SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gathumbi, S. M.
Right arrow Articles by Giller, K. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Gathumbi, S. M.
Right arrow Articles by Giller, K. E.
Agricola
Right arrow Articles by Gathumbi, S. M.
Right arrow Articles by Giller, K. E.
Related Collections
Right arrow Crop Rotation Systems
Right arrow Nutrient Cycling
Right arrow Isotopes
Soil Science Society of America Journal 67:573-582 (2003)
© 2003 Soil Science Society of America

DIVISION S-4—SOIL FERTILITY & PLANT NUTRITION

Subsoil Nitrogen Capture in Mixed Legume Stands as Assessed by Deep Nitrogen-15 Placement

Stanley M. Gathumbia, Georg Cadisch*,b, Roland J. Bureshc and Ken E. Gillerd

a MacArthur Agro-Ecology Research Center, 300 Buck Island Ranch Road, Lake Placid, FL 33852
b Department of Agricultural Sciences, Imperial College at Wye, University of London, Wye, Kent, TN25 5AH, UK
c International Rice Research Institute, DAPO Box 7777, Metro Manila, Philippines
d Plant Production Systems, Department of Plant Sciences, Wageningen University, P.O. Box 430, 6700 AK Wageningen, The Netherlands

* Corresponding author (g.cadisch{at}ic.ac.uk)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The rotation of crops with planted N2–fixing legumes (improved fallows) is a promising agroforestry innovation for replenishing soil fertility in the tropics. We postulated that woody and herbaceous legumes with different rooting and growth patterns could be mixed in improved fallows to maximize utilization of belowground growth resources. We used a method of injecting a solution of 15N-labeled (NH4)2SO4 into soil at 0.15- and 1.0-m depths to measure soil mineral N acquisition by sesbania [Sesbania sesban (L) Merr], crotalaria [Crotalaria grahamiana Wight and Arn.], and the understory herbaceous legume siratro [Macroptilium atropurpureum (DC.) Urb.] grown in mixed stands on a Kandiudalfic Eutrudox soil in western Kenya. Crotalaria had the highest root length in the topsoil. Sesbania on the other hand had nearly half its total root length below 0.3 m at 0.3- to 1.5-m depth; sesbania took up more added 15N than crotalaria and siratro from the 1.0-m depth. Mixed sesbania and crotalaria stands, as compared with growing species in monocultures, increased root length at the 0.3- to 1.2-m depth. Sesbania mixed with siratro was more effective than sesbania mixed with crotalaria in uptake of 15N at 1.0-m depth but not at 0.15-m depth. At 2 mo after injection, the 15N was concentrated immediately below the injection point with little lateral movement. This confirmed the utility of the methodology in determining temporal N uptake for species in mixed stands. Our results suggest that opportunities exist for enhanced subsoil N retrieval through the mixing of leguminous species, which can influence root distribution and increase rooting in the subsoil.

Abbreviations: DM, dry matter • Lrv, root-length density • PVC, polyvinyl chloride • SE, standard error of the means • SRL, specific root length


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
A PRIMARY OBJECTIVE for establishing fast-growing legumes, as a replacement for traditional natural fallows in tropical crop-fallow rotations is to restore fertility of nutrient-depleted soils within a short duration (Prinz, 1986). Leguminous species achieve this, not only by sourcing N from the atmosphere through N2–fixation, but also through their ability to explore subsoil nutrient pools and to capture available nutrients through their extensive root systems (van Noordwijk et al., 1996; Buresh and Tian, 1997; Gathumbi et al., 2002b). Appreciable amounts of N in the form of nitrates can accumulate in agricultural soils, especially when N inputs from fertilizer application and mineralization of soil organic matter in the topsoil layers exceed crop N demand and uptake (Guillard et al., 1995). Since NO3–N is highly mobile in soils, unutilized NO3–N is subject to rapid leaching (Suprayogo et al., 2002). Additionally, appreciable amounts of N can apparently be mineralized in lower soil layers in deep and clay-rich soils. It has been hypothesized that active tree and shrub roots can intercept leaching nutrients and also capture leached nutrients or nutrients from deep horizons resulting from weathering (van Noordwijk et al., 1996; Rowe et al., 1999). These two concepts have been referred to in the recent literature as "the safety net hypothesis" and the "nutrient pump hypothesis," respectively. Recent studies conducted in an oxisol in western Kenya have shown that sesbania fallows can reduce the subsoil nitrate bulge (accumulated nitrates) through plant uptake (Mekonnen et al., 1997) and also through a reduction of leaching (Hartemink et al., 1996).

The safety net and nutrient pump hypotheses can effectively be tested by use of 15N tracer techniques (IAEA, 1975). Rowe et al. (1999) successfully used 15N-labeling methods to test the safety net role and root activity of two species with different rooting patterns in a hedgerow intercropping agroforestry system in Indonesia. Nitrogen-15 uptake by tree roots varied with time after application and between species, but the 15N signatures were well above the background natural 15N abundance. Atkinson et al. (1978) reported similar variation of 15N uptake while using deep placement techniques in studies on root distribution and N uptake by fruit trees. Deep placement of 15N was also used in studies on subsoil N uptake by pearl millet [Pennisetum glaucum (L.) R. Br.] on a Kandiudult soil of the southeastern USA (Menezes et al., 1997). Similarly, Huang et al. (1996) used this technique during an investigation of the relative recoveries of deep-point injected soil 15N by switchgrass (Panicum virgatum L.), alfalfa (Medicago sativa L.), and maize (Zea mays L.).

The results of the above studies indicated applicability of these 15N-tracer methods in monitoring rooting systems and utilization of soil N by crops and trees in agroforestry systems. However, the main limitation of this technique is the high variability in plant root distribution relative to the 15N application points. This results in varying signatures of 15N uptake, which is a major source of experimental variability.

All earlier 15N injection experiments had examined single species stands. In this study we mixed species and tested the hypothesis that when legume species are grown in mixtures, soil N uptake is maximized at different soil depths through their different rooting patterns and distribution within the soil horizons. We also assessed lateral and vertical movement of the 15N-tracer in the field to evaluate the validity of the 15N placement methodology. The objectives were to assess rooting pattern and distribution of fallow legume species in sole and mixed fallow systems and to assess uptake and recovery of 15N-enriched fertilizer in mixed species stands.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
This study was conducted in a farmer's field in western Kenya (0°06'N lat., 34°34'E long., 1330 m above sea level). Rainfall in the study area is distributed in two crop-growing seasons per year with an annual mean of 1800 mm. The long rainy cropping season extends from March to August and the short rainy cropping season extends from September to January. Daily rainfall pattern recorded during the course of the 15N injection experiment is shown in Fig. 1 . Soils are highly weathered and are generally classified as very fine, kaolinitic, isohyperthermic Kandiudalfic Eutrudox. The initial soil physical and chemical characteristics at a depth of 0 to 0.15 m were: pH = 5.6 (in a 1:2.5 soil/water suspension); organic C = 14.0 g kg-1, by wet oxidation with heated acidified dichromate followed by colorimetric determination of Cr3+ (Anderson and Ingram, 1993); extractable P = 1.3 mg kg-1 and exchangeable K = 0.3 cmolckg-1 (by extraction with 0.5 M NaHCO3 + 0.01 M ethylenediaminetetraacetic acid, pH 8.5); and exchangeable Ca = 5.4 cmolckg-1, Mg = 1.70 cmolckg-1, and exchangeable acidity = 0.5 cmolckg-1 (by 1 M KCl extraction); sand = 27%; clay = 52%; silt = 21%; and bulk density = 1.3 g cm-3.



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 1. Daily rainfall received from January through March 1998 during the period the 15N injection experiment was conducted in western Kenya.

 
Experimental Design
The experiment was superimposed within a larger legume screening experiment, consisting of single and mixed species stands, established in October 1997 (short rainy season) (Gathumbi et al., 2002b). Two fallow mixture types were selected for the 15N injection study based on their contrasting growth and above and potential belowground resource acquisition patterns. They were sesbania + crotalaria (50% sesbania and 50% crotalaria planted in alternate rows) and sesbania (100%) + siratro (as an understory plant) (Gathumbi et al., 2002b). Plant spacing was 0.75 by 0.75 m for sesbania + crotalaria, 0.75 by 0.75 m for sesbania in sesbania + siratro, and 0.375 by 0.75 m for undersown siratro (Fig. 2) . Two subplots for 15N injection each measuring 0.75 by 0.75 m were systematically laid out at diagonal corners of the main plot, which measured 6 by 5.25 m. To reduce risk of cross-contamination two replicates of the same application depth were located on the same main fallow plot. The experimental layout hence consisted of 16 15N placement subplots, eight of each 0.15- and 1.0-m application depths. Each 15N placement depth was tested in two species stands and replicated four times.



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 2. Field layout of 15N-application subplots in two improved fallow systems in western Kenya.

 
Nitrogen-15 Depth Placement Procedures
Four months after fallow establishment (15NH4)2SO4, with an enrichment of 30.0 atom% 15N, was injected in equal aliquots through five polyvinyl chloride (PVC) tubes with a 2-cm i.d. previously installed in subplots at 0.15- or 1-m depths (Fig. 2). For each tube the 15N solution was applied through a funnel with an attached tube reaching to the bottom of the PVC tube at a rate of 25 mL (containing 0.4 g of N (15NH4)2SO4) and then flushed with an extra 25 mL of deionized water.

Plant Sampling and Nitrogen-15 Recovery Calculations
To assess the time course of 15N uptake, the first four fully developed leaflets from the shoot tips from the center plants of the application subplot were sampled at 7, 14, 21, 28, and 35 d after 15N application. Final harvesting of the plants in the 15N application sites was performed on 24 March 1998, about 5 wk after the 15N application. Recovery of the applied 15N was measured only for the aboveground plant components. Center plants of woody species (n = 4) were harvested individually while border plants (n = 12) were bulked to give composite samples. For siratro center plants (n = 3) and borders (n = 18), all vines for individual plants were tracked down and harvested. All plant samples were separated into plant components (main stem, branches, and foliage) and total fresh weight determined. The stems and branches were then chopped into small pieces. All plant materials were thoroughly mixed before taking representative subsamples of each plant component using the quartering procedure. Subsamples weighing approximately 500 g were taken for oven-dry weight determination.

Additional young fully developed leaf samples were taken from plants located at 0.75, 1.06, 1.68, 2.37, and 3.09 m from the injection point to generate data on uptake of labeled N by lateral roots and the minimum spatial distance required to separate 15N placement sites and avoid interference between plots (Fig. 2). The minimum distance between the sampled plant in the outer row and the nearest corner of the 15N site was calculated using the formula:

[1]
where d is the minimum distance between sampled plant and the nearest corner of 15N injection site, x is the distance between the corner of 15N injection site and the nearest plant in the outer row (0.75 m), and y is the distance between the nearest plant and the sampled plant.

All plant samples were oven dried (50°C, 72 h) and ground in order of expected enrichment to avoid cross contamination using a micro hammer mill. The percentage of N and the 15N enrichment of the samples were then determined using a 20-20 stable isotope mass spectrometer (PDZ formerly Europa Scientific, Crewe, UK) coupled to an automated CN analyzer (Crewe, UK).

The 15N enrichment of all materials (fertilizer and plant samples) was expressed as atom% 15N excess, which was determined from the natural 15N abundance of the soil (0.3668) or the respective plant N pool (Rowe and Cadisch, 2002) measured before application of 15N enriched fertilizer. The amount of 15N applied was calculated as 15N excess over the background abundance of 15N in soil, since it is the soil N pool to which enriched fertilizer was applied. The proportion of the total applied labeled N recovered by target plants was calculated using the formula:

[2]
where i = plant components (n = 3) of center (n = 4) or border (n = 1 composite) plants.

Soil Sampling at Nitrogen-15 Application Sites
Soil samples were taken from under the 15N injection points on 9 Apr. 1998 (2 mo after application), to generate information on the possible vertical and lateral movement of the labeled fertilizer. Sampling was performed so that the auger hole was directly over the injection point. For the 0.15-m placement depth, vertical sampling was done from 0.15 to 1.15 m in 0.1-m intervals, and lateral sampling was done at 0.1-m intervals up to a 0.5-m distance. Lateral sampling was done at perpendicular distances away from the 15N application site with reference to one outer 15N injection tube at the 0.15-m application depth. For the 1.0-m placement depth, the same sampling intervals were used but starting at 1.0 m deep. The soil samples were air dried and finely ground and analyzed for percentage of N and 15N.

Root Sampling and Analysis
Root studies were performed between 27 Feb. and 1 Mar. 1998 on replicated soil profiles for four selected treatments within the same main experiment (Gathumbi et al., 2002a), which included sesbania + crotalaria and sesbania + siratro among the mixed species fallows. In addition, root samples were also taken from sesbania and crotalaria sole species fallow plots to offer additional data on rooting patterns of the same species when grown in monoculture arrangement. Root sampling was conducted on a profile wall exposed from a pit dug in one of the corners of each plot. The pits were dug 0.10 m from the base of the target plants and extended to 1.8-m depth. The orientation of the profile wall was such that it was diagonal to the line of trees and shrubs to be sampled. This was to ensure that the sampled area was representative of the unit area and soil covered by each plant following the suggestions of Rao and Coe (1991). The profile wall was smoothed using a machete.

Soil-root sampling was done using a monolith metal box sampler with measurements of 15 by 15 by 10 cm, giving a volume of 2250 cm3. The root-sampling depths were 0 to 0.15, 0.15 to 0.30, 0.30 to 0.45, 0.45 to 0.60, 0.60 to 0.90, 0.90 to 1.20, and 1.20 to 1.50 m. Soil-root samples were soaked in water for at least 12 h following the procedures of Bohm (1979). The stirred soil-root suspension was then passed through a 0.5-mm sieve. Roots and organic debris retained on the sieve were stored at 5°C in plastic bags containing 17% (v/v) acetic acid solution. The roots were separated from organic debris. In mixed fallow plots, roots of the respective species could not be recognized with certainty and hence the calculations were based on the composite root sample for each depth. The separated roots were stained with methyl violet solution (0.1% in 10% [v/v] ethanol), spread on a glass tray, and scanned with Delta-T Scan system (Delta-T Devices, Ltd., Cambridge, UK). Root images were obtained with a Hewlett Packard scanner (Hewlett Packard Analytical Group, Palo Alto, CA) and Aldus Photostyler image analysis software (Aldus Corp., Seattle, WA) at a resolution of 95 dots per centimeter. Root-length density (Lrv) was calculated by the formula of Newman (1966) when the tray contained nonoverlapping roots and by the formula of Harris and Campbell (1989) when the tray contained overlapping roots. Immediately after scanning, the root samples were oven dried (70°C, 48 h) and weighed.

Statistical Data Analysis
Root data were tested for normality by plotting residuals against fitted values. They were not normally distributed and hence a square root transformation was performed for the percentage data; all other data sets were log transformed (logarithm to base 10) before analysis (Gomez and Gomez, 1984). Analysis of variance (ANOVA) was then performed using the general linear model procedure of the SAS program (SAS Institute, 1990). All data, however, are reported on an untransformed scale for comparison of treatment means.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Nitrogen-15 Uptake Patterns
The 15N enrichment of young leaves from the 0.15-m 15N placement depth was highest for sesbania followed by crotalaria and siratro (Fig. 3a) . Even after only 1 wk, significant 15N enrichment was detected showing highly active plant roots at that soil depth. The 15N enrichments of young leaves increased progressively with time, but increments decreased in sesbania after 21 d. At all sampling times the 15N leaf enrichment was highest for sesbania and lowest for siratro.



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 3. Nitrogen-15 concentration in young leaves after application of 15N at two soil depths in western Kenya.

 
At 1-m placement depth, the 15N leaf enrichment remained low for the first 14 d after 15N placement for all species tested, but thereafter started to increase significantly, except for siratro for which 15N enrichment declined after 28 d (Fig. 3b). The leaf 15N enrichment of sesbania was less with 1-m placement than with 0.15-m placement, while enrichments were unaffected by placement depth with crotalaria and siratro at 35 d. The highest 15N enrichment was again obtained from sesbania at 35 d after 15N placement at 1-m.

Nitrogen-15 Uptake at Lateral Distances From the Application Site
Young leaf 15N enrichment for sesbania and crotalaria plants sampled at varying lateral distances from 15N application sites showed a decreasing enrichment with increasing distance (Fig. 4) . At any one distance, the 15N concentration was higher for sesbania than crotalaria. The 15N enrichment for the leaf materials sampled at 0.75 m from the nearest sampling point was 0.01 and 0.07 atom% 15N excess for crotalaria and sesbania, respectively. No significant 15N enrichment was detected in plant samples taken >1.7 m (crotalaria) and >2.4 m (sesbania) away from the injection point when compared with the background 15N enrichment of plant materials grown in unenriched soil.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 4. Young leaf 15N concentration for sesbania and crotalaria plants sampled at various distances from the 15N injection point at 9 wk after application in the field.

 
Lateral and Vertical Residual Soil Nitrogen-15
Five weeks after 15N-enriched fertilizer application, most of the residual 15N was observed to be within 0.10-m lateral distance relative to the injection point, which averaged 0.5 atom% 15N excess. Soil samples taken 0.20 to 0.50 m away from the 15N injection point had 15N enrichment similar to the background soil.

Vertical atom% 15N excess distribution down the soil profile decreased with depth indicating some gradual downward movement (leaching) of the applied 15N relative to the application point (Fig. 5) . Nitrogen-15 enrichment detected in soil samples taken at 0.40 m below the 0.15-m application depth and 0.80 m below the 1.0-m application depth were similar to the background enrichment of the soil (0.3668 atom% 15N excess).



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 5. Residual soil atom% 15N excess at vertical distances below the 0.15- and 1.00-m depths for 15N injection at 9 wk after application.

 
Plant Root Distribution
Total root length decreased with profile depth (Table 1). The highest total root length to 1.5-m depth was observed in crotalaria + sesbania (6.7 km m-2) and the lowest was with sesbania + siratro (2.2 km m-2). Root data from several depths were pooled because of high variability of the data. The 0- to 0.3-m soil layer was observed to contain the highest proportion of the total root length, the highest being 76% for crotalaria. At 0.3- to 0.6-m depth a higher proportion of roots was observed for mixtures of sesbania + crotalaria (26%) compared with sole sesbania (16%) and sole crotalaria (10%) treatments. Sole crotalaria treatments had a proportionally lower root length at 0.6- to 1.5-m soil depth, accounting for only 14% of the total root length compared with 25 to 33% of the other fallow treatments. With respect to total root length, sesbania + siratro mixtures had lower (P < 0.05) root length at 0- to 0.3-m depth compared with crotalaria. No statistical differences in root length among treatments were observed below 0.6 m.


View this table:
[in this window]
[in a new window]
  		Table 1. Distribution of root length, specific root length, and root biomass with soil depth for different species.

 
As with root length, the Lrv (cm cm-3) decreased with depth. The highest Lrv in the 0- to 0.15-m soil layers was with crotalaria and the lowest was with sesbania + siratro (Fig. 6) . The Lrv of sesbania + crotalaria was observed to be larger in the 0.30- to 0.60-m soil layer as compared with other treatments at the same depth.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 6. Effect of soil depth on root length density (Lrv) of pure and mixed species.

 
Statistical differences in specific root length (SRL, m g-1) between treatments were observed only in the top 0.3 m where SRL of sesbania was significantly less (P < 0.05) than that of crotalaria. With the exception of sesbania + crotalaria at the 0.3- to 0.6-m depth, the SRL of crotalaria was consistently higher in all soil depths compared with other treatments, suggesting that this species had a higher proportion of fine roots than coarse roots at all layers.

The sesbania + siratro mixed fallow was among the treatments observed to have the least root biomass at all soil depths. Total root biomass to 1.5-m soil depth was significantly higher (P < 0.05) for sesbania + crotalaria mixtures (1343 kg ha-1) compared with other treatments.

Nitrogen-15 Recovery by Sesbania, Crotalaria, and Siratro
Total biomass yield determined from all plants harvested from the 15N-injection subplots was not different (P > 0.05) between the fallow types. Total biomass for the harvested center and border plants in sesbania + crotalaria mixture was 252 (±17 SE) g m-2 (56% sesbania, 44% crotalaria) and for sesbania + siratro was 372 (±67 SE) g m-2 (91% sesbania, 9% siratro). In both mixed stands sesbania trees contributed a higher proportion of the biomass than the companion species crotalaria or the understory plant siratro.

In sesbania and crotalaria mixtures, 15N recovered by center trees was 64 mg 15N site-1 for sesbania and 78 mg 15N site-1 for crotalaria when labeled N was injected at the 0.15-m depth (Table 2). Crotalaria border plants recovered less (7 mg) of the applied N than sesbania border trees (13 mg). Total mean 15N recovery from the 0.15-m application depth was 162 mg, representing 27% of the applied N. Sesbania and crotalaria contributed 13 and 14%, respectively (Table 3).


View this table:
[in this window]
[in a new window]
  		Table 2. Recovery of applied 15N from 0.15- and 1.0-m placement depths by sesbania, crotalaria, and siratro grown in mixtures as determined from "center" and "border" plants relative to the 15N injection point in the field in western Kenya.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Recovery of the labeled N from 0.15- and 1.0-m placement depths by sesbania, crotalaria, and siratro grown in mixtures in the field in western Kenya.

 
Sesbania and siratro in mixture recovered in total 22% of the applied N at 0.15 m with sesbania and siratro contributing 16 and 6% respectively. While sesbania center plants contributed most to the 15N recovery, the stoloniferous siratro border plants were equally active in 15N recovery as were sesbania border plants (Table 2). At 1.0-m application depth, sesbania grown in crotalaria + sesbania mixtures had a higher recovery of 15N (32 mg 15N) than crotalaria (17 mg 15N) for the center plants. Also sesbania border plants recovered more 15N than crotalaria border plants. Total mean recovery was 13% of the applied N. The sesbania + siratro mixture recovered a similar amount of 15N from the 1.0 m as with the 0.15-m application depth. However, most of the 15N was recovered by sesbania center plants (57 mg 15N), compared with only 9 mg 15N recovered by siratro. Siratro border plants recovered a higher proportion of the applied N compared with the center plants. Total mean 15N recovery from 1 m was 18% (107 mg 15N site-1) of the applied N (Table 3).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Use of Nitrogen-15 Placement to Differentiate Uptake in Mixed Species Stands
The recovery of applied N by plants depends on location and spatial distribution of active roots relative to the 15N application point. In this study there was a large variation in vertical root distribution for all the species within the profile and also across the four replications. As such the standard errors for the means in 15N uptake were large as would be expected (Table 2). Large variations in root distribution were also reported for gliricidia [Gliricidium sepium (Jacq.) Kunth ex Walp.], peltophorum [Peltophorum dasyrrhachis (Miq.) Kurz.] (Rowe et al., 1999), and sesbania (Jama et al., 1998; Mekonnen et al., 1999). This variability resulted from the sporadic rooting patterns of the trees. Thus, sufficient replications must be used if small differences in 15N uptake are to be detected.

Residual 15N enrichment of the soil samples taken at various lateral and vertical distances from the 15N injection points revealed more downward than lateral movement of the injected 15N (Fig. 5). This could have been as a result of leaching, although there remained a pronounced enrichment of 15N close to the injection point even 9 wk after 15N application. To reduce 15N movement in soil, (15NH4)2SO4 could be applied in combination with a C source such as sucrose at a C/N ratio of 10:1 to facilitate microbial immobilization of the 15N, and thus restricting it more closely to the applied depth (Witty, 1983; Rowe et al., 1999). Alternatively, 15N can be applied in the form of resin-adsorbed 15N whereby weakly acidic cation adsorption resins are saturated with enriched fertilizer; however, difficulties are often encountered in achieving sufficient 15N saturation in resins, which leads to inefficient use of the isotope.

The vertical movement (leaching) of the placed 15N fertilizer leads to some of it being taken up by plant roots at soil layers below the placement point and hence the uptake should be considered to be from that soil layer rather than from a point. The application depth thus represents the minimum depth at which the labeled N is placed. The amount of 15N applied should provide sufficient enrichment but should not significantly increase the available soil N pool around the injection points as this could alter the uptake pattern and thereby introduce some errors (Witty, 1983). Assuming that applied N extended horizontally largely in a radius of 0.20 m (as indicated by our results), the application rate was thus equivalent to around 40 kg N ha-1 within this area. This corresponds close to the 27 to 30 kg inorganic N (NO3–N + NH4–N) ha-1 found at the two depths at the start of the experiment (Gathumbi et al., 2002b). Potential interferences of applied N with root activity patterns could be further reduced by applying less fertilizer N but at higher enrichment (e.g., 99 atom% 15N).

The 15N placement method was successful in quantifying and separating N uptake patterns from different depths by individual species in mixed stands. The 15N recovery for the species reported here were higher than those reported for peltophorum and gliricidia (Rowe et al., 2001) ors Calliandra calothyrsus Meissn. and Leucaena leucocephala (Lam.) deWit. (Mugendi et al., 1999) in hedgerow intercropping systems. In those studies the 15N enriched materials were applied away from the tree rows, as opposed to this study where the 15N fertilizer was applied very close to the target plants.

The low 15N recovery by border tree plants at all placement depths suggested that the plant roots extended more vertically than horizontally relative to the application sites in these relatively dense stands. Plant roots in monoculture or in mixed species systems are likely to have variable horizontal growth and distribution depending on the nutrient and water availability at topsoil layers and also the degree of competition with the neighboring plants (Fitter, 1991; Caldwell, 1994; van Noordwijk and Purnomosidhi, 1995; Schroth, 1998). Trees are known to have more extended horizontal root growth than shrubby species or even annual crops and hence may source nutrients far from their growing position. As such, in studies on nutrient uptake by plant roots using 15N techniques, it is important to know horizontal root activity of the specific species before setting up the experiment, to reduce the chances of plants scavenging the applied 15N from the neighboring plot. In the current study, the enrichment of the leaf samples obtained at various distances from the injection point revealed that in this type of experimental setup, the minimum spatial placement distances between two 15N application sites should be >4 m for crotalaria and >6 m for sesbania (Fig. 4). Rowe and Cadisch (2002) recommended a minimum interplot spacing of 6 to 8 m for short-term 15N experiments with gliricidia and peltophorum hedgerow intercropping systems. Trees that are substantially larger than those studied here, however, are likely to have more extensive root systems and larger distances will be required. Tree rooting extent is also likely to be greater on sites with greater moisture deficits, more infertile soils, or shallower rooting depths (Helliwell, 1986).

Complementarity in Soil Nitrogen Acquisition
The differences in 15N uptake showed that sesbania had more root activity at lower soil depths than either siratro or crotalaria, despite sesbania and crotalaria having similar Lrvs at 1-m depth in single species stands (Fig. 6). However, mixing sesbania with crotalaria increased the amount and distribution of roots, particularly in the 0.30- to 0.60-m depth, but also at lower soil depths up to 1.2 m. This suggests that mixing sesbania with crotalaria may have increased competition for growth resources (nutrients and water) in the topsoil layer and hence this overlapping of feeding niche stimulated downward development of roots to lower soil depths. In other words, more roots needed to be established at lower layers to increase subsoil nutrient capture driven by the N demand within each species. Given the higher 15N uptake of sesbania at the lower soil depths, it may be assumed that sesbania Lrv was increased in this case. Unfortunately, it was not possible to distinguish the two root systems with certainty to verify this hypothesis.

Van Noordwijk et al. (1996) noted that increased root establishment in species mixtures could enhance complementarity in resource use at subsoil horizons in an attempt to acquire an equilibrium between growth resource availability and plant-root uptake because the vertical distribution of roots is greatly influenced by the relative success of roots in taking up the most limiting resource in the upper soil profile. The higher Lrv for crotalaria than sesbania in topsoil and high 15N recovery from topsoil by crotalaria supports the hypothesis that sesbania root development at lower soil depths was stimulated by competition from crotalaria. This is further corroborated by the fact that on a per plant basis sesbania took up more 15N from 1-m depth when mixed with crotalaria than when mixed with siratro.

The higher topsoil recovery of the applied N by the sesbania + crotalaria mixture compared with the sesbania + siratro mixture corresponded with the higher total root length (Table 1) and Lrv (Fig. 6) in the topsoil (0–0.30 m). Because of the similar distribution of fine roots for both mixed species fallow treatments as depicted by the SRL measurements, the higher uptake of 15N by sesbania + crotalaria treatments could largely be attributed to the Lrv, which is reported to greatly influence the soil nutrient accessibility to plants (van Noordwijk and Brouwer, 1991; Schroth and Zech, 1995; Schroth, 1998).

Root data obtained for sesbania was comparable with those reported in previous studies conducted in the study area. For instance, root length recorded for the 0- to 0.3-m soil depth for 11- and 15-mo-old sesbania fallows planted at a spacing of 1.0 by 1.0 m was 1.3 km m-2 (Jama et al., 1998) and 1.5 km m-2 (Mekonnen et al., 1997), respectively, compared with 1.5 km m-2 reported in this study. The high SRL for sesbania in this study for 0- to 0.3-m soil depth (35 m g-1) compared with 12 m g-1 reported for the 11-mo-old sesbania fallow reported by Jama et al. (1998) could be attributed to the younger age and the higher plant density. Specific root length is an indication of the relationship between root biomass and the ability of the root systems of crops to penetrate a given soil (van Noordwijk and Brouwer, 1991). The values obtained for SRL measurements observed in most of the fallow tree and shrub species indicate that these species require less root biomass to produce a unit length of roots compared with other species with a higher proportion of coarse roots. In other words, more fine roots were observed for all the species, which can be attributed to their relatively young age.

Elucidation of Temporal Changes in Nitrogen Uptake Patterns
The recovery of the labeled N per site or even for individual species did not necessarily correspond with the amount of total biomass or N yield accumulated. This could be because of the fact that 15N uptake measurements corresponded only for the later growth phase of the fallows and was partly influenced by the onset of the dry season (Fig. 1). Thus 15N uptake pattern are not only influenced by Lrv and 15N distribution but also strongly affected by plant N demand and accessibility of water and N during this phase. The onset of the dry season could explain why a high proportion of 15N was taken up from a depth of 1 m. This is further corroborated by the decline in 15N uptake in the topsoil toward the end of the experiment, while the opposite occurred in the lower soil layer (Fig. 3). Thus the 15N data gives additional insight into the temporal pattern of N uptake at various soil depths, which are not revealed by root mapping observations. Although 15N enrichments of young leaves give useful insights into the temporal uptake patterns of labeled N of a given species, it must be taken into consideration that 15N enrichment in young leaves is not an indicator of total plant 15N uptake, as this depends on total plant N yield and 15N distributions within the plant (Rowe et al., 1999). The index hence has to be used with caution and put in the context of other parameters, for example, in our case, the relative differences in 15N uptake patterns between topsoil and subsoil (Fig. 3) correspond with the actual total 15N uptake (Table 3).

A further reason why uptake of soil mineral N by legumes is not necessarily correlated to growth demand is the potential of legumes to acquire N from atmospheric N2 by means of their symbiotic relationship with Rhizobium. Both sesbania and crotalaria have been shown to obtain over 50% of their N demand from symbiotic N2 fixation (Gathumbi et al., 2002a). In fact, crotalaria obtained over 75% of its N from N2 fixation compared with around 50% by sesbania, suggesting that this species needed to acquire more soil N and this may explain the higher 15N uptake activity. It may also be argued that the much faster early uptake of 15N by sesbania compared with crotalaria (Fig. 3) further suggests a higher dependency of sesbania on soil mineral N.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
The 15N placement technique for assessing subsoil N uptake by the test legume species can be used with an appreciable degree of certainty in tracing 15N uptake by legume species in fallow systems and also in studies on rooting patterns. The method appears particularly useful in mixed species stands to distinguish root activity between species at a given depth and to assess root plasticity in response to competition by the companion species. In this study, the uptake of the labeled 15N from different soil depths by different species in fallows corresponded with total root length and the Lrv. However, when designing such experiments, species rooting patterns and distribution should be considered because these constitute the highest source of error as has been observed in this study. Increasing the number of treatment replications and sampling points may partly reduce the experimental errors associated with plant root heterogeneity; while increasing 15N enrichment of N applied will allow reducing the amount of N applied hence avoiding preferential root responses to added N.

Whereas legume species evaluated in this study could be used for subsoil N retrieval when grown in sole species fallows, the root, and 15N recovery data accruing from this study suggest that there is complementarity in the system when they are grown in mixtures. For instance, when sesbania and crotalaria were mixed there was more subsoil N uptake by sesbania while crotalaria is reported to source a higher proportion of its N from N2–fixation and topsoil N uptake. Again it appears that mixing crotalaria and sesbania introduces competition of growth resources in the topsoil and thus more roots are established at lower zones, expanding the nutrient uptake zone for the two species.


   ACKNOWLEDGMENTS
 
We gratefully acknowledge the assistance of both the field and laboratory research teams at the ICRAF/KEFRI/KARI Maseno Agroforestry Research Centre and Nyabeda field site. This publication is an output from two projects: Agroforestry Research Network for Africa (AFRENA) funded by the European Union and NRSP R7056 funded by the UK Department for International Development (DfID) for the benefit of developing countries. The views expressed are not necessarily those of the funding agencies.

Received for publication February 12, 2002.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via ISI Web of Science (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gathumbi, S. M.
Right arrow Articles by Giller, K. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Gathumbi, S. M.
Right arrow Articles by Giller, K. E.
Agricola
Right arrow Articles by Gathumbi, S. M.
Right arrow Articles by Giller, K. E.
Related Collections
Right arrow Crop Rotation Systems
Right arrow Nutrient Cycling
Right arrow Isotopes

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