| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a MacArthur Agro-Ecology Res. Cent., 300 Buck Island Ranch Rd., Lake Placid, FL 33852
b Kenya Forestry Res. Inst., P.O. Box 20412, Nairobi, Kenya
c Dep. of Agric. Sci., Imperial College at Wye, Univ. of London, Wye, Kent, TN25 5AH, UK
* Corresponding author (sgathumbi{at}archbold-station.org)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: LAI, leaf area index • PAR, photosynthetically active radiation
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Mixing species with compatible and complementary rooting and/or shoot growth patterns in fallows may have multiple advantages. Undersowing herbaceous or shrubby legume species under the tall, open canopy of sesbania may improve light capture and lead to gains in net primary productivity of the fallow. Planting shallow-rooted species together with species that are able to root to depth may enhance uptake of water and nutrients from the soil profile, preventing losses of nutrients by leaching into the subsoil. Mixing species may also increase fallow residual benefits for the next crop by improving the synchrony of nutrient supply and crop demand if the species grown in mixtures differ in their residue quality (Handayanto et al., 1997; Palm, 1995; Swift et al., 1979).
Fallows comprised of species mixtures may therefore be superior in their potential to replenish soil fertility and can provide a wider range of fodder and wood products than pure-species fallows (Kerr, 1999; Rao et al., 1997). Mixing fallow species may also reduce the risk of failure in establishment or productivity due to constraints such as drought, disease, or pest attack.
The objectives of this study were to assess growth compatibility and productivity, soil water and N uptake, and aboveground plant N for a range of promising shrubby legumes in both pure and mixed fallows. The feasibility of undersowing groundnut or the creeping forage legume siratro as understories between the rows of the woody fallow legumes was also explored as this may be an attractive option for farmers. Three hypotheses were tested: (i) Mixing species in fallows can increase aboveground resource capture, (ii) mixing species in fallows can increase belowground resource capture, and (iii) mixing species in fallows can provide a wider variety of useful products for farmers.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
Plant Establishment
One-month-old plant seedlings grown in a nursery were transplanted to the field on 20 Oct. 1997. Plot size was 6 by 5.25 m (31.5 m2). Tree and shrubby legumes were planted at a spacing of 0.75 by 0.75 m in both pure- and mixed-species fallows, with species in mixed stands planted in alternate rows. Siratro and groundnut in pure-stand treatments were planted at a spacing of 0.375 by 0.75 m for siratro and 0.20 by 0.375 m for groundnut. In mixed fallows, these species were planted in between rows of tree and/or shrub legumes at the same interrow spacing as in pure-stand treatments.
Measurements during Establishment
Root collar diameter and plant height were monitored on a monthly basis after transplanting. Photosynthetically active radiation (PAR) was measured 2 and 3 mo after transplanting using a portable ceptometer consisting of a narrow, 80-cm-long probe with 80 sensors. Canopy transmittance and the total incoming PAR in open areas were used to calculate the leaf area index (LAI) using the Beer–Lambert equation where Q1 = canopy PAR transmittance, Q0 = total incoming PAR, and k = canopy light extinction coefficient (Vose and Swank, 1990). In this method, the calculation is based on the assumption that the foliage and individual leaves are randomly distributed. Five ceptometer readings were taken for both above and below the canopy at five locations within the treatment plots, i.e., four measurement points at 1 m from each corner of the plots and one at the center.
Fallow Harvesting Procedures
Fallow harvesting was carried out between 6 and 8 Apr. 1998 on a block basis, except for groundnut, which had been harvested on 20 Feb. 1998. For trees and shrubby fallow species, and for siratro, the net harvest area was 2.25 by 3 m (6.75 m2). Net harvest area for groundnut was 3.75 by 3.75 m (14 m2). Plants were cut at ground level and separated into stems, branches (>5 mm), and foliage (leaves and pods). Fresh weight was immediately determined in the field for each plant component in the harvest area. Subsamples of each plant component were taken and fresh weight determined. Subsamples were then placed in plastic bags, packed in a cool box, and transported back to the laboratory where they were oven-dried at 70°C for 48 h, weighed for dry matter determination, and then finely ground using a micro hammer mill. Groundnut plants were separated into pods (grains and husks), foliage, and roots. All the pods in the harvest area were weighed fresh in the field, sun-dried, and subsequently shelled.
Dry plant samples were analyzed for total N using the Kjeldahl procedure (Anderson and Ingram, 1993), and NH4 was then determined colorimetrically using an SFA-2 Burkard scientific autoanalyser (Bukard Sci., Uxbridge, Middlesex, UK).
Plant mixtures show the result of interspecies competition compared with intraspecies competition in sole stands. We determined the effect of interspecies competition by comparing aboveground plant N for individual species in sole and mixed stands. We calculated plant N differences based on same plant density. The effect of competition on changes in plant N (PNdiff, expressed in kg N ha-1) for a given species when mixed with other woody species grown at 50:50 plant populations was calculated using the equation:
 |
 |
Soil Sampling and Analysis for Mineral Nitrogen Determination
Initial soil sampling was carried out in September 1997 just before the fallow establishment, and the final soil sampling was done between 26 March and 4 April 1998. Using an Edelman auger with a diameter of 7 cm, soils were sampled to 200-cm depth at the following six depths: 0 to 15, 15 to 30, 30 to 50, 50 to 100, 100 to 150, and 150 to 200 cm. On both sampling times, all of the treatment plots were sampled individually. Soil augering was done at six spots per plot and samples bulked for layers above 100 cm and four spots for layers below 100 cm. Soil samples were analyzed for gravimetric soil water content and for mineral N (NH4–N and NO3–N), with two subsamples for each sample, after extraction of 20 g of field moist soil with 2 M KCl. The resulting extract was analyzed for NH4–N by a colorimetric method following the procedure of Anderson and Ingram (1993) and for NO3–N and NO2–N by Cd reduction (Dorich and Nelson, 1984), with subsequent colorimetric determination of NO2 (Hilsheimer and Harwig, 1976). Cores were taken at each selected depth to determine bulk density, which was later used to convert N values from mg kg-1 into kg ha-1 and gravimetric water content to m3m-3.
Data Analysis
All data on growth parameters, biomass, and plant N were subjected to one-way analysis of variance (ANOVA) using Genstat statistical software (Payne et al., 1987). Wherever appropriate, data were grouped according to species and their respective mixtures. Mineral N data were not normally distributed, and hence were square root–transformed before conducting the ANOVA using the procedures of Gomez and Gomez (1984). Tests of significance between treatment means were performed using the Tukey test at P = 0.05. Standard errors of the difference are reported to enable comparison of treatment means.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
1.5 m tall) and 3 mo after fallow establishment, already had a completely closed canopy so that light penetration was minimal (Fig. 2)
. Sesbania, which grew slowly initially, was the tallest tree at the end of the experiment, reaching almost 3 m. However, sesbania and pigeonpea fallows had a less dense aboveground canopy at the early fallow growth phase compared with tephrosia and crotalaria (Fig. 2). Fallows consisting of crotalaria, tephrosia, and siratro completely smothered the undergrowth of weeds because of reduced light penetration due to their dense canopies.
|
Sole groundnut yielded 1.02 Mg ha-1, which was significantly (P < 0.05) more than in the intercrop treatments. Groundnut yielded significantly less grain (0.2–0.3 Mg ha-1) in crotalaria and tephrosia intercrops than in sesbania and pigeonpea intercrops (0.4 Mg ha-1).
Fallow Aboveground Biomass Production
Total aboveground dry matter production in the improved fallows ranged between 8.7 and 13.2 Mg ha-1 in woody single-species treatments and between 8.2 and 15.2 Mg ha-1 in mixed-species treatments (Fig. 3a)
. In most cases, there were no significant differences in the aboveground biomass production between pure- and mixed-species plots.
|
Wood dry matter production (the difference between total biomass and foliage biomass) varied strongly among the species: Tephrosia yielded the largest amount of wood (9.3 Mg ha-1) compared with sesbania (7.8), crotalaria (5.7), and pigeonpea (5.2). Among the mixtures, pigeonpea + siratro (3.0 Mg ha-1) and sesbania + siratro (11.8 Mg ha-1) gave the smallest and the largest wood yields, respectively. Pigeonpea yielded more wood in mixtures, except when combined with siratro. By contrast, wood yield decreased from 4.6 Mg ha-1 to 2 to 3.5 Mg ha-1 with tephrosia mixtures, whereas wood production of crotalaria and sesbania was little affected by mixing.
The high wood/foliage ratio could explain the relatively small foliage yields in both sole sesbania stands and in its mixtures compared with other fallow species. A low wood/foliage ratio and relatively high N content in shoots (30 g kg-1) led to the large biomass and plant N of crotalaria. All fallow mixtures that included siratro had a greater foliage component (22–59%) of the total biomass production as the density of the companion species was not altered when siratro was included as an understory. Siratro produced substantially more biomass in pure stands (5 Mg ha-1) than when intercropped with the shrubby species (1.2–2.5 Mg ha-1) but is an attractive option for inclusion in fallows for soil fertility management by farmers due to its added value as forage.
Fallow Aboveground Plant Nitrogen
Among the improved fallow tree and/or shrub monocultures, sesbania accumulated the least aboveground N and crotalaria the most (Table 2). Among the mixtures, total aboveground plant N was largest in the tephrosia + siratro mixture (224 kg N ha-1) and least in the sesbania + groundnut mixture (98 kg N ha-1). No significant differences were observed in total plant N among the crotalaria and pigeonpea mixtures.
|
Mixing pigeonpea with crotalaria resulted in 11 kg ha-1 more plant N for pigeonpea than when pigeonpea was grown in monoculture at equal plant density (Fig. 4) . Wood N changes for pigeonpea grown in mixtures ranged between 24 and -12 kg N ha-1 when grown with sesbania and with siratro, respectively. Crotalaria was most strongly negatively affected by competition from other species in absolute terms (Fig. 4), resulting in total N reductions of 20 to 71 kg ha-1 when mixed with tephrosia and groundnut, respectively. However, crotalaria wood N, unlike foliage N, was not strongly affected by competition. Mixing sesbania with pigeonpea, tephrosia, and siratro resulted in increased sesbania foliage N compared with the monoculture. Similarly, sesbania wood N increased in most of the mixtures, except with pigeonpea where it was reduced by about 8 kg ha-1.
|
28 kg N ha-1 in wood (Table 2). In fallow mixtures, the smallest amount of wood N was found in crotalaria + groundnut plots and the largest amounts in crotalaria + sesbania. Wood N in mixtures was not significantly different from that of respective single-species wood, except for the crotalaria + sesbania mixture, which contained significantly more N in the wood than crotalaria alone.
Effect of Fallows on Mineral Nitrogen and Water Storage in the Soil
Initial mineral N content for the soil profile to a depth of 200 cm totaled 152 kg N ha-1. The surface 15 cm had similar amounts of mineral N as the deepest layer sampled (150–200 cm), and less NO3–N and total mineral N were present at 30 to 50 cm than in both the upper and deeper layers.
After 6 mo, there was less mineral N in the surface 50 cm with all treatments, except in the calliandra and groundnut plots (Fig. 5) . The amount of mineral N had increased at depths between 30 and 100 cm in most of the fallow systems but was often depleted in the deepest layers sampled. Subsoil N (150–200 cm) was decreased by 8.1 kg ha-1 in sole sesbania fallows compared with a decrease of 4.1 kg ha-1 in the sole crotalaria plots and a net increase in pigeonpea fallows. Conversely, topsoil (0–15 cm) mineral N was decreased more strongly in sole pigeonpea, crotalaria, and siratro plots compared with sole sesbania. The crotalaria + sesbania mixture resulted in relatively more decreases in both topsoil and subsoil N compared with sole crotalaria and in topsoil compared with sole sesbania. Combinations of tephrosia and pigeonpea also resulted in slightly more topsoil and subsoil N extraction compared with their respective monocultures. Results also suggested slightly less topsoil N present after sole pigeonpea than after tephrosia but more subsoil N after pigeonpea than tephrosia.
|
In April 1998, 6 mo after fallow establishment, the soil water content at lower soil depths was less in fallow treatments than under maize or natural fallow treatments or than the initial soil water content (Fig. 6) . The difference increased with soil depth, particularly with respect to maize, which showed the lowest soil water uptake below 100 cm. Sesbania fallows depleted soil water content in the subsoil most strongly while the highest soil water content among the improved fallow treatments was found under the sole siratro plots.
|
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The differences observed in total aboveground biomass and plant N for the specific species in mixtures can be attributed to the degree of complementarity and competitive ability compared with the companion species. For instance, inclusion of sesbania in mixtures resulted in reduction in total biomass yield of crotalaria (14%), tephrosia (20%), and siratro (50%) compared with monocultures at equal plant densities. On the other hand, sesbania profited from being associated with the other species; the increases in wood biomass and plant N (Fig. 4) and changes in soil mineral N profiles suggest that this was achieved through a better exploitation of the upper soil layers by the mixtures. Thus, in most cases, reduction in biomass and plant N in the less competitive species was compensated for by increased yield of the more competitive species in the mixture. This largely explains the lack of significant differences in the overall fallow biomass and N performance between sole-species fallows and their respective mixtures in the majority of cases.
The lower-growing species, such as siratro (a creeping legume) and crotalaria (a short shrub), suffered most from competition as could be expected. However, other results suggest that the balance of competition and complementarity is often fragile. For example, in this study, sesbania was very competitive, but it has been observed that when the fallows were established by direct seeding, sesbania develops slowly and suffers from competition by the associated species.
Complementarity in Belowground Resource Capture
Accumulation of mineral N for the entire profile to 200 cm-depth at the end of the fallow period showed that the different systems either depleted or spared N present in the soil mineral N pool. Amounts of mineral N were decreased in the surface and deepest soil layers after 6 mo under the different improved fallow systems, i.e., there was a depletion in mineral N at depths of 0 to 30 cm and 150 to 200 cm for most of the improved fallow systems (Fig. 6). The depletion of the soil mineral N could be attributed to the net active N uptake by plant roots and/or downward movement (leaching) although the latter is thought to be small in unfertilized fallow systems where plants are actively absorbing N from the soil solution. The resulting net effect on soil mineral N could be attributed to the total soil N uptake by the plants; continuous mineralization of the N from the soil N pool; and some mineralization of the roots, nodules, and fallen litter that may occur during the fallow (although litter fall in these short-term fallows was neglible). For example, groundnut plots in which the harvested biomass was returned to the plot after harvesting 2 mo before the final soil sampling showed a net positive mineral N accumulation in sole and intercropped stands in improved fallows. In contrast, the large accumulation of soil mineral N after the calliandra fallow (256 kg N ha-1 total profile mineral at harvest) resulted from its slow growth, and hence limited plant N uptake (45 kg N ha-1), demonstrating the considerable potential of these soils to supply mineral N.
A similar degree of complementarity in belowground resource capture was observed as for aboveground resource utilization. Mixing sesbania and crotalaria resulted in increased topsoil and subsoil N uptake compared with sole crotalaria. The two species in monocrop fallows already exhibited potential for complementarity in soil mineral uptake at different soil depths. Sesbania reduced subsoil N at 100 to 200 cm twice as much as crotalaria while the latter extracted more of the topsoil N (0–15 cm). The increased mineral N uptake from both topsoil and subsoil by sesbania + crotalaria mixture contributed to a larger total N in aboveground biomass compared with sole sesbania fallow. A similar phenomenon was observed for pigeonpea and tephrosia in sole-species fallows where pigeonpea extracted less subsoil N than tephrosia and the reverse was true for the topsoil N. Mixing pigeonpea + tephrosia resulted in increased exploitation of subsoil N compared with their respective monocultures.
Cadisch et al. (2002) confirmed the higher subsoil N uptake activity by sesbania compared with crotalaria by injection of 15N into the subsoil in the same experiment. However, subsequent results also revealed that such differences may not always be consistent and indeed may reverse in mixtures if one species establishes poorly or is affected by pests or diseases. Thus, mixing species may also provide a means to reduce risks where adverse conditions affect one species more than the other.
The different fallow systems had varying abilities to extract or conserve soil water, and these results complemented and confirmed observations on mineral N depletion patterns discussed above. For instance, the greater subsoil water extraction by sesbania compared with crotalaria closely reflected the mineral N depletion patterns of the two (more subsoil mineral N was depleted by sesbania), demonstrating that extraction of water and N are closely linked (Fig. 5 and 6).
The soil water stored under sole-siratro plots after 6 mo was substantially greater in the top 50 cm than the initial soil water content and was much greater than in the sole sesbania fallows in the same soil layers. Mixing the two species improved the soil water retention in the top 15 cm compared with the initial soil water content. This can partly be explained by the dense ground cover by siratro, which decreased water losses from the soil surface through reduction of surface runoff and evaporation and enhanced infiltration. On the other hand, sesbania and crotalaria fallows either in sole stands or mixtures extracted more soil water throughout the soil profile. Similarly, Hartemink et al. (1996) concluded that pure sesbania fallows were more efficient in extracting subsoil water than maize and natural weed fallows. In the current study, the topsoil under maize and sole-groundnut treatments was drier than the deeper soil layers, which had not changed much from the initial soil water content (data not presented). This can be attributed to the fact that the two crops were harvested 2 mo before soil water determination and the plots were bare, leading to reduced plant water uptake and increased surface water loss. This results in a higher rainfall requirement to recharge the soil water content and, thus, potentially delayed planting of subsequent crops if rains are scarce.
Complementarity of Uses: Soil Fertility Improvement, Wood, Fodder, and Grain
Among the different tree fallow species and fallow types investigated, the accumulation of woody biomass was, in all cases, greater than foliage production. The wood component of the fallow species is often removed from the plot after fallow harvest, and this contributed substantially to the net nutrient export from the system (e.g., 9–61 kg N ha-1). Sesbania in pure and mixed systems produced the highest proportion of woody biomass (72–78%) (Fig. 3a and 3b). As a secondary product for a fallow intended for soil fertility improvement, the woody component of the fallow can be used as domestic firewood, stakes, and fencing materials. The quality and quantity of wood varies among species, and this can influence a farmer's decision as to which species to adopt. For instance, sesbania wood could be utilized for simple construction and fencing purposes while wood from crotalaria, pigeonpea, and tephrosia could be used as stakes as well as firewood.
Mixtures that included siratro had the highest production of total foliage biomass because including the species as an understory did not alter the density of the companion species. Compared with the pure-siratro plots, siratro in the intercrops had relatively lower biomass production (24–50%). It remains to be tested how the removal of fodder will affect the subsequent impact on soil fertility, but based on the current observations, it appears that the remaining fallow species would provide sufficient inputs for provision of mineralizable N for the next maize crop. Measurements of N2 fixation of these species have also suggested inputs of 70 to 150 kg N ha-1 from the atmosphere, reinforcing their value for soil fertility improvement apart from improved recycling of deep soil N (Gathumbi et al., 2002).
Inclusion of grain legumes as an understory within a fallow system consisting of legume tree and/or shrubby species offers the extra incentive to farmers of an immediate return in cash, food, or both. Groundnut yields were substantially better when intercropped with species with an open canopy structure, such as pigeonpea and sesbania compared with crotalaria and tephrosia, which developed a more dense aboveground canopy within 2 mo after fallow seedling transplanting, as demonstrated by their larger LAI (Fig. 2). Pigeonpea can also produce pods and edible grains in fallows of longer duration than the ones tested here.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The second hypothesis that belowground resource capture can be increased by mixing species was confirmed for selected species with complementarity in root activity patterns. Complementarity was depicted, for example, by differences in soil water depletion at different depth, such as the case of sesbania and crotalaria, which led to an efficient depletion of soil mineral N to depth in the mixed fallow.
The third hypothesis that mixed species could provide a better basket of secondary products while maintaining a high production of foliage biomass for high supply of subsequent soil mineral N was also shown possible with several of the mixtures evaluated: Combining sesbania and crotalaria gave a substantial wood component, undersowing the forage legume siratro provided additional fodder, and groundnut produced a moderate yield of grain.
Based on the results obtained from the different sole species and mixtures, it can be concluded that there is a wide variety of species and fallow types (combined species fallows) from which farmers may choose, depending on their specific needs and preferences. Mixing species requires a good understanding of the early establishment phase as certain species were competitive although sequential timing of planting may alleviate such problems.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Sharma, M. K. Aneja, J. Mayer, J. C. Munch, and M. Schloter Diversity of Transcripts of Nitrite Reductase Genes (nirK and nirS) in Rhizospheres of Grain Legumes Appl. Envir. Microbiol., April 1, 2005; 71(4): 2001 - 2007. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Gathumbi, G. Cadisch, R. J. Buresh, and K. E. Giller Subsoil Nitrogen Capture in Mixed Legume Stands as Assessed by Deep Nitrogen-15 Placement Soil Sci. Soc. Am. J., March 1, 2003; 67(2): 573 - 582. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. H. Graham and C. P. Vance Legumes: Importance and Constraints to Greater Use Plant Physiology, March 1, 2003; 131(3): 872 - 877. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
