkshire from 1944 to 1946 on farms where crops were reported to be unsatisfactory. Take-all ( Ophiobolus graminis (Sacc.) Sacc.) and eyespot ( Cercosporella herpotrichoides Fron.) were both found to be present; the latter is the most serious trouble on the better wheat lands.
The variation in the incidence of these diseases on selected farms during the three seasons has been compared.
The effect of rotation has been examined; both diseases were found to be encouraged by too frequent cropping with wheat or barley. The incidence of disease in wheat crops following a 1-year seeds ley was found to be influenced by the nurse crop used to undersow the seeds, and also by the time at which the ley was ploughed up. Oats were found to be preferable to either wheat or barley as a nurse crop, and in 1946 late ploughing considerably reduced the amount of eyespot disease present in the following wheat crop.
A 1-year ley is considered to be of too short a duration to ensure the disappearance from the soil of Ophiobolus graminis and Cercosporella herpotrichoides surviving from the previous crop, and a period of 2–3 years is suggested as being desirable.     

Modeling long-term compost application effects on nitrate leaching

Application of compost to agricultural soils may be beneficial for crop production by increasing soil fertility and supplying plant nutrients, however, any raise of the soil organic matter content may increase the potential for unintended groundwater contamination by nitrate leaching. In this paper, the effect of long-term compost applications on nitrate leaching, soil organic matter content, and crop production is analyzed using results of simulation scenarios for agricultural sites with loamy and sandy soils. Simulations were carried out using the Danish Nitrogen Simulation System (DAISY) which describes the nitrogen balance in a one-dimensional soil-plant-atmosphere system and considers compost type and application rate as well as management and cropping practice. Estimations of hydraulic and solute transport parameters are based on pedotransfer functions. Data from a 4.5-year period of field experiments with compost applications in northern Germany as well as from laboratory experiments with compost amended soil are used for model calibration. Simulation results suggest that: (i) with respect to nitrogen turnover the differences between compost types (i.e., non-matured and matured) are small compared to site-specific properties (i.e., soil temperature and water balance) and management practices (i.e., crop rotation) when considering a compost application period of 50 years; (ii) with respect to nitrate leaching the effect of different compost application scenarios is highly sensitive at the sand and relatively small at the loam site; and (iii) relatively high crop yields and acceptably low nitrate concentrations in the drainage water are obtained at the sand site when applying a combination of 10 t compost/ha/yr and soil- N_min adjusted mineral nitrogen additions of about 20 kg/ha/yr to a winter-grain dominated crop rotation. Further optimization may be possible by applying reduced rates of the (economically interesting) non-matured compost. This revised version was published online in June 2006 with corrections to the Cover Date. This revised version was published online in June 2006 with corrections to the Cover Date. This revised version was published online in June 2006 with corrections to the Cover Date.     

Weed performance in crop rotations with and without leys and at different nitrogen levels

Weed populations were studied from a 26-year-old field experiment in southern Sweden with three different 6-year crop rotations, each with four rates of nitrogen application. The rotations differed in that one had a two-year legume-grass ley, another had a two-year grass ley, and that the third had spring wheat followed by a repeatedly harrowed fallow. The leys and the fallow were followed by turnip rape, winter wheat, oats and barley which was undersown in the two ley rotations. Data on weed biomass, collected in one season, were subjected to multivariate analysis.
Winter turnip rape had the highest weed biomass. However, of the several weed species, only Matricaria perforata Merat was important in wheat (the crop following turnip rape in the rotation). The weed flora did not differ consistently between rotations. We conclude that none of the three rotations had developed any major weed problems under the past weed management regime (herbicides applied to cereal crops).
There was no consistent effect of nitrogen fertilisation on total weed biomass in any of the three rotations. However, when comparing the weed floras in winter wheat, turnip rape and oats, the unfertilised plots differed from the plots receiving nitrogen. In the two latter crops, the abundant, low-growing annual Stellaria media (L.) Vil. performed best in fertilised plots with dense stands. Equisetum arvense L., the most abundant perennial weed, was important only in unfertilised plots.     

Growth analysis of soil-grown spinach plants at different N-regimes

Spinach plants were grown in pots under controlled conditions in three different soils (a loamy sand, a silt loam at low mineral-N level and a silt loam at the double mineral-N level). The nitrogen uptake pattern varied considerably between the three soil types and was used to validate an equation between the relative growth rate and nitrogen content. This equation is based on the growth response of spinach plants grown hydroponically at equal environmental conditions either at optimum nitrogen supply (complete nutrient solution) or with a relative nitrate addition rate of 0.30 day^–1, 0.225 day^–1 or 0.15 day^–1 effecting an exponential increase in nitrogen uptake. Growth in potted soil was slightly overestimated. Part of this bias was explained by the lower shoot weight ratio observed for the soil grown plants. This was demonstrated by the improvement in growth predictions when using net assimilation rate rather than relative growth rate as the driving variable in the model.     

Leaching losses of nitrogen from a clay soil under grass and cereal crops in Finland

Summary A 16-plot experimental field was established in 1975 on a clay soil in Jokioinen, Finland. The water discharge through tile drains was measured and its ammonium and nitrate N contents determined for each plot separately. The surface runoff was also measured and analysed. The annual runoff and the N leached from the surface of moderately fertilized (100 kg/ha/y N) cereal plots varied during 1976–1982 from 21 to 301 mm and from 2 to 7 kg/ha, respectively. The discharge of water and leaching of N through subdrains varied from 65 to 225 mm and from 1 to 38 kg/ha, respectively. The highest leaching was probably caused by a previous fallow. The annual N uptake by the crop varied between 41 and 122 kg/ha.Of the fertilizer-N used for cereals, 20% of that applied in the autumn was lost, but only 1 to 4 per cent was lost when applied in the spring. There was much less N leaching from ley than from barley plots, although the former was given twice as much N. The rate of N fertilization had only a very slight effect on N leaching from both ley and barley plots.The results were compared with those obtained in lysimeters filled with clay, silt, sand and peat soils. No definite conclusions can be drawn because the lysimeter experimental data are only for the first year.     

Modelling and measuring the effect of nitrogen catch crops on the nitrogen supply for succeeding crops

Nitrogen catch crops are grown to absorb nitrogen from the rooting zone during autumn and winter. The uptake of N (N_upt) from the soil inorganic N pool (N_min) to a pool of catch crop nitrogen, will protect the nitrogen against leaching. After incorporation, a fraction (m) of the catch crop nitrogen is mineralized and becomes available again. However, not all available nitrogen present in the soil in the autumn is lost by leaching during winter. A fraction (r) of the nitrogen absorbed by the catch crop would, without a catch crop, have been retained within the rooting zone. The first year nitrogen beneficial effect (N_eff) of a catch crop may then be expressed b N _eff = m*N _upt - r* N _upt The soil-plant simulation model DAISY was evaluated for its ability to simulate the effects of catch crops on spring N_min and N_eff. Based on incubation studies, parameter values were assigned to a number of catch crop materials, and these parameter values were then used to simulate spring Nmin. The model was able to predict much of the vairiation in the measured spring Nmin (r² = 0.48***) and there was good agreement between the measured and the simulated effect of winter precipitation on spring N_min and Neff.Scenarios including variable soil and climate conditions, and variable root depth of the succeeding crop were simulated. It is illustrated that the effect of catch crops on nitrogen availability for the succeeding crop depends strongly on the rooting depth of the succeeding crop. If the succeeding crop is deep rooted and the leaching intensity is low, there is a high risk that a catch crop will have a negative effect on nitrogen availability. The simulations showed that the strategy for the growing of catch crops should be adapted to the actual situation, especially to the expected leaching intensity and to the rooting depth of the succeeding crop.     

DNDC模型在农田氮素渗漏淋失估算中的应用

采用田间原装渗漏计测定了山东省济南市典型种植模式下,2008年冬小麦整个生长季水分和 氮素淋失量,并利用该数据对DNDC（DeNitrification-DeComposion）模型进行检验和敏感性分析.结果表明：模型对该地区农田土壤水分运动的模拟效果较好,模拟结果的精度总体上可以接受,但对氮素淋失量的模拟结果存在一定偏差,冬小麦整个生长季的氮素淋失量模拟值（18.35 kg N·hm^-2）比实测值（14.89 kg N·hm^-2）高3.46 kg N·hm^-2,相对偏差在20%左右,模型参数尚需作进一步调整.模型敏感性分析显示,农田氮素淋失更易受到灌水量和施肥量的影响.DNDC模型在该地区有一定的适用性.     

Vegetation development and nitrogen dynamics of sown and spontaneous set-aside on arable land

The area of arable land devoted to ecological compensation in Switzerland has increased markedly in the last decade. We studied the influence of plant cover upon nitrogen (N) dynamics in the uppermost soil layer in experimental plots representing three types of ecological compensation area (set-aside sown with two different seed mixtures and without sowing) and a grass-clover ley. Potential losses, apparent uptake and net mineralisation of N were investigated during critical periods: during the early development of the vegetation, after a first mowing, and before and after ploughing and sowing with winter wheat.The four plant-cover types differed in species richness as well as in the proportions of forbs, grasses and legumes, and annuals and perennials. N dynamics varied with time of sampling, but differences between plant-cover types were small, and neither early mowing nor ploughing had a significant effect. We conclude that as far as N dynamics in the uppermost soil layer is concerned, both species-poor and species-rich seed mixtures and natural regeneration can be recommended for set-aside.     

Assessing nitrate leaching during the three‐first years of Miscanthus × giganteus from on‐farm measurements and modeling

Miscanthus × giganteus is often regarded as one of the most promising crops to produce sustainable bioenergy. This perennial crop, renowned for its high productivity associated with low input requirements, in particular regarding fertilizers, is thought to have low environmental impacts, but few data are available to confirm this. Our study aimed at assessing nitrate leaching from Miscanthus × giganteus crops in farmers' fields, thus including a wide range of soil and cropping system conditions. We focused on the first years of growth after planting as experimental studies have suggested that Miscanthus × giganteus, once established, results in low nitrate leaching. We combined on‐farm measurements and modeling to estimate drainage, leached nitrogen, and nitrate concentration in drainage water in 38 fields located in Center‐East France during two winters (November 2010 to March 2011, November 2011 to March 2012). Nitrate leaching and nitrate concentration in drainage water were on average very low. Nitrate leaching averaged 6 kg N ha^?1 whereas nitrate concentration averaged 12 mg l^?1. These low values are attributable to the low estimates of drainage water (mean = 166 mm) but also to the low soil mineral nitrogen contents measured at the beginning of winter (mean = 37 kg N ha^?1). Our results were, however, very variable, mainly due to the crop age: nitrate leaching and nitrate concentration were critically higher during the winter following the first growth year of Miscanthus × giganteus, reflecting the low development of the crop. This variability was also explained by the range of soil and cropping conditions explored in the on‐farm design: shallow and/or sandy soils as well as fields where establishment failed had a higher risk of nitrate leaching.     

Is low rooting density of faba beans a cause of high residual nitrate content of soil at harvest ?

It was the aim of this study was to evaluate the hypothesis that low rooting density of faba beans is the major reason for the comparable low depletion of N_min-nitrogen from the rooted soil volume during the vegetation period. Therefore a simulation study was carried out using data from a two-year field experiment with faba beans and the reference crop oats. Since the nitrate dynamics in the soil is closely coupled with the water budget, the model simulated also the water uptake by plants, movement and content in the soil applying a numerical solution of the Richard's equation. The nitrogen budget part of the model includes calculation of vertical nitrate movement in the soil, mineralisation of nitrate from organic matter and nitrate uptake by the crop. Vertical nitrate movement was simulated with the convection-dispersion equation. Mineralisation was computed from a simple first order kinetic approach using only one fraction of mineralisable organic matter. Nitrate uptake was assumed to be determined either by the nitrogen demand of the crop, which was estimated from a logistic growth equation that was fitted to measured data of N-accumulation, or by the maximum nitrate transport rate towards the root surface. The latter was computed from a steady state solution of the diffusion - mass flow equation for cylindrical co-ordinates.For oats the model calculated a maximum nitrate transport rate towards roots that was quite close to the measured N-uptake of that crop. For faba beans, however, the calculated maximum nitrate transport towards roots was much lower than total N-uptake and lower than for oats. Consequently, simulated N_min-contents below faba beans were during the growing season about 20-30 kg N ha^–1 higher than below oats. This difference matches quite close with the observed differences between the two crops. Therefore it was concluded that low nitrate uptake resulting from low rooting density is the main reason for higher residual nitrate contents below faba beans at harvest time.