Published online 19 September 2005
Published in Agron J 97:1396-1402 (2005)
DOI: 10.2134/agronj2004.0250
© 2005 American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
Nutrient Cycling and Uptake
Foliage Residue Decomposition and Nutrient Release in Peach Palm (Bactris gasipaes Kunth) Plantations for Heart-of-Palm Production in Costa Rica
G. Sotoa,
P. Luna-Oreab,
M. G. Waggerb,*,
T. J. Smythb and
A. Alvaradoa
a Cent. for Agron. Investigations, Univ. of Costa Rica, San José, Costa Rica
b Dep. of Soil Sci., North Carolina State Univ., Raleigh, NC 27695
* Corresponding author (michael_wagger{at}ncsu.edu)
Received for publication October 2, 2004.
 |
ABSTRACT
|
|---|
Peach palm (Bactris gasipaes Kunth) for heart-of-palm production exports relatively low amounts of nutrients from the field and leaves considerable amounts of residue and nutrients on the ground as mulch. The primary objective of this study was to gain an understanding of residue decomposition and nutrient release patterns of peach palm foliage in a mature plantation in Costa Rica. The study was conducted within a 16-yr commercial peach palm stand during two typically seasonal wet periods and one typically dry period. The third leaf of the five leaves cut during harvest of the stem portion were placed in 1-mm mesh nylon bags. Bags were placed on the soil surface and retrieved at 1, 2, 4, 8, 16, 24, 32, 40, and 48 wk. Initial residue N concentrations over the three placement periods ranged from 24.2 to 28.1 g kg–1, C to N ratios between 16:1 and 17:1, cellulose from 251 to 325 g kg–1, and lignin from 80 to 104 g kg–1. There was no effect of seasonal periods on residue decomposition and N, P, and K release. Residue decomposition and nutrient release were best fitted by single-exponential, three-parameter models. The residue decomposition rate was 0.1472 wk–1 while nutrient release rates ranged from 0.0297 to 0.2998 wk–1. The potentially available nutrient pools in 4- and 8-yr peach palm stands from a companion experiment ranged on an annual basis from 93 to 107 kg N ha–1, 14 to 15 kg P ha–1, 90 to 116 kg K ha–1, 19 to 23 kg Ca ha–1, and 13 to 14 kg Mg ha–1. The relatively rapid decomposition and nutrient release rates would seem to be ideal for this perennial cropping system where plants are continuously absorbing nutrients during the year to support the growth of offshoots that eventually become harvestable stems.
|
INTRODUCTION
|
|---|
PEACH PALM, a native plant from the humid tropics of Latin America, has been used extensively for centuries by natives of this region (Clement, 1986). It became an attractive perennial cash crop for farmers four decades ago when local and international markets opened for heart-of-palm (Mora Urpí et al., 1997; Deenik et al., 2000). Heart-of-palm is extracted from several palm genera and species; however, those from the genus Euterpe are preferred because they are abundant, palatable, highly productive, and they lack toxic components and grow fast (Bovi, 1997). The major country exporters in order of importance are Brazil, Costa Rica, Colombia, Venezuela, and Peru (Quintana de la Flor et al., 1993). In Costa Rica, peach palm used for heart-of-palm production is the second most important economic crop, covering an area of 12000 ha. The total area in Central and South America is 24000 ha (Mora-Urpí et al., 1997). Recently, the demand for heart-of-palm has increased in Europe and USA (Clement et al., 1996), leading to substantial export increases and income receipts. Peach palm is intensively cultivated in Oxisols, Ultisols, and Andisols in Central and South America, with pH as low as 4.5, high aluminum saturation, and high organic matter contents. It is best suited to well-drained areas with rainfall between 2000 and 3000 mm and elevations between 5 and 700 m above sea level (Clement, 1989). Peach palm for heart-of-palm is planted at spacing of 2 by 0.5, 2 by 1, or 2 by 1.5 m (Herrera, 1989; Clement et al., 1996) and starts producing heart-of-palm at about 1 to 1.5 yr after transplanting (Ares et al., 2002; Clement et al., 1996). A plant unit contains about four stems and six to eight suckers (Deenik et al., 2000). On average, every plant produces 1.5 heart-of-palm per year in Brazil compared with three hearts in Costa Rica (Deenik et al., 2000). In Hawaii, a 1.3-yr-old stand with 5000 plants ha–1 produced from 2 to 7 Mg dry biomass ha–1 (Clement, 1995), whereas in Costa Rica, 3- to 5-yr-old plantations produced from 7.1 to 11.4 Mg standing dry biomass ha–1 (Herrera, 1989).
Peach palm for heart-of-palm, interestingly, exports only a small portion of the aboveground biomass and accompanying nutrients from the field. A mature plantation of heart-of-palm in Costa Rica exported 1.76 Mg ha–1 yr–1 of dry biomass that contained 28 kg of N, 4.8 kg of P, and 31 kg of K (Herrera, 1989). In Brazil, 2.6 Mg dry biomass ha–1 was removed during harvest, containing 32.3 kg of N, 6.4 kg of P, and 45.2 kg of K (Bovi, 1998). The remaining biomass can be recycled, protecting the soil, returning nutrients, and building soil organic matter (McGrath, 1998). In a mature peach palm plantation, 91 to 97 Mg C ha–1 was calculated for the top 20 cm of soil (Ares et al., 2002).
Despite the fact that research on peach palm used for heart-of-palm production has been conducted over the last 20 yr, there is still a need for experimentally based fertilizer recommendations (Deenik et al., 2000). In Sao Paulo, Brazil, fertilizer applications per hectare per year range from 110 to 300 kg of N, 0 to 35 kg of P, 17 to 133 kg of K, 20 to 50 kg of S, and 1 to 2 kg of B (Bovi and Cantarella, 1997; Bovi, 1998). Common fertilization rates in Costa Rica per hectare per year are 300 kg of N, 82 kg of P, and 250 kg of K; however, in some commercial plantations, up to 500 kg N ha–1 yr–1 is applied (Herrera, 1989). Fertilizer use has also been reported to affect soil chemical properties in this cropping system. In Costa Rica, N fertilization at 400 kg ha–1 yr–1 lowered pH from 5.4 to 4.5 and increased exchangeable acidity from 0.5 to 2 cmol kg–1 soil (Herrera, 1989). In Central Amazon under fertilized peach palm trees, elevated soil nutrient concentrations were accompanied by increased concentrations of Al and low pH values (Schroth et al., 2000). Soil degradation is considered minor, as the products harvested represent only a small portion of the system's total organic matter and nutrient stores (Jordan, 1988).
The aforementioned studies illustrate the potential benefits derived from the recycling of biomass and nutrients in the heart-of-palm production system, along with heart-of-palm's potential for a sustainable perennial cropping system for the humid tropics. To ensure greater nutrient use efficiency in this cropping system, the objective of the present study was to gain an understanding of residue decomposition and nutrient release patterns of peach palm foliage left in the field after harvest under a tropical, wet environment in a mature peach palm plantation at Guápiles, Costa Rica. A secondary objective was to use the equations developed in Objective 1 to predict the cumulative annual release of nutrients based on data collected in a companion study.
|
MATERIALS AND METHODS
|
|---|
The study was conducted at Guápiles, Costa Rica. This site represents a tropical, wet climate (Af) according to the Köppen classification system (Pidwirny, 2002). Its elevation was 210 m, with a mean annual temperature of 25.5°C and rainfall of 4577 mm. A distinct dry season from February to April was considered to be a major factor governing residue decomposition, and thus the study was designed to cover two "wet" rainfall periods starting in July (W1) and September (W2) and the "dry" period between February and April (D) Rainfall during the dry period was historically characterized as receiving less than 100 mm mo–1. The soil was classified as a Typic Hapludands (cindery, amorphous, isohyperthermic). Selected soil chemical characteristics of the top 0 to 20 cm were as follows: pH, 5.7 (1:2.5, soil:water); effective cation exchange capacity, 12.35 cmol (+) L–1 (Ca, Mg, and exchangeable acidity extracted with 1 M KCl; K extracted by modified Olsen); P, 55 mg L–1 (modified Olsen); and organic matter, 84.2 g kg–1 (wet digestion).
A 16-yr-old commercial peach palm plantation with 5000 plants ha–1 (1 by 2 m) was used to evaluate our main objective. Plants generally are clumps of four, 2- to 3-m-tall stems and six to eight smaller stems (offshoots of the larger stems), which eventually substitute for the harvested stems. Stems were harvested when they attained 9-cm diam. at 10 cm above the ground level. The standard fertilization of the plantation, on an annual basis, consisted of equal bimonthly applications banded near the base of the plant, totaling 250 to 300 kg N ha–1, 22 to 44 kg P ha–1, 42 to 167 kg K ha–1, 10 to 36 kg Mg ha–1, 40 to 80 kg S ha–1, and 500 to 2000 kg calcitic lime ha–1.
To monitor residue decomposition during each seasonal period described above, the third leaf was selected to represent the average dry weight and nutrient concentration of the five leaves cut during the harvest of the stem portion for heart-of-palm. Leaves were air-dried on greenhouse benches, and then one complete leaf was bent and placed in a 1-mm mesh nylon bag (90 by 45 cm). The corresponding residue rate on a dry weight basis was 4.0, 3.0, and 2.7 Mg ha–1 for placement periods W1, W2, and D, respectively. Each period defined had nine sampling times (1, 2, 4, 8, 16, 24, 32, 40, and 48 wk after field placement) and four replicate bags were collected and analyzed at each sampling time. The mesh bags were placed on the soil surface, 2 m apart in the intra-row area on 31 July (W1), and 24 September (W2) 1998, and 11 March 1999 (D). This intra-row placement was approximately 1 m away from the bi-monthly fertilization regime previously described. Residues added during subsequent heart-of-palm harvests partially or completely covered the bags. At each sampling time, the bag contents were dried at 70°C, weighed, and analyzed for N, P, K, Ca, and Mg concentrations. Residue-remaining samples were digested by wet ashing with HNO3 and HClO4 for determination of P, K, Ca, and Mg. Phosphorus was determined by spectrophotometry after stanous chloride reduction of the phospho-molybdate complex (Chapman and Pratt, 1973). Potassium, Ca, and Mg were determined by atomic absorption spectrophotometry. Samples for N were digested by a modified Kjeldahl procedure, and the NH4 in the digest was determined with a Lachat flow injection analyzer. Tissue weight remaining at any sampling date is reported on an ash-free basis.
In conjunction with the leaves obtained for mesh bags, 10 additional leaf samples were collected for each seasonal period and analyzed for N, P, K, Ca, and Mg in the same manner as previously described. Cellulose and lignin plant fractions were determined using the procedure of Van Soest and Wine (1968).
Percentages of the original dry weight, N, P, K, Ca, and Mg remaining at each retrieval date were regressed on time using nonlinear regression procedures of the Statistical Analysis System (SAS Inst., 1998). The decreases in tissue dry weight and nutrient contents with time generally followed an exponential function; thus, the following exponential models were fit to each response variable for each wet and dry period: (i) single two-term model, (ii) double four-term model, and (iii) single three-term model. A single exponential two-term model (Eq. [1]) assumes that all of the fresh foliage tissue or nutrient addition will decompose or release at the same rate from the beginning to the end while the double exponential four-term model (Eq. [2]) considers that the fresh foliage material or nutrient addition is decomposed or released at two different rates. The single exponential three-term model (Eq. [3]) assumes that a fraction of foliage tissue or nutrient addition does not decompose or release during the time the study was conducted. The general form of the equations is as follows:
 | [1] |
 | [2] |
 | [3] |
where Y is the percentage dry weight or nutrient remaining at time x (in weeks), ß0 and ß1 represent the tissue dry weight or nutrient pool, k is the decomposition or nutrient release rate constant, and 
is the random error.
The criteria for model selection were based on the lowest mean square error values. For the purpose of model comparison among the three seasonal placement periods, reduced models were compared to full models only when seasons were fitted by the same type of model. A reduced model would fit data from more than one designated season, whereas a full model would only fit data from a single season. The significance of the increment in the error sum of squares caused by the parameter reduction in the reduced models was tested by an approximate F test (Ratkowsky, 1983). The hypothesis that the curves for a particular variable under study were the same regardless of the season was
 |
The residue nutrient release models obtained in this study were also used to predict cumulative release of nutrients for the initial 48 wk of foliage dry matter and nutrient content data collected in a companion study conducted in the same region, during harvest of offshoots at 4-wk intervals in 4- and 8-yr commercial heart-of-palm stands. Measurements in each peach palm stand were taken in six 10-m-long by 2-m-wide plots. Both stands had similar cumulative numbers of harvested offshoots and biomass during the study period, as well as developing offshoot biomass.
|
RESULTS AND DISCUSSION
|
|---|
The three selected season placements of litterbags did not always correspond with the expected rainfall. Periods W1 and W2 were intended to capture peak rainfall activity during the subsequent 3 to 6 mo after placement, but erratic rainfall distribution occurred (Fig. 1)
. The dry period (D) was absent, and ultimately, total rainfall received during this period was the greatest of the three periods. Moreover, continuous additions of fresh foliage tissue in an environment with high relative humidity should provide microorganisms with nutrients and sufficient moisture to proceed with foliage residue decomposition regardless of the designated seasonal periods.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 1. Rainfall distribution at Guápiles, Costa Rica during each of the three periods (W1, W2, and D) of litterbag placement. Values in parentheses represent total rainfall for the respective periods.
|
|
Chemical Composition of Leaf Dry Matter
Chemical characteristics of the third leaf at the beginning of each season averaged 25.5 g N kg–1, 2.0 g P kg–1, 12.9 g K kg–1, 3.9 g Ca kg–1, and 2.0 g Mg kg–1 (Table 1). The initial N concentrations (20 to 24 g N kg–1) were above the theoretical values (17 to 18 g N kg–1), and corresponding C to N ratios (16:1 to 17:1) were below the theoretical value (25:1) for net N mineralization to occur (Constantinides and Fownes, 1994; Aber and Melillo, 1980; Allison, 1966). Structural carbohydrate concentrations ranged from 251 to 325 g cellulose kg–1 and 80 to 104 g lignin kg–1. In addition to the C to N ratio as an indicator of how quickly residues would decompose, low lignin/N ratios have also significantly correlated with N release from decomposing plant residues (Ranells and Wagger, 1992). Collectively, the tissue chemical characteristic data were indicative of a resource quality supporting nutrient release from decomposing heart-of-palm foliage residues. The chemical composition of peach palm residues in a study conducted in a peach palm–cupuassu [Theobroma grandiflorum (Willdenow ex Sprengel) Schumann] agroforestry system (McGrath et al., 2000) reflected lower concentrations of N (9.7 g kg–1) and P (0.5 g kg–1) compared with our results but a similar concentration of C (446 g kg–1). Due to the lower N and P tissue concentrations in the McGrath et al. study, the C to P and C to N ratios were about four- and threefold greater than the ratios obtained in our study.
Leaf Residue Decomposition
Comparison of full vs. reduced models indicated that regardless of period, decomposition of the heart-of-palm residues was similar; thus, instead of having full models for each season, a reduced model fitted the three seasons (Table 2, Fig. 2a)
. This result is not unexpected since not a single month had less than 100 mm of rainfall during the study, and water availability was likely not a limiting factor on residue decomposition—nor was nutrient concentration. A single exponential three-term model best described dry weight loss from decomposing leaves, with 50% of the dry weight remaining approximately 7 wk after field placement and only 30% remaining by Week 15.
View this table:
[in this window]
[in a new window]
|
Table 2. Parameter coefficients of the models describing decomposition and nutrient release of peach palm foliage residue from stems harvested for heart-of-palm.
|
|
View larger version (34K):
[in this window]
[in a new window]
|
Fig. 2. Percentage of initial peach palm foliage (a) dry matter and (b–f) nutrient contents (N, P, K, Ca, and Mg) remaining in litterbags as a function of time after placement in the field for each of the three periods investigated. Symbols correspond to mean values for four replications, and the lines correspond to the prediction models described in Table 2.
|
|
The residue decomposition rate reported by McGrath et al. (2000) for peach palm residues was approximately 14-fold greater than that obtained in our study (2.1 bi-wk–1 vs. 0.1472 wk–1), with only 40% of the original residue remaining after 1 yr compared with 30% of the original residue remaining in our study at Week 15. The difference in results may be associated with the fitted models and the system used since McGrath et al. (2000) measured decomposition and nutrient release biweekly in an agroforestry system and utilized a single two-term exponential model while we worked with a monocrop and used a single three-term exponential model.
Leaf Residue Nutrient Release Patterns
Reduced models for all three placement periods were justified for N, P, and K since model coefficients were similar among the wet and dry periods (Table 2). A full model best described Ca release in period W1, and a reduced model represented periods W2 and D. Magnesium release was best fit by full models in each period. A single exponential three-term model provided the lowest mean square error value for N, P, K, Ca (W2, D), and Mg release. A single exponential two-term model best described the release of Ca from residues in period W1. The inherent variability in the Ca data sets (Luna-Orea et al., 1996; Njunie et al., 2004) was reflected in considerably larger mean square error values.
Similar to the dry matter disappearance results, the release of nutrients from heart-of-palm residues was relatively rapid and in the order of N = K > P > Mg > Ca (Table 2). With the exception of Ca and Mg (W1), more than 50% of the nutrients (i.e., N, P, K, and Mg) had been released by 6 wk (Fig. 2b–2f). Ten weeks after field placement, the mean estimated nutrient content percentages remaining in heart-of-palm leaf residues were 23% N, 33% P, 9% K, 44% Ca (W2, D), and 36% Mg.
The relatively fast release of K (0.2805 wk–1) is understandable, given the role of this nutrient in the cell solution. Once cell wall lysis occurs, K is subject to release. Similarly, the relatively fast release of the other primary nutrients (N = 0.2819 wk–1 and P = 0.2282 wk–1) observed in our work is consistent with the results from other decomposition studies involving leguminous plant residues (Luna-Orea et al., 1996; Palm and Sanchéz, 1990; Thomas and Asakawa, 1993). These results suggest that decomposition rate and nutrient release are governed by resource quality, most notably C to N ratio. It is interesting to note that P, Ca (W1, W2), and Mg (D), despite having relatively rapid decomposition rates, still reflected 24 to 34% of these nutrients not released during the course of the study (Table 2).
Leaf Residue Nutrient Availability Estimates
To fully understand the effect that decomposing heart-of-palm leaf residues may exert on potential nutrient contributions to the heart-of palm production system, it is important to quantify the amount of released nutrients. The nonlinear equations in Table 2 were used to predict cumulative annual release of nutrients for the initial 48 wk of foliage dry matter and nutrient content data collected in a companion study during harvest of offshoots at 4-wk intervals in 4- and 8-yr heart-of-palm stands (Table 3). The reduced models for periods W2 and D were used for Ca, and the release of Mg was predicted with the W2-period model. Predictions for the 4-yr heart-of-palm stand shown in Fig. 3
indicate that sizeable nutrient contributions are associated with the leaves remaining in the field after harvest of the stems. Estimates for cumulative additions of foliage dry matter reflect variations in the number of stems and associated leaves harvested across the 4-wk intervals, including values of zero at Weeks 0 and 20. The potentially available residue-derived nutrient pools by 24 wk, averaged across the two heart-of-palm stands, were 66 kg N ha–1, 9 kg P ha–1, 62 kg K ha–1, 13 kg Ca ha–1, and 9 kg Mg ha–1. Over the course of the 52-wk study period, nutrient release estimates for the two heart-of-palm stands range from 93 to 107 kg N ha–1, 14 to 15 kg P ha–1, 90 to 116 kg K ha–1, 19 to 23 kg Ca ha–1, and 13 to 14 kg Mg ha–1 (Table 3).
View this table:
[in this window]
[in a new window]
|
Table 3. Comparisons among nutrients in fertilizer inputs, accumulation in harvested and pruned offshoots, and release from foliage residue mulch in 4- and 8-year peach palm stands.
|
|
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 3. Cumulative additions of foliage residue and predicted nutrient release during 48 consecutive weeks of heart-of-palm harvests in a 4-yr peach palm stand at Guápiles, Costa Rica. Foliage dry matter and nutrient content data are from a companion study.
|
|
Potential fertilizer uptake for the 4- and 8-yr heart-of-palm stands was estimated as the portion of nutrient uptake in harvested offshoots that exceeded the cumulative nutrient release from decomposing foliage residues. The potential fertilizer applied, however, may be larger if we consider that nutrient fluxes with throughfall and stemflow likely occurred under this tropical wet climate. For example, Schroth et al. (1999) reported that palms have very high stemflow, which may increase nutrient leaching when fertilizer is applied close to the stem. Additionally, Schroth et al. (2001) found in Central Amazonia that the crowns of heart-of-palm monoculture increased total N fluxes by 4 to 5 kg ha–1, mostly in organic form, and added substantial amounts of mineral and organic P to the rainwater. Potassium on the other hand, increased the flux in the rainwater five- to sevenfold. The flux contribution to the total nutrient return was 14% N, 77% K, and from 5 to 10% for P, Ca, and Mg.
Another aspect that has to be considered is the peach palm recovery of N from fertilizer and residues. Haggar et al. (1993) in Costa Rica reported corn N recovery percentages from mulches in an 8-yr-old alley cropping system ranging from 5 to 10%. Peach palm develops a root mat that rises as much as a meter above the soil surface (Mora-Urpí et al., 1997; McGrath et al., 2000), with most of this root biomass concentrated in the top 20-cm soil layer underneath the crown area (Lópes Morales and Sancho Vargas, 1990; Ferreira et al., 1980; Ferreira et al., 1995). With root/shoot ratios between 0.86 and 1.88 (Haag, 1997), nutrient recovery is generally considered to be quite efficient. Juo and Lal (1977) demonstrated that returning crop residues to the soil might reduce the decline in soil N during cropping. Haggar et al. (1993) concluded that the ability of the alley cropping system to maintain high N availability longer than sole cropping systems is primarily due to the abundant organic matter inputs from the trees. In the present study, the 16-yr-old peach palm plantation is expected to have accumulated readily mineralizable N forms and augmented the organic-bound nutrients in the soil as a result of adding high resource quality organic matter for years (Anderson and Ingram, 1989). Thus, the potential nutrient contributions from leaf residues can be of paramount interest in reducing commercial fertilizer inputs. In a perennial cropping system such as heart-of-palm, it would appear that the potential exists for the near-optimal synchronization between release of nutrients from decomposing leaf residues and crop nutrient requirements. The fact that this perennial crop is producing harvestable yield on 2- to 4-wk intervals suggests an enhanced all-year-long synchronization potential between plant requirements and nutrient release from decomposing leaf residues because the plant is continuously accumulating nutrients either to be used or stored in the well-developed root mat and associated microorganisms. Comparison between fertilizer inputs and potential fertilizer uptake values in Table 3 provides an indication of the potential nutrient contributions from foliage residues. These comparisons suggest a surplus of inputs for N, P, K, and Mg and potential reductions in soil Ca reserves. The certainties of these figures, however, need the support of additional research on the fate of released nutrients and soil organic matter dynamics in this cropping system.
|
CONCLUSIONS
|
|---|
Under the wet tropical conditions that occurred during the course of this study, leaf residues left on peach palm plantations used for heart-of-palm production decompose at the same rate, irrespective of seasonal rainfall distribution patterns. Additionally, the resource quality of the residues, primarily reflected by the C to N ratios, did not impose constraints to decomposition. Nevertheless, the use of single-exponential two- and three-term models indicates that some other factors are controlling the complete decomposition of residue and release of nutrients during the course of the study. Nutrients remaining in the field as crop residues and subsequently released may meet the nutrient needs of the plant, but more information is needed on the recovery of fertilizer- and residue-derived nutrients by peach palm, as well as the nutrient-supplying capacity of the soil. The fact that the harvest proceeds throughout the year and for each harvest a portion of residue and nutrients are not decomposed or released might lead to the accumulation of organic nutrient pools from where respective nutrients would be released through mineralization. Thus, this study also points to the need for a more ambitious multidisciplinary project, one which could address in more detail the nutrient dynamics of this perennial cropping system, thereby leading to critical decisions of soil and crop management.
|
ACKNOWLEDGMENTS
|
|---|
This investigation was supported by the project on Decision Aids for Soil Nutrient Management of the Soil Management Collaborative Research Support Program and funded in part by grant no. LAG-G-00-97-00002-00 from the U.S. Agency for International Development. We thank the staff of DEMASA, S.A. for access to peach palm plantations on their farm at Guápiles, Costa Rica.
|
REFERENCES
|
|---|
- Aber, J.D., and M.J. Melillo. 1980. Litter decomposition: Measuring relative contributions of organic matter and nitrogen in forest soils. Can. J. Bot. 58:416–421.
- Allison, F.E. 1966. The fate of nitrogen applied to soils. Adv. Agron. 18:219–258.
- Anderson, J.M., and J.S.I. Ingram. 1989. Tropical soil biology and fertility: A handbook of methods. CAB Int., Wallingford.
- Ares, A., J. Boniche, E. Molina, and R.S. Yost. 2002. Bactris gasipaes agroecosystems for heart-of-palm production in Costa Rica: Changes in biomass, nutrient and carbon pools with stand age and plant density. Field Crops Res. 74:13–22.
- Bovi, M.L.A. 1997. Expansão do cultivo da pupunheira para palmito no Brasil. Hortic. Bras. 15(suppl.):183–185.
- Bovi, M.L.A. 1998. Palmito pupunha: Informações básicas para cultivo. Boletin Técnico 173. Instituto Agronômico, Campinas, Brazil.
- Bovi, M.L.A., and H. Cantarella. 1997. Pupunha para extração de palmito. In B. Van Raij, H. Cantarella, J.A.Quaggio, and A.M.Cangiani (ed.) Recomendações de adubação e calagem para o estado de São Paulo. Boletim Técnico 100. Instituto Agronômico, Campinas, Brazil.
- Chapman, H.D., and P.F. Pratt. 1973. Métodos de análisis para suelos, plantas y aguas. Editorial Trillas, Mexico D.F., Mexico.
- Clement, C.R. 1986. The pejibaye palm (Bactris gasipaes H.B.K.) as an agroforestry component. Agrofor. Syst. 4:205–219.
- Clement, C.R. 1989. The potential use of the pejibaye palm in agroforestry systems. Agrofor. Syst. 7:201–212.
- Clement, C.R. 1995. Growth and genetic analysis of pejibaye (Bactris gasipaes Kunth, Palmae) in Hawaii. Ph.D. diss. Univ. of Hawaii, Manoa.
- Clement, C.R., R.M. Manshardt, J. DeFrank, C.G. Cavaletto, and N.Y. Nagai. 1996. Introduction of pejibaye for heart-of-palm in Hawaii. HortScience 31:765–768.[Free Full Text]
- Constantinides, M., and J.H. Fownes. 1994. Nitrogen mineralization from leaves and litter of tropical plants: Relationship to nitrogen, lignin, and soluble polyphenol concentrations. Soil Biol. Biochem. 26:49–55.[CrossRef]
- Deenik, J., A. Ares, and R.S. Yost. 2000. Fertilization response and nutrient diagnosis in peach palm (Bactris gasipaes): A review. Nutr. Cycling Agroecosyst. 56:195–207.
- Ferreira, S.A.N., C.R. Clement, and G. Ranzani. 1980. Contribução para o conhecimiento do sistema radicular da pupunheira (Bactris gasipaes (H.B.K.)-Guilielma gasipaes (H.B.K.) Bailey): I. Solo Latossolo Amarelo, textura media. Acta Amazonica 10:245–249.
- Ferreira, S.A.N., C.R. Clement, G. Ranzani, and S.S. Costa. 1995. Contribução para o conhecimiento do sistema radicular da pupunheira (Bactris gasipaes (H.B.K.)–Guilielma gasipaes (H.B.K.) Bailey): II. Solo Amarelo, textura argilosa. Acta Amazonica 25:161–170.
- Haag, D. 1997. Root distribution patterns in a polycultural system with local tree crops on an acid upland soil in Central Amazonia. M.Sc. thesis. Universitat Bayreuth, Bayreuth, Germany.
- Haggar, J.P., E.V.J. Tanner, J.W. Beer, and D.C.L. Kass. 1993. Nitrogen dynamics of tropical agroforestry and annual cropping systems. Soil Biol. Biochem. 25(10):1363–1378.[CrossRef]
- Herrera, W. 1989. Fertilización del pejibaye para palmito. Serie Técnica Pejibaye, Boletín Informativo, Vol. 1. Universidad de Costa Rica, San José.
- Jordan, C.F. 1988. Permanent plots for agriculture and forestry. p. 58–89. In C.F. Jordan (ed.) Amazonian rainforests: Ecosystem disturbance and recovery. Springer-Verlag, Berlin, Heidelberg, Germany.
- Juo, A.S.R., and R. Lal. 1977. Effects of fallow and continuous cultivation on the chemical and physical properties of an alfisol in W. Nigeria. Plant Soil 47:567–584.
- Lópes Morales, A., and H. Sancho Vargas. 1990. Observaciones sobre la distribución radical del pejibaye (Bactris gasipaes H.B.K.) para palmito en un andosol. Rev. Corporación Bananera Nacional 14:9–15.
- Luna-Orea, P., M.G. Wagger, and M.L. Gumpertz. 1996. Decomposition and nutrient release dynamics of two tropical legume cover crops. Agron. J. 88:758–764.[Abstract/Free Full Text]
- McGrath, D.A. 1998. Ecological sustainability in Amazonian agroforests: An on-farm study of phosphorus and nitrogen dynamics following native forest conversion. Ph.D. diss., Univ. of Florida, Gainesville.
- McGrath, D.A., N.B. Comerford, and M.L. Duryea. 2000. Litter dynamics and monthly fluctuations in soil phosphorus availability in an Amazonian agroforest. For. Ecol. Manage. 131:167–181.
- Mora Urpí, J., J.C. Weber, and C.R. Clement. 1997. Peach palm, Bactris gasipaes Kunth. Promoting the conservation and use of underutilized and neglected crops. Vol. 20. Int. Plant Genetic Resour. Inst., Rome.
- Njunie, M.N., M.G. Wagger, and P. Luna-Orea. 2004. Residue decomposition and nutrient release dynamics from two tropical forage legumes in a Kenyan environment. Agron. J. 96:1073–1081.[Abstract/Free Full Text]
- Palm, C.A., and P.A. Sanchéz. 1990. Decomposition and nutrient release patterns of the leaves of tropical legumes. Biotropica 22(4):330–338.[CrossRef]
- Pidwirny, M.J. 2002. Introduction to meteorology and climatology: V. Climate classification and climatic regions of the world. Fundamentals of Physical Geography. Version 1.3. Dep. of Geography, Okanagan Univ. College, Kelowna, BC, Canada.
- Quintana de la Flor, C., R. Valeriani-Vela, L. Martel-Mori, and C. Huaman De la Cruz. 1993. Projecto de pre-factibilidad de produción y exportación de conservas de palmito. p. 375–402. In J. Mora-Urpí, L.T. Szott, M. Murillo, V.M. Patiño (ed.) IV Congreso Internacional Sobre Biología, Agronomía e Industrialización del Pijuayo, Iquitos, Peru. November 1991. Editorial de la Universidad de Costa Rica, San José, Costa Rica.
- Ranells, N., and M.G. Wagger. 1992. Nitrogen release from crimson clover in relation to plant growth stage and composition. Agron. J. 84:424–430.[Abstract/Free Full Text]
- Ratkowsky, D.A. 1983. Nonlinear regression modeling: A unified practical approach. M. Dekker, New York.
- SAS Institute. 1998. The NLIN procedure. SAS Inst., Cary, NC.
- Schroth, G., L.F. da Silva, M.A. Wolf, W.G. Teixeira, and W. Zech. 1999. Distribution of throughfall and stemflow in multi-strata agroforestry, perennial monoculture, fallow and primary forest in central Amazonia, Brazil. Hydrol. Processes 13:1423–1436.
- Schroth, G., M.E.A. Elias, K. Uguen, R. Seixas, and W. Zech. 2001. Nutrient fluxes in rainfall, throughfall and stemflow in tree-based land use systems and spontaneous tree vegetation of central Amazonia. Agric. Ecosyst. Environ. 87:37–49.[CrossRef]
- Schroth, G., W.G. Teixeira, R. Seixas, L.F. da Silva, M. Shaller, J.L.V. Macêdo, and W. Zech. 2000. Effect of five tree crops and a cover crop in multi-strata agroforestry at two fertilization levels on soil fertility and soil solution chemistry in central Amazonia. Plant Soil 221:143–156.
- Thomas, R.J., and N.M. Asakawa. 1993. Decomposition of leaf litter from tropical forage grasses and legumes. Soil Biol. Biochem. 25(10):1351–1361.[CrossRef]
- Van Soest, P.J., and R.H. Wine. 1968. Determination of lignin and cellulose in acid-detergent fiber with permanganate. J. Assoc. Off. Anal. Chem. 51:780–785.
TOP
ABSTRACT
INTRODUCTION
