Modelling agroforestry systems of cacao (Theobroma cacao) with laurel (Cordia alliodora) and poro (Erythrina poeppigiana) in Costa Rica III. Cycles of organic matter and nutrients

Models for cycles for organic matter and nutrients element (N, P, K, Ca and Mg) are presented for the agroforestry systems of cacao (Theobroma cacao) withCordia alliodora orErythrina poeppigiana in Turrialba, Costa Rica. For the models, system reserves (soil, humus, vegetation divided into leaves, branches, stems, fine roots, fruits) and transference between compartments (production and decomposition of litter residues) inputs (fertilizer, rainfall) and outputs (harvests) of the system are considered. The implications of the models are discussed in detail. Aspects of net primary production in the systems studied are considered. N fixation is calculated on the basis of balances. Analysis of soil water showed high variations that coincided with rainfall patterns and pruning of theE. poeppigiana. For part I see Vol. 4, No. 3, 1986 For part II see this issue Agroforestry Project, CATIE/GTZ (Tropical Agricultural Research and Training Center/Gesellschaft für Technische Zusammenarbeit), Turrialba, Costa Rica     

Modelling agroforestry systems of cacao (Theobroma cacao) with laurel (Cordia alliodora) and poro (Erythrina poeppigiana) in Costa Rica

The agroforestry systems of cacao (Theobroma cacao) under laurel (Cordia alliodora) and cacao under poro (Erythrina poeppigiana) were studied at CATIE, Turrialba, Costa Rica. An inventory was taken of the organic matter and nutrients (N, P, K, Ca, and Mg) separating the species into their compartments (leaves, branches, trunks and roots). Studies of the litter and of the mineral soil (0–45 cm) yielded these results: Patterns of nutrient accumulation are discussed in relation to the characteristics of these agroforestry systems.     

Litter production and nutrient cycling in coffee (Coffea arabica) or cacao (Theobroma cacao) plantations with shade trees

The relative importance of N fixation, organic material inputs and nutrient inputs in litterfall, as justifications for including shade trees in plantations of coffee or cacao, is discussed. According to existing data, N fixation by leguminous shade trees does not exceed 60 kg.N/ha/a. However, these trees contribute 5,000–10,000 kg. organic material/ha/a.Comparisons are made between the leguminous shade tree Erythrina poeppigiana and the non-leguminous timber tree Cordia alliodora. The former, when pruned 2 or 3 times/a., can return to the litter layer the same amount of nutrients that are applied to coffee plantations via inorganic fertilizers, even at the highest recommended rates for Costa Rica of 270 kg.N, 60 kg.P, 150 kg.K/ha/a. The annual nutrient return in this litterfall represents 90–100 percent of the nutrient store in above-ground biomass of E. poeppigiana, and hence the consequences of competition with the crop should not be a serious limitation. In the case of C. alliodora, which is not pruned, nutrient storage in the tree stems, especially of K, is a potential limiting factor to both crop and tree productivity.It is concluded that, in fertilized plantations of cacao and coffee, litter productivity is a more important shade tree characteristic than N fixation.An early version of this review was presented at the CATIE-IUFRO meeting Los Arboles de Uso Multiple en Sistemas Agroforestales, June 1985, Turrialba, Costa Rica.     

Modelling agroforestry systems of cacao (Theobroma cacao) with laurel (Cordia alliodora) and poro (Erythrina poeppigiana) in Costa Rica II. Cacao and wood production,litter production and decomposition

During 7 years (1979–1985) cacao harvests (beans and husks) have been recorded for the agroforestry systems ofTheobroma cacao underCordia alliodora andErythrina poeppigiana shade trees. The mean oven dry cacao yields were 626 and 712 kg.ha^–1.a^–1 cocoa beans underC. alliodora andE. poeppigiana respectively. Harvests have gradually increased over the years and the plantation has now reached maturity.Annual extraction of N, P, K, Ca and Mg in fruits, which is relatively small, was calculated on the basis of chemical analyses. The following average values were found (kg.ha^–1.a^–1): At the age of 8 years, theC. alliodora trees have reached 26.7 cm diameter (DBH) and 14.0 m in height. Mean annual growth (from age 5 to 7) is 14.6 m³.ha^–1.a^–1.Natural plant residue production has been measured for 4 years (Nov. 1981–Oct. 1985). UnderE. poeppigiana it has reached a value of 8.91 t.ha^–1.a^–1 and underC. alliodora 7.07 t.ha^–1.a^–1. The shade trees have contributed 57 and 47% respectively. Transference and decomposition rates are high and important in the nutrient cycles.The nutrient content of the litter was analysed and corresponding average yearly transfers were (kg.ha^–1.a^–1): For part I see Vol. 4, No. 3, 1986.Agroforestry Project, CATIE/GTZ (Tropical Agricultural Research and Training Center/Gesselschaft für Technische Zusammenarbeit), Turrialba, Costa Rica     

Modelling agroforestry systems of cacao (Theobroma cacao) with laurel (Cordia alliodora) and poro (Erythrina poeppigiana) in Costa Rica II. Cacao and wood production,litter production and decomposition

During 7 years (1979–1985) cacao harvests (beans and husks) have been recorded for the agroforestry systems ofTheobroma cacao underCordia alliodora andErythrina poeppigiana shade trees. The mean oven dry cacao yields were 626 and 712 kg.ha⁻¹.a⁻¹ cocoa beans underC. alliodora andE. poeppigiana respectively. Harvests have gradually increased over the years and the plantation has now reached maturity. Annual extraction of N, P, K, Ca and Mg in fruits, which is relatively small, was calculated on the basis of chemical analyses. The following average values were found (kg.ha⁻¹.a⁻¹): At the age of 8 years, theC. alliodora trees have reached 26.7 cm diameter (DBH) and 14.0 m in height. Mean annual growth (from age 5 to 7) is 14.6 m³.ha⁻¹.a⁻¹. Natural plant residue production has been measured for 4 years (Nov. 1981–Oct. 1985). UnderE. poeppigiana it has reached a value of 8.91 t.ha⁻¹.a⁻¹ and underC. alliodora 7.07 t.ha⁻¹.a⁻¹. The shade trees have contributed 57 and 47% respectively. Transference and decomposition rates are high and important in the nutrient cycles. The nutrient content of the litter was analysed and corresponding average yearly transfers were (kg.ha⁻¹.a⁻¹): For part I see Vol. 4, No. 3, 1986. Agroforestry Project, CATIE/GTZ (Tropical Agricultural Research and Training Center/Gesselschaft für Technische Zusammenarbeit), Turrialba, Costa Rica     

Fine root dynamics of shaded cacao plantations in Costa Rica

Root turnover may contribute a significant proportion of recycled nutrients in agroforestry systems and competition between trees and crops for nutrients and water may depend on temporal fine root regrowth patterns. Fine root biomass ( 2 mm) and fine root productivity were measured during one year in plantations of cacao (Theobroma cacao) shaded by Erythrina poeppigiana or Cordia alliodora planted on a deep alluvial soil in Turrialba, Costa Rica. Fine root biomass of approximately 1.0 Mg ha^–1 varied little during the year with maximum values at the beginning of the rainy season of 1.85 Mg ha^–1 in the cacao-C. alliodora system compared to 1.20 Mg ha^–1 for cacao-E. poeppigiana. Fine root productivity of C. alliodora and E. poeppigiana (maximum of 205 and 120 kg ha^–1 4 week^–1, respectively) was greatest at the end of the rainy season, while for cacao it was greatest at the beginning of the rainy season (34–68 kg ha^–1 4 week^–1), which suggests that if nutrient competition occurs between the shade trees and the cacao, it could be minimized by early fertilization during the beginning of the rains immediately after pruning the shade trees. Annual fine root turnover was close to 1.0 in both systems. Assuming that fine root biomass in these mature plantations was constant on an annual basis, nutrient inputs from fine root turnover were estimated as 23–24 (N), 2 (P), 14–16 (K), 7–11 (Ca) and 3–10 (Mg) kg ha^–1 year^–1, representing 6–13% and 3–6% of total nutrient input in organic matter in the C. alliodora and E. poeppigiana systems, respectively.     

Decomposition of Erythrina poeppigiana leaves in 3-, 9-, and 18-year-old alleycropping systems in Costa Rica

Timing the application of organic residues and therefore the release of nutrients during decomposition may be critical to the growing crop in tropical alleycropping agroforestry systems. Field experiments were carried out in Turrialba, Costa Rica, to determine differences in Erythrina poeppigiana (Walp.) O.F. Cook leaf decomposition in 3, 9 and 18-year alleycropped agroforestry systems. Treatments consisted of mulch-only, and mulch plus Arachis pintoi Krapov. and W. Gregory var. CIAT 18347 in 3 and 9-year old alleycrops under no-till cultivation. The 18-year old site consisted of treatments with mulch-only and mulch plus chicken manure under disk plow cultivation. Litterbags, filled with E. poeppigiana leaves from 3, 9 and 18-year old trees, were placed on the soil surface and collected over a period of 84 days. Results showed no significant differences in the amount of plant residues remaining after 84 days in the 3-, 9-, and 18-year-old systems, or between the manure and mulch-only treatments. Comparing mulch-only treatments, leaves in the 18-year old system decomposed most rapidly which may be due to disk-plow cultivation practices where litterbags were in direct contact with the soil as opposed to the no-till system in the younger alleycrops.     

Effect of pollarding frequency on biomass of Erythrina poeppigiana as a coffee shade tree

The use of pollarded Erythrina poeppigiana as shade tree in coffee plantations is apparently an old practice in Costa Rica. The tree is not native to this country but was introduced between late 19th and 20th century and was rapidly dispersed in the coffee and cacao areas. Currently, the Erythrina tree is widespread in the Turrialba Valley (elevation 600—1300m) and in the Central Valley (elevation 1200m) where the species is always associated with present or past coffee crops. Pollarding carried out by Costa Rican farmers constitutes a long dated and functional practice, hence the objective of this study was to evaluate the amount of biomass produced by pollarding of Erythrina poeppigiana used as shade in coffee crop planted at a density of 280 trees/hectare under different pollarding frequencies. Results showed that by pollarding once a year, 18,470 kg of dry matter per hectare are produced; with two pollardings per year 11,800 kg/ha are produced and with three pollarding per year 7,850 kg/ha are produced. The total amount of nitrogen removed is very similar for pollarding once and twice a year, but is lower for three times a year. The amount of nitrogen removed was approximately 230 kg/ha/year in the first two cases and 170 kg/ha/year in the last one.The above observations suggest that a considerable supply of nutrients exist in the systems with shade trees, when they are periodically pollarded.Finally some conclusions and follow up activities related to research on the species are suggested, such as higher biomass production techniques, appropriate planting practices, selection of genetic material, nutrient depletion when biomass is harvested, conversion of leaves to marketable feed sources (flour, pellets), alley cropping and green manure production and restoration of degraded areas and improductive savannas by planting large cuttings that would improve the soil by adding biomass and shade out undesirable grasses.This work is part of a Ph.D. Dissertation submitted to the Southeastern University, New Orleans, Lousiana by R.O. Russo.     

Modelling agroforestry systems of cacao (Theobroma cacao) with laurel (Cordia alliodora) or poro (Erythrina poeppigiana) in Costa Rica

Predictive models were developed for Cordia alliodora branch and Theobroma cacao branch or leaf biomass,based on branch basal areas (r² 0.79) but the model of C. alliodora leaf biomass, although significant, was of very low accuracy (r² = 0.09) due to annual leaf fall. At age 10 years, shade tree stem biomass accounted for 80% of the total above-ground biomass of either tree. However, between the ages of 6 and 10 years, the biomass increment of T. cacao branches (3–4t.ha^–1.a^–1) was similar to that of the shade tree stems. During the same period, the net primary productivity was 35 and 28 t.ha^–1.a^–1, for the Erythrina poepigiana and and C. alliodora systems, respectively.Cocoa production under either of the shade trees C. alliodora or E. poeppigiana was 1000 kg.ha^–1.a^–1 (oven-dry; ages 6–10 yr). During the same period, C. alliodora timber production was 9 m³.ha^–1.a^–1 whilst the leguminous shade tree E. poeppigiana does not produce timber. Litterfall over the same 5 years, including crop and/or shade tree pruning residues, averages 11 and 23 t.ha^–1.a^–1, respectively. The main difference was due to E. poeppigiana pruning residues (10t.ha^–1.a^–1).Soil organic material reserves (0–45 cm) increased over 10 years from 198 to 240 t.ha^–1 in the E. poeppigiana plots and from 168–184 t.ha^–1 in the C. alliodora plots. These values, together with the productivity indices presented, provide evidence that the systems are sustainable.For economic reasons, the use of C. alliodora is recommended under the experimental conditions. however, on less fertile soils without fertilization, the greater biomass and hence nutrient return to the soil surface under E. poeppigiana, might make this the preferable shade tree.     

Application of the pipe model theory to non-destructive estimation of leaf biomass and leaf area of pruned agroforestry trees

The relationship of branch cross sectional area (CS) to leaf biomass (LM) and leaf area (LA) was studied in three agroforestry tree species,Calliandra calothyrsus Maissn.,Erythrina berteroana Urban andErythrina poeppigiana (Walpers) O.F. Cook, to develop a non-destructive method for the estimation of LM and LA for trees managed with periodic pruning. Variation in these relationships was observed according to the bifurcation level and, in theErythrina spp., by clone. All the relationships were linear except the CS-LM relation in small branches ofE. poeppigiana, where it was initially exponential. At main branch level the relationship of CS to LM and LA was linear in all cases but the regression parameter values varied between species and clones, with determination coefficient (R²) 0.88–0.99. It was concluded that the ratio of main branch CS to LM and LA can be used for non-destructive estimation of the latter variables. The method has the additional benefit that the regression parameter value reflects the allocation of dry matter within a tree and, consequently, may give indications about its possible uses in different agroforestry systems.Work carried out at the Centro Agronómico Tropical de Investigación y Enseñanza (CATIE), Turrialba, Costa Rica.     

Improving rumen fermentation and milk production with legume-tree fodder in the tropics

Feeding leguminous fodder that is high in protein can improve rumen fermentation parameters leading to increased digestibility and intake of low quality fibrous feeds, and hence improved animal production. The effect on dry matter (DM) degradation and rumen fermentation parameters of feeding urea or tree legume foliage (Erythrina poeppigiana or Gliricidia sepium), as protein supplements to a basal diet of Hyparrhenia rufa hay, rice bran and molasses, was studied with rumen fistulated steers (Zebu × Creole crossbreeds). In a second experiment, Jersey × Creole crossbreeds were used to determine the effect of the above three treatments on milk production. Both experiments were conducted at CATIE's experimental farm, Turrialba, Costa Rica (9°58 N; 83°31 W; 639 m above sea level; mean annual precipitation of 2600 mm; mean daily temperature 22 °C). The disappearance of hay DM in the rumen increased with time to 144 hours and potential degradation of hay DM was significantly higher (P < 0.05) for the legume diets compared to urea but there were no differences between legume species. Legume based rations resulted in total volatile fatty acid concentrations of rumen licor that were 15 to 24% higher than for urea based rations (P < 0.05). The concentration of rumen NH₃-N was highest two hours after feeding and tended to decrease in time. The urea ration resulted in significantly higher NH₃-N than the legume rations, reaching 45 mg/100 mL two hours after feeding. Milk yields did not differ (P > 0.05) when cows received either E. poeppigiana or G. sepium foliages as protein supplements (7.3 and 7.4 kg milk/cow/day, respectively) but they were superior (P < 0.05) to urea supplementation (6.7 kg milk/cow/day).     

Alley cropping and mulching withErythrina poeppigiana (Walp.) O. F. Cook andGliricidia sepium (Jacq.) Walp.: effects on maize/weed competition

The potential of allye cropping systems to sustain a high productivity with low external inputs and the reduction of maize/weed competition through weed suppression in different alley cropping and sole-cropped mulched systems was studied in Costa Rica at CATIE. Data were recorded eight years after establishment of the experiment. Plant residues ofErythrina poeppigiana trees (10 t/ha dry matter) planted at 6 by 3 m reduced weed biomass by 52%, whileGliricidia sepium trees (12 t/ha dry matter) planted at 6 by 0.5 m reduced weed biomass by 28%, in comparison to controls.Erythrina had a considerable impact on grass weeds, whileGliricidia reduced the incidence of some dicot weeds. Weed competition significantly reduced maize yield in all systems. Nevertheless weed suppression contributed to the higher maize grain yield underErythrina andGliricidia alley cropping of 3.8 t per hectare as opposed to the unmulched control yield of 2.0 t per hectare.     

Soil characteristics below <Emphasis Type="Italic">Erythrina poeppigiana</Emphasis> in organic and conventional Costa Rican coffee plantations

The impact of Erythrina poeppigiana on soil characteristics, at three different positions relative to the shade tree and from three different soil depths, was evaluated in pairs of comparable Costa Rican coffee farms (organic and conventional) in 2000 and 2004. In the conventional system at 0–5 cm, higher C and N concentrations were found close to the shade tree versus the positions 2 m from the trunk (5.04 vs. 4.18%). This positive effect could influence only 20% of the farm area when high population of E. poeppigiana were used. This finding highlighted the importance of E. poeppigiana in maintaining SOM levels. In contrast, the organic system showed similar C and N concentrations for all positions probably due to an even distribution of pruning residues and to the use of organic amendments. A trend to higher total C and N concentrations for organic farms in comparison to conventional farms was found. No significant temporal changes in soil C or N concentrations were found between 2000 and 2004.     

Hopper (Homoptera: Auchenorrhyncha) diversity in shaded coffee systems of Turrialba, Costa Rica

Diversity of hopper species (suborder Auchenorryncha) in coffee (Coffea arabica) plantations with no shade (C) was compared with the diversity in plantations with shade of either poró (Erythrina poeppigiana) (CP) or poró plus laurel (Cordia alliodora) (CPL) in Turrialba, Costa Rica. Species-abundance and rarefaction curves were plotted for each system, and indices of diversity (Shannon-Wiener), dominance (Simpson), species evenness, and similarity (Jaccard) were calculated. The majority of hopper species and individuals belonged to the Cicadellidae family. A particular species dominated in each system: Graphocephala sp. (C), Fusigonalia lativittata (CP) and Hebralebra nicaraguensis (CPL). The richness and diversity of hopper species were highest in the CP system, followed by the CPL and C systems. Species similarity was closest between the CP and CPL systems, but varied considerably according to plant component and geographic location of each plot. Even though hoppers have not been reported as coffee pests in Mesoamerica, some of them cause serious problems elsewhere. This revised version was published online in June 2006 with corrections to the Cover Date.     

Transpiration of Arabica Coffee and Associated Shade Tree Species in Sub-optimal, Low-altitude Conditions of Costa Rica

Sap flows of coffee (Coffea arabica L. cv ‘Costa Rica 95’) and associated timber trees (Eucalyptus deglupta or Terminalia ivorensis) or leguminous tree (Erythrina poeppigiana) were measured simultaneously during 12 months in 4-year-old coffee agroforestry systems in sub-optimal ecological conditions of Costa Rica. In the wet period, coffee and shade tree transpiration followed the daily patterns of photosynthetic photon flux density (PPFD) and reference evapotranspiration (ETo) while their transpiration was restricted at higher air VPD values (>1.5 kPa) registered during the dry period. Coffee transpired more per unit leaf area in full sun than under shade, an indication of higher environmental coffee stress in non shaded conditions. Nonetheless, coffee daily water consumption per hectare was generally higher under shade than in full sun due higher vegetative growth of shade-grown coffee plants. Minimum and maximum daily transpiration were 0.74 and 4.08 mm for coffee, 0.35 and 1.06 mm for E. deglupta, 0.70 and 2.10 mm for T. ivorensis and 0.13 and 0.79 mm for E. poeppigiana. Estimation of the annual combined water transpiration by coffee and shade trees was 20–250% higher than that of coffee grown in full sun. Nevertheless, there was no evidence that water use by associated trees decreased soil water availability for coffee and hence limited coffee transpiration in the dry season due to its relatively short length (3 months) and the high annual rainfall (over 3100 mm). In the sub-optimal, low altitude conditions of this experiment, E. deglupta was the optimum shade species as it maintained a more constant shade level throughout the year and ensured a better protection to coffee underneath than T. ivorensis and E. poeppigiana which underwent a complete defoliation during the adverse meteorological conditions of the dry period.     

Modelling agroforestry systems of cacao (Theobroma cacao) with laurel (Cordia alliodora) and cacao with poro (Erythrina poeppigiana) in Costa Rica

The hydrological balances for the agroforestry systems of Theobroma cacao with Cordia alliodora or Erythrina poeppigiana were calculated using measurements taken over four years (1983–1987) of the following parameters: climatic (precipitation, evaporation); edaphological (texture, soil moisture tension, hydraulic conductivity, variations in soil moisture content). Water fluxes (interception, simulation of transpiration and percolation) were estimated from one year of measurements (1986–1987). Weekly samples of percolated water were taken from March 1986 – March 1987 using lysimetric capsules at 100 cm soil depth. N, P, K, Ca and Mg concentrations were determined to calculate the annual leaching losses. Nutrient concentration values were extrapolated for the whole 4 years observation period, in order to calculate leaching losses for the whole study period. Precipitation samples were also taken to determine the corresponding annual nutrient inputs.     

Soil carbon dynamics and residue stabilization in a Costa Rican and southern Canadian alley cropping system

Agroforestry systems can play a major role in the sequestration of carbon (C) because of their higher input of organic material to the soil compared to sole crop agroecosystems. This study quantified C input in a 19-year old tropical alley cropping system with E. poeppigiana (Walp.) O.F Cook in Costa Rica and in a 13-year old hybrid poplar (Populus deltoides × nigra DN-177) alley cropping system in southern Canada. Changes in the level of the soil organic carbon (SOC) pool, residue decomposition rate, residue stabilization efficiency, and the annual rate of accumulation of SOC were also quantified in both systems. Carbon input from tree prunings in Costa Rica was 401 g C m⁻² y⁻¹ compared to 117 g C m⁻² y⁻¹ from litterfall at the Canadian site. In southern Canada, crop residue input from maize (Zea mays L.) was 212 g C m⁻² y⁻¹, 83 g C m⁻² y⁻¹ from soybeans (Glycine max L.) and 125 g C m⁻² y⁻¹ for wheat (Triticum aestivum L.), and was not significantly different (p < 0.05) from the sole crop. The average yearly C input from crop residues in Costa Rica was significantly greater (p < 0.05) in the alley crop for maize (134 g C m⁻² y⁻¹) and Phaseolus vulgaris L. bean crops (35 g C m⁻² y⁻¹) compared to the sole crop. The SOC pool was significantly greater (p < 0.05) in the Costa Rican alley crop (9536 g m⁻²) compared to its respective sole crop (6143 g m⁻²) to a 20 cm depth, but no such difference was found for the southern Canadian system. Residue stabilization, defined as the efficiency of the stabilization of added residue (crop residues, tree prunings, litterfall) that is added to the soil C pool, is more efficient in southern Canada (31%) compared to the alley cropping system in Costa Rica (40%). This coincides with a lower organic matter decomposition rate (0.03 y⁻¹) to a 20 cm depth in Canada compared to the Costa Rican system (0.06 y⁻¹). However, the average annual accumulation rate of SOC is greater in Costa Rica (179 g m⁻² y⁻¹) and is likely related to the greater input of organic material derived from tree prunings, compared to that in southern Canada (30 g m⁻² y⁻¹) to a 20 cm depth.     

Nutrient cycling in two traditional Central American agroforestry systems

A preliminary nutrient cycling study quantified total and temporal nutrient inputs via litterfall and pruning residues in two agroforestry systems: (1) Coffea arabica (perennial crop)-Erythrina poeppigiana (leguminous shade tree); and (2) C. arabica-E. poeppigiana-Cordia alliodora with emphasis on the effect of the timber tree C. alliodora. The total annual input of litterfall plus pruning residues was similar in both associations. Total annual input from E. poeppigiana was less than half in the association with C. alliodora than without, but the litterfall from this latter species compensated for the loss. Large differences in the total annual nutrient input of K, Ca and Mg was found between associations, but not for N or P. The amount of nutrients recycled by the associated trees reached the recommended level of fertilizer required for coffee production. The inclusion of C. alliodora within the C. arabica-E. poeppigiana association resulted in a more evenly distributed annual nutrient input.     

Dynamics of non-structural carbohydrate reserves in pruned <Emphasis Type="Italic">Erythrina poeppigiana</Emphasis> and <Emphasis Type="Italic">Gliricidia sepium</Emphasis> trees

In alley cropping systems, fast growing leguminous trees are pruned to reduce competition with crops for light and to provide organic inputs for crop nutrition. Tree regrowth depends on non-structural carbohydrate reserves in the remaining tree parts. In this study, the dynamics of starch and soluble carbohydrates in roots and stems of completely pruned (all shoots removed), partially pruned (one branch retained on the pruned stump) and unpruned Erythrina poeppigiana (Walp.) O.F. Cook and Gliricidia sepium (Jacq.) Kunth ex Walp. trees were studied under humid tropical conditions in Turrialba, Costa Rica. Measurements on starch and soluble carbohydrates in roots and stems were made at 0, 2, 6 and 12 weeks after pruning during both a “rainy” and a “dry” season. In general, the dynamics of non-structural carbohydrates in roots and stems of pruned E. poeppigiana and G. sepium trees were similar. Starch concentration was highest in unpruned trees and higher in roots than in stems of pruned trees. The effect of pruning intensity was first observed in stems, and starch reserves were more depleted in stems than in roots, an effect more evident during the “dry” season. The critical tree regrowth stage for starch mobilisation was that of vigorous sprout development at six or four weeks after pruning particularly in completely pruned trees. At this time, fine root biomass and length and nodule biomass in pruned trees decreased. Survival of fine roots and nodules was greater in partially pruned than in completely pruned trees. Starch accumulation in roots recommenced at 12 weeks after pruning in G. sepium, and later than 12 weeks after pruning in E. poeppigiana roots. This study showed that E. poeppigiana responded better to pruning regimes than G. sepium. Recovery of trees after pruning is better when trees are partially pruned than when completely pruned.     

Natural abundance of <Superscript>15</Superscript>N in two cacao plantations with legume and non-legume shade trees

Natural abundance of ¹⁵N was sampled in young and mature leaves, branches, stem, and coarse roots of trees in a cacao (Theobroma cacao) plantation shaded by legume tree Inga edulis and scattered non-legumes, in a cacao plantation with mixed-species shade (legume Gliricidia sepium and several non-legumes), and in a tree hedgerow bordering the plantations in Guácimo, in the humid Caribbean lowlands of Costa Rica. The deviation of the sample ¹⁵N proportion from that of atmosphere (δ¹⁵N) was similar in non-legumes Cordia alliodora, Posoqueria latifolia, Rollinia pittieri, and T. cacao. Deep-rooted Hieronyma alchorneoides had lower δ¹⁵N than other non-N₂-fixers, which probably reflected uptake from a partially different soil N pool. Gliricidia sepium had low δ¹⁵N. Inga edulis had high δ¹⁵N in leaves and branches but low in stem and coarse roots. The percentage of N fixed from atmosphere out of total tree N (%N_f) in G. sepium varied 56–74%; N₂ fixation was more active in July (the rainiest season) than in March (the relatively dry season). The variation of δ¹⁵N between organs in I. edulis was probably associated to ¹⁵N fractionation in leaves. Stem and coarse root δ¹⁵N was assumed to reflect the actual ratio of N₂ fixation to soil N uptake; stem-based estimates of %N_f in I. edulis were 48–63%. Theobroma cacao below I. edulis had lower δ¹⁵N than T. cacao below mixed-species shade, which may indicate direct N transfer from I. edulis to T. cacao but results so far were inconclusive. Further research should address the ¹⁵N fractionation in the studied species for improving the accuracy of the N transfer estimates. The δ¹⁵N appeared to vary according to ecophysiological characteristics of the trees.