[en] This research is a part of a longstanding effort aimed at the ecological and economic rehabilitation of the anthropogenic savannahs, locally known as 'espinales', which occupy over 2 000 000 ha, and support the livelihoods of more than 350 000 people in the Mediterranean type climate region of central Chile. The field experiment was carried out in the subhumid Mediterranean zone of Chile from 1999 to 2005. The treatments included comparisons of the traditional agroforestry system in the region, consisting of a natural annual pasture associated with a sparse population of Acacia caven; with an alternative agroforestry system composed of annual legume-based pastures associated with Tagasaste (Chamaecytisus proliferus var palmensis). In the treatment that included Tagasaste, an ¹⁵N tree injection technique was used to measure the contribution of tree N to the crop. The agroforestry systems that include annual legumes in rotation with wheat, or Tagasaste with natural pasture in rotation with wheat, have the potential to improve the yield and nitrogen uptake by the wheat crop. (author)

[en] Agroforestry has been identified as a sustainable land use system for degraded lands in the humid tropics. The dynamics of N in a Gliricidia-maize alley-cropping system was studied through the injection of ¹⁵N-labelled fertilizer into Gliricidia trees, and subsequently the ¹⁵N label was traced into the maize crop through Gliricidia root decomposition and prunings incorporated in the surface soil. The experiment was conducted at the University of Peradeniya, Sri Lanka. The roots of 3 year old Gliricidia trees were isolated by lined trenches and the trunks were injected with ¹⁵N-enriched ammonium sulphate (98 atom %, 1.415 g tree¹) and harvested 4 weeks after labelling. Harvested above ground residues were applied to 3 equal compartmentalized sections around isolated trees (2 m radius) to form three treatments. T1: ¹⁵N labelled above ground Gliricidia residue and ¹⁵N-labelled Gliricidia roots remained in the compartment (roots not removed); T2: similar amount of nonlabelled above ground Gliricidia residues and ¹⁵N-labelled roots remained; T3: non-labelled Gliricidia residues and ¹⁵N-labelled Gliricidia roots removed. Four weeks after incorporation of Gliricidia residues to the respective plots, a maize crop was planted. N cycling and corn productivity were measured. The ¹⁵N distribution among different parts of Gliricidia 4 weeks after injection was 71, 15 and 1.7% in the stems, leaves and roots, respectively, with the total recovery of 88%. ¹⁵N recoveries by the subsequent maize crops amounted to 13% from Gliricidia foliage and roots. Approximately 10% of the N recovered by maize was from leaves and stems of Gliricidia residues, while around 2% was recovered from Gliricidia root residues. Surface soil organic C and N contents increased with the addition of Gliricidia residues up to 4-6 weeks and gradually declined with time. Soil C derived from Gliricidia residues ranged from 21-36%. The injection of ¹⁵N into the tree component of an agroforestry system promises to be an effective way to evaluate N cycling in mixed cropping systems. (author)

[en] The dynamics of N released from fresh leaves of Paraserianthes falcataria when applied as surface mulch to an Ultisol, and their effects on soil acidity, were studied under tropical field conditions in Malaysia. ¹⁵N-labelled residues were applied to confined microplots to trace the pathways and fate of mineralized N, and resin bags placed at 20 cm soil depth under the microplots collected leached inorganic N for quantification. A control treatment consisted of no residue application. The microplots were kept free of vegetation. Decomposition of the high quality residue was very fast with the likelihood that crop N demand would not be in synchrony with the rate of N mineralization. Nitrate was readily leached to 20 cm depth suggesting a high potential for N loss and poor N use efficiency in the system. The leaf mulch had a short term positive effect in the 0-10 cm soil depth as an ameliorant of soil acidity. (author)

[en] Agroforestry is a low-input system which combines trees with crops in various combinations or sequences. It is an alternative to intensive cropping systems, which rely on large inputs of manufactured fertilizers and other external inputs to sustain production. Agroforestry also has the potential to reduce risk through diversification of a variety of products, including food, fuelwood and animal fodder. Other perceived benefits include enhanced nutrient and water use efficiencies, reduced nutrient leaching to groundwater and improved soil physical and biological properties. The use of leguminous or actinorhizal trees may further enhance these benefits because of their capacity to fix atmospheric nitrogen. Depending on the type of agroforestry system and the management practices employed, a substantial portion of this fixed nitrogen can be transferred to companion crops and to the soil. In considering the overall productivity of agroforestry systems, it is essential to investigate the competition or complementarity in the capture and partitioning of resources between tree and crop components. This is especially true for nutrients and water, usually the two most limiting factors influencing crop growth. The focus of this coordinated research project (CRP) was to evaluate the efficacy of various agroforestry systems used in Member States in terms of crop productivity, resource use efficiency and improvements in soil properties. The use of isotopes and nuclear techniques was essential for understanding the dynamics of nutrients and water in agroforestry systems. The contribution of nitrogen from fertilizers and leguminous trees to soil and crops was studied using both direct and indirect ¹⁵N labelling techniques. The cycling of carbon from trees or crops to soil was studied using natural variations in the ¹³C signatures of the soils and the different species. The soil moisture neutron probe in conjunction with tensionics was used to monitor soil water status and balance, and differences in the ¹⁸O composition of rain, surface and groundwaters were used to identify the sources of water used by trees and crops. This CRP was implemented through a research network that initially included nine contract holders from Benin, China, Chile, Costa Rica, Kenya, Malaysia, Sri Lanka, Uganda and Zambia, and five agreement holders from Australia, France, Kenya, Nigeria and the UK. The CRP was conducted in collaboration with National Agricultural Research Systems in Africa, Asia and Latin America, and with three CGIAR Centres, the International Centre for Research in Agroforestry (ICRAF, Kenya), the International Institute of Tropical Agriculture (IITA, Nigeria) and CIAT's Tropical Soil Biology and Fertility Institute (TSBF, Kenya). This IAEA-TECDOC contains 15 manuscripts prepared by the project participants

[en] The objective of the research was to study the potential of alley-cropping agroforestry systems to improve degraded lands in the Mid-Country Intermediate Zone of Sri Lanka. A field experiment was carried out at the University of Peradeniya experimental station using Gliricidia as the hedgerow tree species. Gliricidia hedgerows having within row spacing of 0.75 m (8 m long rows) and between row spacing of 7 m were established. ¹⁵N-enriched ammonium sulphate (60 kg N ha¹, 10 atom % ¹⁵N excess) was applied to a subplot (2.25 m wide x 3.5 m long) in the alley, which enclosed 3 Gliricidia trees. Labelled plant material was added to microplots as crop residue only (ML), Gliricidia loppings only (GL), crop residue and Gliricidia loppings (GLML), and no residues (NL) in the subsequent seasons. Soil properties, crop yields and nutrient dynamics were recorded regularly for every growing season.Addition of Gliricidia loppings and crop residues over 5 a improved soil chemical properties including soil organic matter, major (especially N) and minor nutrients and physical properties. There was no significant impact of hedgerows on soil fertility compared with the sole crop or at different distances from the hedgerow. Physiological parameters measured in this study illustrated that hedgerows may influence one or two adjacent crop rows negatively, possibly due to competition for light or water resources. Photosynthesis rates of both maize and cowpea crops were reduced near the hedgerow compared to the sole crops, due to partial shading by the hedgerow. However, leaf photosynthesis and yields were significantly greater in rows in the middle of the alleys compared with the sole-crop (control). This would suggest the existence of complementary interactions from Gliricidia hedgerows through increasing the resource availability and/or making the microenvironment more favorable for the crop species. Addition of Gliricidia and crop residues (MLGL) enhanced growth and yield of crops more than the other treatments. Addition of similar litter amounts to the sole crop also resulted in similar trends. ¹⁵N recovery by the maize crop was 25% in seeds and 23% in leaves and stems (in total 48%). (author)

[en] Tree and crop components of an alley cropping system may compete for nutrients and water in the surface soil while the deeper-rooting tree may have better access to resources in the subsoil. The objectives of the present study in an alley cropping system were to monitor runoff and soil loss, spatial variation of soil water, and tree sap flow to determine competitiveness for water between the tree and peanut. Plant and root growth were also recorded to evaluate competition. The tree competed for water in the surface soil at some critical period as indicated by the spatial variation of monitored soil water content. The tree used deep soil water and reduced deep drainage as indicated by modelling of soil water fluxes and direction of water movement. Trees in the alley cropping system used less soil water than peanut, and more water than trees alone, as indicated by water balance modelling. The magnitude of the reduction of deep drainage increased with tree age. The alley cropping system increased system water use efficiency by using deep soil water as indicated by water balance modelling. The alley cropping system promoted Choerospondias axillaris growth by 50 to 100% and reduced peanut biomass and yield by 20 to 50%. Field management should aim to prevent water competition between trees and crops in the surface soil. (author)

[en] Soil and water management is an essential element in food security, agriculture sector growth and sustainable land management of sub-Saharan Africa (SSA). The increased land degradation and declining fertility of SSA soils contribute to food insecurity and poverty. Previously, agroforestry researchers tended to focus mostly on soil nutrient replenishment as being solely responsible for post-fallow crop yield dynamics. Missing from many studies on soil fertility issues is the recognition of the important role of soil physical properties in agricultural productivity. However, many factors affect soil fertility and some agroforestry measures taken to correct soil nutrient deficiencies can also produce desirable soil physical effects. We hypothesized that planted tree fallows can potentially increase soil N status and improve soil physical properties, thus increasing subsequent crop yields. Field studies were conducted on infertile sandy clay loams at Msekera and Kagoro, Zambia, to determine the effect of contrasting fallows (natural fallow, planted non-coppicing and coppicing tree fallows) and no-tree no-fallow [maize (Zea mays L.) with and without fertilizer] on soil fertility and maize yields. This study attempted to address agricultural productivity by viewing soil fertility in terms of both chemical and physical properties. Hence, this report discusses the implications of improving the nutrient status of soils without correcting soil physical constraints. Data from both tree-(agroforestry) and non-tree-based systems have been used to illustrate important physical and chemical changes that occur in soils as a consequence of varying management regimes or cropping systems. Such data show that the concept of soil productivity refers to more than replacement of the lost nutrients. Other aspects include soil structure, soil water retention, water storage, infiltration and soil penetration resistance. The results imply that standard inputs such as mineral or organic fertilizers can maintain only some elements of soil productivity. Therefore, a broader view that incorporates the role of soil physical properties and water in influencing productivity is appropriate. This research does not attempt to provide a comprehensive treatment of all aspects of soil fertility in agroforestry systems. For example, it does not address the role of soil biota diversity in soil productivity. (author)

[en] Stable isotopes are a powerful tool for examining relationships between plants and their environments. Knowledge gained from studies using isotopes of H, C, N, O and S can inform decisions about management of agroforestry systems - in particular management of water and nutrients. Isotopes of carbon, oxygen and deuterium are now widely used to determine patterns of water use and environmental adaptations. Dual analysis of carbon and oxygen isotopes continues to provide new opportunities to gain a more detailed understanding of variation in ecophysiological processes of plants. For example, coupling a dual isotope approach with other disciplines, particularly dendrochronology, has expanded our ability to examine and compare short- and long-term temporal relationships between plants and their environment. Correlations between carbon and oxygen isotopes of tree rings of Callitris glaucophylla, growing in semi-arid, north-west Australia, reveal an apparent shift in the balance between photosynthesis and stomatal conductance. Between 1920 and 1960, the two isotopes are negatively correlated, while after 1960 they switch rapidly to a positive association. The shift to a positive correlation strongly suggests an increase in stomatal control to regulate water use, possibly a reflection of changes in rainfall patterns in north-west Australia. Detailed modelling of climate-isotope relationships also reveals a number of unusual relationships. Contrary to most studies, we found a significant correlation between δ¹⁸O values and climate of the preceding year. This suggests reduced oxygen re-exchange during cellulose synthesis than reported for other species. Variation in oxygen isotopes largely reflects relative humidity and rainfall early in the growing season, and temperature late in the preceding and current growing season. Again in contrast to other studies, carbon isotopes of tree rings of C. glaucophylla appear less useful than oxygen as a climate proxy, but reveal fundamental information about differences in physiology between conifers and broadleaf species. While carbon and oxygen are commonly used to examine relationships between trees and climate, deuterium isotopes are more commonly used to examine water sourcing by plants. In arid regions, dependence on rainfall or groundwater is largely a reflection of landscape position. For example, species adapted to growing in and along creek lines have access to groundwater and are less dependent on stored soil water. Conversely, species growing on floodplains are highly dependent on rainwater. A difference in access to water translates into differences in physiology and water use: rates of transpiration of creek line species varies little between seasons. These findings have implications for management of mixed systems and their adaptability to long-term changes in rainfall and groundwater supply. (author)

[en] A field experiment was undertaken in western Kenya to (i) determine the amount of legume below ground biomass N at a given time without physically disturbing the soil, (ii) study ¹⁵N recovery by maize from decomposing labelled roots of legume fallows and assess interactions occurring when mixed with the above ground fallow biomass and (iii) determine the distribution of excess ¹⁵N in water stable aggregates, free organic matter and the clay + silt fraction. One year old Sesbania sesban, Calliandra calothyrsus, Senna spectabilis, Cajanus cajan, Crotalaria grahamiana and Tephrosia vogelii were enriched in situ with labelled ammonium sulphate [(¹⁵NH₄)₂SO₄] through stem injection techniques. The total above ground dry matter (DM) of the legumes varied between 589 g tree¹ for Senna and 1572 g tree¹ for Calliandra. The shoot to root ratio ranged from 1.2 for Senna to 3.1 for Calliandra. Total above ground N ranged from 8.6 to 23.1 g tree¹ and was highest for Calliandra and lowest for Senna. Cajanus, Calliandra, Crotalaria and Tephrosia exhibited similar proportions of ¹⁵N enrichment in the leaves and wood, but Senna allocated more ¹⁵N to wood as compared to leaves. Total above ground ¹⁵N recovery ranged from 49% for Crotalaria to 69% for Tephrosia indicating that 27 to 50% is potentially allocated to the roots. About 2.5 to 7.4% of ¹⁵N is potentially recovered from roots less than 5 mm. A large proportion (39%) of below ground N became protected in water stable meso- and macroaggregates while around 20% was associated with the clay-silt sized fraction. This contrasts with the fate of ¹⁵N labelled foliage litter from which only 31% was found in aggregates, but a larger proportion remained in the free organic fraction, suggesting an important role of roots in aggregate formation. ¹⁵N recoveries of below ground N in the catch crop were small suggesting that these inputs play a more important role in soil structure rather than in nutrient supply. (author)

[en] An agroforestry system that included coffee (Coffea arabica) plants associated with two species of legume trees, Erythrina poeppigiana and Gliricidia sepium was established. The experiment included three treatments, coffee without legume trees, coffee and Erythrina and coffee with Gliricidia in two different coffee growing areas of Costa Rica. Shade trees were planted in May 1999 on a 2 year old coffee plantation. Several studies were completed on the experimental areas including N fertilizer efficiency, N leaching through the soil profile, and N cycling from legume prunings. Labelled ¹⁵N was injected into Gliricidia and Erythrina trees in 2003 in order to study the dynamics of N from the legume tree residues. Very low recoveries of N by coffee were obtained from fertilizer sources (3.2 % for urea and <1% for nitrate), and different sinks of plant N were identified depending on the N source. Nitrate fertilizer had the highest leaching potential especially on the Heredia site, and cannot be recommended for coffee. The N₂ fixation capacity of Erythrina was around 50%. With the tree injection technique, the time required for N in the legume residues to be recovered by coffee was 30 d for the initial appearance with a peak around 60 d. The modified system with 500 legume trees ha¹ produced the highest coffee yield and foliar N values. The amount of biomass produced by Erythrina was around 4500 kg ha¹, containing 200 kg N ha¹. Part of that N comes from biological N₂ fixation and part from soil N being recycled by the tree. The Erythrina density of 500 trees ha¹ is recommended as the best system to substitute for N fertilizer (author)