| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Danish Institute of Agricultural Sciences, Dep. of Horticulture, Kirstinebjergvej 10, DK-5792 Årslev, Denmark
* Corresponding author (Hanne.Kristensen{at}agrsci.dk).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The efficiency of catch crops to retain NO3 differs depending on factors such as length of growing season and capacity for N uptake (e.g., Richards et al., 1996). The importance of catch crop root depth in reducing NO3 leaching to ground water has been largely neglected, with most studies of soil NO3 depletion covering only the top 1 m or less of the soil profile (e.g., Vos et al., 1998). Cultivated plants are thought to have the major part of their root system within this depth, but because of the high mobility of the NO3 ion in soil solution it is worthwhile studying root growth in deeper soil layers even if root density is low. Over the time span of a growing season, just a few roots may be enough for the uptake of large amounts of NO3 (Robinson, 1986; Strebel and Duynisveld, 1989), and thus for reducing NO3 leaching to ground water (Thorup-Kristensen, 2001).
Studies have shown that many species extend roots to considerable depths. In a literature review, the average maximum rooting depth of arable crops on a global scale was estimated to be 2.1 m (Canadell et al., 1996); another review found the expected maximum rooting depth in the range 1 to 3 m in 44 out of 53 annual crops (Borg and Grimes, 1986). Studies of the root growth of catch crops below 1-m depth have shown large differences in root depth and distribution from one species to the next (Barraclough, 1989; Materechera et al., 1993; Thorup-Kristensen, 2001).
Investigations of soil NO3 and plant N uptake have also indicated root activity and N uptake by annual crops and catch crops in soils below 1-m depth. Significant N uptake has been found at 0.9 to 1.5 m for winter wheat (Triticum aestivum L.), winter barley (Hordeum vulgare L.), sugarbeet (Beta vulgaris L. var. altissima Döll), corn (Zea mays L.), and fodder radish (Kuhlmann et al., 1989; Strebel and Duynisveld, 1989; Wiesler and Horst, 1994; Huang et al., 1996; Thorup-Kristensen, 2001). Studies from Nebraska have shown N uptake from 1.8 m by corn and from as deep as 2.4 m by sugarbeet (Gass et al., 1971; Peterson et al., 1979). These findings were based either on indirect evidence from changes in residual soil N pools, or on study of 15N uptake from different soil layers over long periods, such as a growing season. In only a few studies has N uptake been compared with root growth at more than 1-m depth (Kuhlmann et al., 1989; Strebel and Duynisveld, 1989; Wiesler and Horst, 1994; Thorup-Kristensen, 2001), and the experimental approaches in these studies were not optimal for more detailed examination of root system–plant N uptake relationships. The long-term nature of this kind of experiment has an unknown influence from N leaching and N microbial processes such as mineralization-immobilization and denitrification.
In studying root growth, the use of angled minirhizotrons enables frequent and nondestructive sampling of information on root depth and distribution (Smit et al., 2000). The method involves inserting transparent tubes into the ground and in situ counting of roots at the interface between the tube wall and the soil. The use of angled minirhizotrons has been questioned, however, in comparison with destructive core sampling, where roots are washed from the soil samples. It has been concluded that root density in deeper soil layers tends to be overestimated with the minirhizotron method, presumably due to preferential growth of roots in voids along minirhizotrons (e.g., Parker et al., 1991; Heeraman and Juma, 1993). In such studies, however, measurements rarely extend to the bottom of the root zone, and the validity of the conclusions that can be drawn from comparisons is thus weakened. A more appropriate test of the use of minirhizotrons would be to compare root depth and distribution estimated by minirhizotrons with the results of plant N uptake from different soil layers (Thorup-Kristensen, 2001). This would show the effective root depth and distribution in terms of N uptake. Plant N uptake could be measured using 15N injection methodology as described by Gass et al. (1971) and Huang et al. (1996), who injected 15N labeled NO3 solution into different soil layers to show relative differences in N uptake between layers. Contrary to these studies, such a 15N experiment could be short-term to enable comparison of root distribution with N uptake rates at a certain point in time. Furthermore, this would minimize the influences on results from N mineralization processes and N leaching.
In this study, we combined minirhizotron studies of root growth with 15N studies of N uptake from various soil layers over a 6-d period. The aims were threefold: (i) to test the relevance of root measurements obtained by the minirhizotron method against measured root system N uptake; (ii) to investigate root growth and efficiency for N uptake of three catch crops expected to have shallow, intermediate, and deep root growth; and (iii) to quantify the relationship between root distribution and short-term N uptake by catch crops in deep soil layers.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
The video films of roots were used to register three different measures of root growth: root depth, root intensity, and root frequency. Root depth was registered as the deepest root observed in each of the two counting grids on each rhizotron. Root intensity was registered as the total number of roots crossing the lines in each 40 by 40 mm cross (total of 80-mm line). This was calculated as number of root intersections per meter line (intersections m–1) in a soil layer of 34.6-mm depth [= cos(30°) x 40 mm] due to the position of the tube 30° from vertical. For calculation of root frequency, it was registered if any roots were crossing the lines in each 40 by 40 mm cross. The root frequency was calculated as the percentage of 40 by 40 mm crosses where roots had been observed within a given soil layer (0.25 m). The four measures obtained within each plot (two minirhizotrons with two grids each) were averaged for each plot. Root depth registration was initiated 4 wk after sowing and ended after the 15N experiment on 3 Nov. 2000. Root intensity and frequency were measured on 23 Oct. 2000; that is, 2 d before the 15N injection experiment. Root depth penetration rates (mm d–1 °C–1) were calculated, following Barraclough and Leigh (1984), as the slope of regression lines of the average root depth versus accumulated average daily temperature from sowing (average of hourly measurements, base temperature of 0°C).
Deep Point Nitrogen-15 Injection
Uptake of N was studied by deep point injection of 15NO3 at four different depths under the catch crops, followed by sampling of plant biomass and analysis of 15N enrichment in the plant N pool. The 15NO3 was applied on 25–26 Oct. 2000 in four subplots within each plot, giving four replicate subplots for each depth and species. For each species, the injection depths were chosen to represent soil layers with medium, low, very low, and no roots, based on the preceding minirhizotron measurements of root depth. The depths for ryegrass were 0.4, 0.6, 0.8, and 1 m; for winter rye 0.5, 0.8, 1.1, and 1.4 m; and for fodder radish 1, 1.5, 2, and 2.5 m. The subplots (0.9 by 0.8 m, equivalent to seven plant rows) used for injection at the four depths were randomly placed within each plot at a minimum distance of 2 m. As shown in Fig. 2
, the 15NO3 was injected into each subplot through four holes at a distance in all directions of at least 0.3 m from the deep 15N placement points to the subplot border. The holes were placed at an angle 30° from the vertical using a piston steel rod with a diameter of 20 mm. This was done to minimize damage to the plants within the subplots while drilling.
|
Soil water content at the time of the 15N injection experiment was 19.8, 18.6, 19.0, 19.0, and 16.5% of soil dry weight in 0.5-m increments from the soil surface to 2.5-m depth, which was close to field capacity. Based on these values, the ground water table was judged to be below the 2.5 m-depth during the experiment.
Plant Biomass and Soil Sampling
Six days after 15N injection (31 Oct.–1 Nov. 2000), the aboveground plant biomass in each subplot was harvested for 15N, total N, and C analysis. Ryegrass and winter rye were cut 10 mm below the soil surface, and fodder radish, including the tap-root, was pulled from the ground. The plant samples were kept in plastic bags at 1°C, rinsed in water to remove soil, chopped into coarse pieces, and dried at 80°C within a week after harvest. Additional samples of biomass of the three catch crop species were taken for analysis of background 15N levels in the plant material. Additional samples of biomass were taken immediately adjacent to the border of selected subplots (winter rye 1.4-m and fodder radish 2.5-m depth of injection) to check for 15N uptake in the plants surrounding the subplots. The 15N enrichment in these samples was found to be within the range of the background 15N level.
On the day after harvesting was finished (2 Nov. 2000), soil was sampled for analysis of inorganic N content below the three catch crops. Nine replicate samples were taken randomly in each plot with a soil piston auger with an inner diameter of 14 mm. The samples were divided into depth intervals of 0.25 or 0.5 m from the soil surface to 2.5-m depth, and pooled to one composite sample for each depth and plot, making a total of four replicate samples per depth interval and species. The composite samples were thoroughly mixed, placed at 1°C, and a subsample was frozen at 18°C within 24 h from sampling for later analysis.
Sample and Data Analysis
The samples of plant biomass were milled and a subsample was finely ground (<0.5 mm) and analyzed for 15N, total N, and C content on a Continuous Flow Isotope Ratio Mass Spectrometer consisting of an Automatic Nitrogen and Carbon Analyzer coupled to a 20-20 mass spectrometer (both Europa Scientific Ltd., Crewe, UK).
The 15N plant uptake was calculated from the 15N results, as the experimental design was equivalent to the "negative discard method" described by Powlson and Barraclough (1993). This method is different from the most common approach, which includes homogeneous application of 15N in a well-defined soil layer and detailed knowledge of the resulting 15N enrichment, neither of which is obtained with deep point 15N injection (Gass et al., 1971). Instead, the negative discard method requires that: (i) the crop is sampled from an area that is larger than the labeled area and judged to be sufficiently large to include any labeled N that could have moved outside the original application zone, and (ii) the entire sample is thoroughly mixed before subsampling for 15N analysis (Powlson and Barraclough, 1993). These requirements were met in the present experiment. A third requirement is of unlabeled N application outside plots at rates equal to 15N application rates inside experimental plots. This requirement was not met here, but since 15N application rates were low (1.4 kg N ha–1) compared with residual soil N pools (Table 1), the fertilizer effect of the 15N application could be regarded as negligible. The 15N plant uptake was calculated as excess plant 15N by subtraction of background 15N abundance determined for each species.
|
Statistical significance of differences in root distribution, plant, and soil pools between species or soil layers was tested by analysis of variance (F-test), followed by pairwise comparisons by Tukey's student range test (Proc GLM, SAS Institute Inc., Cary, NC). All results were transformed before analysis by the function y = log(x) to obtain homogeneity of variance. To avoid observations that were negative or equal to zero, which cannot be log-transformed, 0.1 was added to root intensity and frequency and 0.2 to plant 15N uptake. In assessing differences between results, tests with P < 0.05 were considered statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Root Depth
All three species differed in rate of root depth development. Winter rye and fodder radish were found to extend their average root depth to >1-m depth (Fig. 3a)
. Winter rye had reached 1 m by 16 October after an accumulative temperature of 977 d °C from sowing, whereas fodder radish had already reached 1 m by 20 September, only 626 d °C after sowing (Fig. 3b). Ryegrass had obtained an average root depth of 0.64 m just before the 15N injection experiment on 23 October, which increased to 0.76 m after termination of the 15N injection experiment on 3 November. Root depth of winter rye was 1.06 and 1.15 m on the two dates, while the fodder radish had obtained a root depth of 2.24 m by 23 October and 2.27 m by 3 November. However, the results for fodder radish were influenced by the fact that roots in at least three of the eight replicate minirhizotrons had reached the maximum measuring depth (2.42 m) of the minirhizotrons by 23 October. The lack of results below 2.42-m depth caused the average estimate of root depth to level off for the fodder radish by the last two measuring dates. These data points were therefore deleted from the regression analysis in Fig. 3b. The r2 values of the regression analyses of root depth versus accumulated daily temperature from sowing were close to 1, and the depth penetration rates were 0.8, 1.3, and 3.5 mm d–1 °C–1 for ryegrass, winter rye, and fodder radish, respectively.
|
|
Plant Nitrogen-15 Uptake
For all three species, 15N uptake was highest at the shallowest 15N injection depths (P < 0.004), decreasing to zero or close to zero at the deepest injection depth (Fig. 4c). In general, 0 to 21% of added 15N was taken up during the 6 d of the experiment, and the 15N uptake profiles differed between the three species. The 15N uptake in ryegrass from the shallowest depth of 0.4 m was 21.4 mg N subplot–1, decreasing sharply to 2.9, 0.3, and 0.0 mg N subplot–1 at the 0.6-, 0.8-, and 1-m depths, respectively. The 15N uptake of winter rye from the shallowest injection depth of 0.5 m was only 1.3 mg N subplot–1, decreasing gradually to 0.5, 0.1, and 0.0 mg N subplot–1 at the 0.8-, 1.1-, and 1.4-m depths, respectively. For fodder radish, a 15N uptake of 6.2 mg N subplot–1 was observed at 1-m depth, decreasing to 4.4, 1.1, and 0.8 mg N subplot–1 at the 1.5-, 2-, and 2.5-m depths, respectively.
Residual Soil Nitrate
The distribution of soil NO3 in the 0- to 2.5-m soil profile differed among the catch crops at the time of the 15N injection experiment (Fig. 4d). Winter rye had left a higher concentration of 2.1 mg NO3–N kg–1 soil in the 0- to 0.25-m layer compared with ryegrass with 1.1 mg NO3–N kg–1 soil (P = 0.04), whereas all three species had equal levels of 1.3 to 1.4 mg NO3–N kg–1 soil in the 0.25- to 0.5-m layers (0- to 0.5-m layer for fodder radish) (Fig. 4d). The NO3 concentration under ryegrass increased to a maximum value of 4.5 mg N kg–1 in the 0.5- to 0.75-m layer, that is, significantly higher than under winter rye (P = 0.01). In the 1.25- to 1.5-m layer, the concentration decreased to 2.1 mg N kg–1, that is, still significantly higher than under fodder radish but not compared with winter rye (P < 0.05). Winter rye showed a similar peak in NO3 concentration of 2.7 mg NO3–N kg–1 around the 0.75- to 1-m layer followed by a decrease to 0.7- to 1.0 mg NO3–N kg–1 in the 1.5- to 2.5-m layer. The NO3 level under fodder radish was found to be at a constant low level of 0.2 to 0.3 mg N kg–1 in the 0.5- to 2.5-m layers, which was significantly lower than for winter rye and ryegrass at all comparable layers. Ammonium levels were found to be at the same level under the three catch crops with values of 1.3 to 1.8 mg NH4–N kg–1 in the 0- to 0.5-m layer and 0.1 to 0.6 mg NH4–N kg–1 in the layers from 0.5 to 2.5 m (results not shown).
Regression Analysis of Nitrogen-15 Uptake and Root Distribution
Regression analyses were performed between the amount of 15N plant uptake from the four 15N injection depths for each species and root intensity or frequency in the 0.2-m (±0.1 m) soil layer surrounding the point of 15N injection (Fig. 5a,b)
. Good correlation (r2 = 0.94–1.00) was found between both 15N plant uptake and root intensity as well as between 15N plant uptake and root frequency (r2 = 0.87–1.00).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
It has been suggested that only part of a root system may be active in N uptake at any one time (Robinson et al., 1991), and that the activity level of each root may change over time (Henriksen et al., 1992) along with root growth and distribution. These changes can be triggered by changes in local N supply (Van Vuuren et al., 1996; Hodge et al., 1999). At the scale of whole root systems in the field it is nevertheless reasonable to expect a fairly linear relationship between root density in different soil layers and plant N uptake from those layers, under conditions such as in the present experiment. These conditions include a sudden excess N supply in the given soil volume as well as measurements of 15N uptake over a few days. The linear relationship is expected because the quantity of roots is thought to be a major factor governing plant N uptake under these conditions (Robinson, 1986). Such a linear relationship was found in the regression analysis of plant 15N uptake and root intensity shown in Fig. 5a. Several studies have shown that root intensities observed with the minirhizotron method often do not correlate well with root length densities determined by soil sampling and root extraction techniques, and the validity of the minirhizotron technique has been questioned on this basis (e.g., Parker et al., 1991; Heeraman and Juma, 1993). The good correlation between plant N uptake and root intensity obtained with minirhizotrons in this study, however, shows that the minirhizotron methodology gave information on root distribution that was meaningful in terms of the ability of the root systems to take up NO3 from different depths.
It should be kept in mind that the root intensities at the 15N injection points covaried with depth for all species. The relationship could therefore be related to differences in depth instead of root intensity. With larger injection depth, the distance that 15N must be transported through the root system to aboveground plant parts increases. The effect on 15N uptake could therefore be the result of slow translocation of 15N relative to the distance between shallow and deep injection depths, as only 6 d were allowed for 15N uptake and translocation before sampling of aboveground plant parts. Nitrogen can be translocated from root to shoot within a matter of hours or a few days (Rao et al., 1993; Rossato et al., 2001), but the rate of translocation depends among other factors on species, transpiration, and C-supply, most of which were unknown in this experiment. Comparison of 15N uptake profiles for the three species, however, does not indicate an influence of injection depths. In this case a plot of 15N uptake against injection level (medium, low, and very low root abundance) for each species would show a steeper gradient the greater the distance between injection depths. These were 0.2, 0.3, and 0.5 m for ryegrass, winter rye, and fodder radish, respectively. Instead, the steepest gradient was found in ryegrass and the flattest in winter rye (results not shown).
Root Intensity and Frequency Measures
A high linear correlation was found not only for plant 15N uptake and root intensity but also for root frequency (Fig. 5b). Use of the root frequency measurement (Thorup-Kristensen, 2001) is based on the theory that just a few roots is sufficient for depletion of NO3 in a given soil layer because of the high mobility of the ion in soil solution (Robinson, 1986). Over the short term and under excess NO3 availability, as in the present study, it can be expected that root intensity will correlate better than root frequency with N uptake, because the number of roots will be the limiting factor for NO3 uptake. However, this assumes that the N demand of the plant is high (Robinson, 1986), which may not have been the case here. A relatively low plant N demand was indicated by the high biomass N concentrations for winter rye and fodder radish of 4.25 and 3.98%, respectively (Table 1), compared with 2.94 and 2.86% in another study with relatively high N supply (Thorup-Kristensen, 2001). Furthermore, high amounts of residual N were left within the root zone of winter rye and ryegrass, which indicated adequate N supply during the growing season (Fig. 4d, Table 1). Another cause of the equally good correlation between 15N uptake and root intensity or frequency could be the strong covariation between root intensity and root frequency. Thus, under the conditions of the present study, root frequency was found to be a useful alternative to root intensity as a measure of root distribution in the study of plant N uptake. This was also found by Thorup-Kristensen (2001) when studying residual soil NO3 under catch crops. Root frequency has the advantage that it can be measured much more quickly than root intensity, which demands counting of all roots along the minirhizotron grid.
Root Depth and Distribution
The use of long minirhizotrons reaching a depth of 2.42 m revealed that the three catch crops had very different root development, that is, shallow (ryegrass), intermediate (winter rye), and deep (fodder radish) rooted species (Fig. 3a). The results for the three species concur with the conclusion of Thorup-Kristensen (2001) that crucifers are characterized by faster root depth development compared with monocots, and are in accordance with studies showing a larger maximum root depth in many dicots than in monocots (Kutschera and Lichtenegger, 1982, 1992; Sun et al., 1997). The root depth penetration rates of 0.8 and 1.3 mm d–1 °C–1 found for ryegrass and winter rye (Fig. 3b) were comparable with those of 1.1 and 1.2 mm d–1 °C–1 for the two species on the same soil during two subsequent years when similar root methodology was used (Thorup-Kristensen, 2001). The root depth penetration rate of 3.5 mm d–1 °C–1 for fodder radish (Fig. 3b) was higher than penetration rates of 2.0 and 2.3 previously reported for fodder radish and winter rape, another cruciferous species. In this previous study, however, minirhizotrons only reached 1.2 m (Thorup-Kristensen, 2001). The discrepancy between the results for fodder radish could be due to the very fast root development of this species, since the use of shorter minirhizotrons allowed only a short period for measurement before the roots reached the bottom of the minirhizotrons.
The differences between the two monocots and fodder radish were also evident in the distribution of roots in the soil profile, with monocots having the highest root intensity in the surface soil layer, declining gradually to the bottom of the root zone (Fig. 4a). Contrary to the monocots, fodder radish showed consistently higher root intensity from the soil surface to 1.5-m depth. Similar differences between monocot and dicot species have been found by Materechera et al. (1993).
Catch Crop Efficiency for Nitrate Uptake
The fodder radish had practically depleted the soil of NO3 down to 2.5-m depth at the time of the 15N experiment, leaving only 18 kg NO3–N ha–1 compared with 87 kg NO3–N ha–1 under ryegrass (Table 1). This showed that fodder radish is a very efficient catch crop for soil N depletion. Much of the difference was found below a depth of 1 m, which is the maximum measuring depth in most root and residual soil N studies. Based on results from the 0- to 1-m layer, the efficiency of fodder radish in depleting NO3 was 36 kg N ha–1 higher than that of ryegrass with an additional difference of 33 kg N ha–1 in the 1- to 2.5-m layer. This emphasizes the importance of choosing the right measuring depth for studies of N leaching and cycling depending on the root depth of the plant species under study.
Ryegrass left much NO3 at the border of its root zone in the 0.5- to 1-m soil layer (Fig. 4d), but less in the soil layers below 1 m. The lower amount of residual NO3 left below the root zone indicated that NO3 had been low there from the beginning of the growing season. Winter rye was more efficient for NO3 retention than ryegrass, even below the 1.5-m depth, which is below the root depth of winter rye (Fig. 4d). This could be due to leaching of soil solution with lower NO3 concentration from the root zone.
In general, root depth of the catch crops was found to be a good indicator of how effective the catch crops were in depleting NO3, whereas aboveground biomass N was not. No correlation was found between this pool and residual soil NO3 under the three catch crops (Table 1). The discrepancy between amounts of NO3 taken from the soil and N found in aboveground biomass can probably be explained by differences in N pools such as root systems (Jackson et al., 1993) and dead soil organic matter originating from leaves shedded during growth (Rossato et al., 2001). Based on the relationship between root depth and NO3 depletion there is reason to believe that other species with deep root depth penetration may be efficient as catch crops for N uptake from deep soil layers, for example, winter rape (Thorup-Kristensen, 2001). When designing crop rotations these deep-rooted species should be used where high amounts of N have been leached to deeper soil layers. Many fields have tile drains at 1 m-depth, which remove large amounts of soil solution after the ground water table rises in winter. The major proportion of catch crop N uptake takes place during autumn, however, when the ground water table is much deeper. Nitrate taken up by deep-rooted catch crops below the root zone of the following crop therefore represents a net input of N into the system that would otherwise have been lost through leaching. The use of deep-rooted catch crops could therefore substantially improve the N use efficiency of crop rotations.
Root Lengths and Nitrogen Inflows
Root intensity estimated by counting the number of roots intersecting with the grid on the minirhizotron surface can be used for estimation of root-length density (Heeraman and Juma, 1993). This was done following the modified Newman-line–intersect method for estimating root length (Tennant, 1975), and by assuming a depth of view of 2 mm into the soil surrounding the minirhizotron to calculate the volumetric root-length density (Taylor et al., 1970; Heeraman and Juma, 1993). Total root lengths were estimated to be 17, 31, and 92 km m–2 for winter rye, ryegrass, and fodder radish, respectively. For comparison, values of 5, 12, 19 to 29, and 11 to 25 km m–2 for sugarbeet, winter barley, winter wheat, and winter rape, respectively, have been found in field studies by other methods (profile wall and soil coring) (Barraclough and Leigh, 1984; Barraclough, 1989; Strebel and Duynisveld, 1989). The root-length density of 92 km m–2 for fodder radish seems very high. The maximum root-length densities of 5 to 6 cm cm–3 calculated for 0.25-m soil layers under fodder radish (results not shown), however, are in range with values found in other field studies for fodder radish (6 cm cm–3) (Vos et al., 1998) and winter rape (4 and 10 cm cm–3) (Barraclough, 1989; Vos et al., 1998). The very high estimate of total root length may therefore be the result of high root-length density extending to deep depths.
The 15N uptake profiles for the three catch crops can be used to make a rough calculation of plant N uptake rates from a given soil volume (Table 2). It was assumed that 15N at each injection point was distributed in 1 L of soil. The 15N enrichment in this soil volume was then adjusted for the dilution effect of the soil NO3 pool (atom% assumed to be natural abundance of 0.366% 15N). Root/shoot N ratios found by Jackson et al. (1993) of 0.27, 0.17, and 0.38 for ryegrass, winter rye, and fodder radish, respectively, were used to include root biomass 15N in total plant N uptake rates for the 6-d period. The N uptake rates and root length densities were used for calculating N inflow rates at the depths at which 15N was injected (Table 2). Fodder radish showed inflow rates in the 1-, 1.5-, and 2-m depths of 3 to 5 pmol m–1 s–1, whereas the rate at 2.5-m depth was one order of magnitude higher. These rates are all in range with those of 4 to 22 pmol m–1 s–1 found for winter rape in the 0- to 0.2-m soil layer during spring and summer (Barraclough, 1989). An explanation for the higher inflow from 2.5-m depth under fodder radish could be that the roots had reached this soil layer within the week immediately before 15N injection. All roots were therefore young and had recently been active in the uptake of small amounts of available NO3 (Fig. 4d). In contrast, the rest of the root system consisted of roots of different age occupying soil layers that had probably been depleted of NO3. Age and NO3 conditions can strongly influence a root's ability to take up NO3 (Lainé et al., 1993).
|
None of the three catch crops showed any tendency toward reduced N inflow rates with depth, thus supporting the notion that the 6-d experimental period from 15N injection to plant sampling was sufficient to ensure translocation of 15N to plant shoots as discussed above.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Differences in root depth were significant in the ability of catch crops to absorb N. It is therefore crucial that the entire root zone of species is measured if the effects on NO3 leaching of different catch crop species are to be compared. Nitrate depletion in deeper soil layers (below 1 m) could be substantial. To increase N use efficiency of crop rotations, the NO3 taken up by deep-rooted catch crops below the root zone of the following crop is especially important. It represents a net input of N to the system, which would otherwise have been lost through leaching. The knowledge on root depth and N uptake from deeper soil layers can be used in the design of crop rotations to improve N use efficiency.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 12, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. S. Boyd, E. B. Brennan, R. F. Smith, and R. Yokota Effect of Seeding Rate and Planting Arrangement on Rye Cover Crop and Weed Growth Agron. J., January 8, 2009; 101(1): 47 - 51. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. G. Poulsen and P. Moldrup Factors affecting water balance and percolate production for a landfill in operation Waste Management Research, February 1, 2005; 23(1): 72 - 78. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
) are situated right under the subplot. The numbers indicate distances in meters.
