Journal of Environmental Quality 30:1848-1852 (2001)
© 2001 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
SHORT COMMUNICATION
Contrasting Nitrate Adsorption in Andisols of Two Coffee Plantations in Costa Rica
M. C. Ryan*,
G. R. Graham and
D. L. Rudolph
Dep. of Earth Sciences, Univ. of Waterloo, Waterloo, ON, N2L 3G1, Canada
* Corresponding author (ryan{at}geo.ucalgary.ca)
Received for publication October 30, 2000.
 |
ABSTRACT
|
|---|
Fertilizer use in coffee plantations is a suspected cause of rising ground water nitrate concentrations in the ground water–dependent Central Valley of Costa Rica. Nitrate adsorption was evaluated beneath two coffee (Coffea arabica L.) plantations in the Central Valley. Previous work at one site had identified unsaturated zone nitrate retardation relative to a tritium tracer. Differences in nitrate adsorption were assessed in cores to 4 m depth in Andisols at this and one other plantation using differences in KCl- and water-extractable nitrate as an index. Significant adsorption was confirmed at the site of the previous tracer test, but not at the second site. Anion exchange capacity, X-ray diffraction data, extractable Al and Si, and soil pH in NaF corroborated that differences in adsorption characteristics were related to subtle differences in clay mineralogy. Soils at the site with significant nitrate adsorption showed an Al-rich allophane clay content compared with a more weathered, Si-rich allophane and halloysite clay mineral content at the site with negligible adsorption. At the site with significant nitrate adsorption, nitrate occupied less than 10% of the total anion adsorption capacity, suggesting that adsorption may provide long-term potential for mitigation or delay of nitrate leaching. Evaluation of nitrate sorption potential of soil at local and landscape scales would be useful in development of nitrogen management practices to reduce nitrate leaching to ground water.
Abbreviations: AEC, anion exchange capacity • XRD, X-ray diffraction
|
INTRODUCTION
|
|---|
NITRATE concentrations have increased steadily during the past few decades in regional aquifers in the Central Valley of Costa Rica. These aquifers provide potable water to about half of the greater metropolitan area of San José (Rodriguez-Estrada and Loaiciga, 1995). Coffee growing has been the traditional economic activity in this region since the early 1800s and remains a major land use despite rapid industrial and urban growth in the last few decades. Coffee plantations are mainly located at moderate elevations (900–1200 m above mean sea level) where much of the ground water recharge to underlying aquifers occurs (British Geological Survey/SENARA, 1988). Fertilizer use in coffee plantations is a suspected source of ground water nitrate. Coffee cultivation uses significant nitrogen fertilizer application rates (270 kg NO3–N ha-1 yr-1 on average; Reynolds-Vargas et al., 1994b).
Previous field investigations to estimate nitrate transport rates in the unsaturated zone beneath a Central Valley coffee plantation provided clear evidence for retardation of 15N-labeled nitrate tracer relative to a tritium tracer (Reynolds-Vargas et al., 1994a). The nitrate retardation occurred between 2 and 2.5 m depth, and adsorption and denitrification were identified as possible mechanisms for the observed behavior.
Nitrate adsorption has been previously reported in tropical Andisols (Kinjo and Pratt, 1971; Matson et al., 1987; Gonzalez-Pradas et al., 1993), which is the predominant soil order in the coffee-growing region of the Central Valley. Clay minerals formed from the weathering of tropical volcanic ash soils include amorphous aluminosilicates such as allophane and non-amorphous aluminosilicates such as halloysite (Wada, 1989; Jongmans et al., 1995). Allophane is typically an initial weathering product (Wada, 1989), and halloysite is an intermediate weathering product. Allophane has a much higher anion exchange capacity (AEC) than halloysite, due mostly to its small particle size, high surface area, and the presence of surface Al–OH–Al groups and defect sites in the mineral structure (Theng et al., 1982; Shoji et al., 1993). In addition, AEC in allophane has a significant pH dependence that is not present in AEC in halloysite (Theng at al., 1982).
We evaluated adsorption as a possible nitrate retardation mechanism at two coffee plantations in the Central Valley: San Pedro de Barva (henceforth referred to as San Pedro) and Santo Tomás. The San Pedro plantation is the site where nitrate retardation was previously observed (Reynolds-Vargas et al., 1994a). The second site, Santo Tomás, was included to provide an indication (albeit limited) of whether the nitrate retardation observed at San Pedro was geographically extensive. Soil samples were collected to 4 m depth at both plantations for mineralogical characterization and evaluation of AEC properties. Nitrate adsorption was estimated by comparing water- vs. KCl-extractable nitrate.
|
Materials and Methods
|
|---|
Site Descriptions
The San Pedro and Santo Tomás plantations were both located within 10 km of San José at about 1210 m elevation, with mean annual temperatures of about 20°C. The region has a distinct January to April dry season. Annual precipitation is about 2500 mm, of which an estimated 40 to 50% is recharged to ground water aquifers (Reynolds-Vargas et al., 1994b; Rodriguez-Estrada and Loaiciga, 1995).
Andisol soils in the region are typically well-structured, deep, and permeable, with low bulk densities and high organic matter contents. They are derived from the weathering of ash from volcanoes that flank the eastern slopes of the Central Valley. Soils at the two sites had high clay contents (about 48%; Graham, 1996), and were dominated by Udands (previously classified as Dystrandepts; Peres et al., 1978).
The San Pedro site was a small (0.4 ha) coffee plantation operated under traditional management. It featured coffee plants intermixed with other plants and trees (Coffea arabica cuttura intermixed with Coffea arabica catuai; Babbar and Zak, 1995). Fertilizer N at the family-operated plantation (in operation for generations) was irregularly applied according to crop needs and availability of funds at a relatively low rate of about 140 kg N ha-1 yr-1. Landowner permission for sampling was also procured at a large (20 ha) coffee monoculture plantation near Santo Tomás. Fertilizer nitrogen at this plantation, which had been in operation for a few decades, was applied at a relatively high rate of 360 kg N ha-1 yr-1.
Soil Sampling and Analysis
Soil samples were taken with an 8-cm-diameter screw-type auger at 20-cm intervals, except near the zone of previously observed nitrate attenuation (between 2 and 3 m depth; Reynolds-Vargas et al., 1994a), where 10-cm intervals were used. Samples were collated from two auger boreholes, located less than 0.5 m apart, to ensure sufficient soil for analyses. Boreholes remained open throughout the depth of sampling. Sample collection was carefully conducted to minimize cross-contamination from borehole sloughing. A marker (a coffee bean) was dropped downhole immediately before auger insertion, and sediment collected above the marker was discarded from the sample. Bulk densities were estimated from samples of known volume taken with a 3-cm-diameter steel punch-tube core sampler at approximately 1-m intervals.
The pH-dependant anion exchange capacity (AEC) was estimated for soils within (2.4 m) and below (3.4 m) the zone of previously observed nitrate retardation. One gram of air-dried soil was saturated with 0.05 M Ca(NO3)2 solution, and pH adjusted by the addition of either 0.1 M HNO3 or 0.05 M Ca(OH)2 (Hendershot et al., 1993). Soil pH was measured in CaCl2 (5:12.5 w/v, 0.01 M CaCl2; Tan, 1993).
The difference between water and KCl-extractable nitrate provided an index of the amount of adsorbed nitrate. Soils were suspended in distilled water or KCl solution (1:10 soil to solution [w/v], 1 M KCl concentration), shaken for 1.5 h, and centrifuged for 5 min (3500 rpm). Extracts were filtered (0.45 µm) and analyzed for nitrate. Nitrate analyses were conducted by automated azo-dye colourimetry (Reynolds-Vargas et al., 1994b) for San Pedro samples, and by cadmium reduction to nitrite with colorimetric detection (Technicon [Tarrytown, NY] Industrial Method 100-70W/B) for Santo Tomás samples. Reagent grade standards were prepared for both methods.
Clay mineralogy was evaluated for the presence of allophane by X-ray diffraction (XRD), and the measurement of soil pH in NaF, and extractable Al and Si. Oriented specimens of the clay-sized fraction were prepared for X-ray diffraction using the paste method (Theisen and Harward, 1962). Specimen pretreatments were conducted with electrolyte free Ca- and K-saturated solutions according to Dudas and Pawluk (1982). The pretreatments included equilibration at 54% relative humidity (RH), solvation with ethylene glycol (EGL), glycerol (GLY), and heating to 300°C and 550°C. The X-ray diffraction patterns for samples from several depths were obtained using Cu-alpha radiation generated at 40 KV and 30 mA. Although not definitive (Shoji et al., 1993), soil pH in NaF (1:50 soil to solution [w/v], 1 M NaF) above 9.4 suggests the presence of amorphous alumino-silicate minerals like allophane (Fields and Perrot, 1966). Ammonium oxalate–extractable Si and Al (McKeague, 1978) were determined using soil samples from 2.3 m at both sites. Allophane contributes to an elevated extractable Al content, which has been correlated with increased anion exchange capacity in amorphous clays (Clark and McBride, 1984). Extractable Al and Si samples were analyzed by inductively coupled plasma–mass spectrometry (ICP–MS).
Results and Discussion
Bulk densities measured from the two sites (average 
b = 1.1 g cm-3, 
= 0.40, and n = 4 for San Pedro, and average b = 1.1 g cm-3, = 0.16, and n = 5 for Santo Tomás) were similar to each other, and to documented values for Andisols (Shoji et al., 1993).
Evidence for Nitrate Sorption
The San Pedro soil exhibited a relatively high AEC (Fig. 1), consistent with previous studies of tropical and subtropical volcanic ash soils (Singh and Kanehiro, 1969; Kinjo and Pratt, 1971; Schalscha et al., 1973; Wong et al., 1990) and similar to values reported for allophane (Wada, 1989). The high AEC and its pH dependence (Fig. 1) suggest a high allophane content in these soils (Theng et al., 1982). Average soil pH (CaCl2) values of 4.4 to 5 at San Pedro correspond to a nitrate adsorption capacity of 300 to 400 µg N g-1 (Fig. 1). The AEC in Santo Tomás soils was substantially lower and exhibited less pH dependence. Typical Santo Tomás soil pH (CaCl2) values (4.5 to 5.5) correspond to a nitrate adsorption capacity of less than 50 µg N g-1 (Fig. 1).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 1. pH-Dependent anion exchange capacity (AEC) for soils from 230 (filled symbols) and 340 cm (open symbols) depth.
|
|
Nitrate concentrations obtained from KCl extracts were significantly higher (110% on average; Fig. 2A) than distilled water extracts at the San Pedro site, indicating the presence of significant amounts of adsorbed nitrate in the upper 3 m of the soil profile. In contrast, similar water- and KCl-extractable nitrate concentrations in the Santo Tomás soils suggest negligible nitrate adsorption occurred at this site (Fig. 2B). Although data scatter suggests soil heterogeneity, there is a clear statistical difference between the sites (Table 1). The statistical analyses are complicated by data sets that are not consistently normally or non-normally distributed. Nonetheless, both parametric and nonparametric analyses consistently indicate a significant difference between KCl- and water-extractable nitrate in San Pedro soils, but not in Santo Tomás soils. Further, the differences between KCl- and water-extractable nitrate (i.e., an index of the amount of adsorbed nitrate) were significantly different between the two sites (Fig. 2C, Table 1). Relatively high soil pH in NaF at the San Pedro site (Fig. 2D) further suggests a high allophanic composition and associated nitrate adsorption capacity, relative to Santo Tomás.
View larger version (32K):
[in this window]
[in a new window]
|
Fig. 2. Soil analyses with depth. Water-extractable nitrate (open symbols) and KCl-extractable nitrate (filled symbols) for (A) Santo Tomás and (B) San Pedro. (C) Difference between KCl- and water-extractable nitrate for Santo Tomás (triangles) and San Pedro (squares). (D) Soil pH in CaCl2 and NaF for Santo Tomás (triangles) and San Pedro (squares).
|
|
View this table:
[in this window]
[in a new window]
|
Table 1. Statistical analysis of KCl- and water-extractable nitrate data from Santo Tomás and San Pedro soils (Figure 2). Nonparametric test (Wilcoxon signed rank test) conducted on difference between KCl- and water-extractable NO3 concentrations.
|
|
Mass balance calculations using the KCl- vs. water-extractable NO3–N data indicate that at least 300 kg NO3–N ha-1 are adsorbed in the 4-m San Pedro profile. Since nitrate leaching estimates are thought to be a fraction of fertilizer application rates, nitrate from numerous years of fertilizer application may be accumulating in the 4-m profile by adsorption. The total adsorption capacity of the 4-m soil profile, estimated using an average AEC of 2 cmol kg-1 (Fig. 1), is more than 10 Mg ha-1. This suggests that only a small percentage (<10%) of the total adsorption capacity is being occupied in the shallow soil profile at San Pedro and that nitrate adsorption may be a long-term process at this site.
Clay Mineralogy
Subtle mineralogical differences corroborate a high nitrate adsorption capacity in San Pedro relative to Santo Tomás. Overall, the XRD patterns are similar and suggest significant amorphous clay content at both sites (Fig. 3). Previous studies of Costa Rican Andisols (Nieuwenhuyse et al., 1994; Jongmans et al., 1995) suggest that allophane should constitute a significant proportion of the amorphous clay content. In XRD results, metahalloysite (resulting from halloysite dehydration during pretreatment) is indicated by diffraction peaks at 0.73 and 0.45 nm (Dixon, 1989). In particular, the peaks at 0.73 nm decrease with depth in the San Pedro samples, but remain more constant with depth at the Santo Tomás samples (Fig. 3). This suggests a greater depth of soil weathering at the Santo Tomás site.
View larger version (37K):
[in this window]
[in a new window]
|
Fig. 3. X-ray diffraction pattern for clay fractions of soil samples from 0.8, 2.3, and 3.4 m depth. Values along top of each pattern indicate selected d-spacings.
|
|
Substantially more ammonium oxalate-extractable aluminum (AlE) and silica (SiE) was in San Pedro soils (AlE = 16.8 g kg-1, SiE = 31.2 g kg-1) relative to Santo Tomás (AlE = 2.8 g kg-1, SiE = 2.7 g kg-1), further suggesting a higher allophane content at San Pedro. The higher concentration of soluble amorphous silica at the San Pedro site is characteristic of less-weathered volcanic ash soils (Shoji et al., 1993) and associated higher AEC (Clark and McBride, 1984). Conversely, lower extractable Al and Si concentrations in the Santo Tomás soil suggest more Si-rich allophane and halloysite clay mineralogy, characteristic of a more weathered volcanic ash soil.
Nitrate adsorption under coffee plantations in the Central Valley of Costa Rica appears to be a significant and potentially long-term, but spatially variable, process. Seemingly subtle changes in soil mineralogy are reflected in substantially different propensities for nitrate retardation. The degree of variability of the nitrate adsorption capacity of the soils in the region should to be investigated at local and landscape scales, and with depths, to assess their adsorption capacities in the context of nitrate leaching to ground water. In addition, the effect of agricultural management practices (e.g., liming) on nitrate adsorption capacity should be considered. Such information is critical for strategies to protect the regional aquifers that supply potable water to the metropolitan area of San Jose.
|
ACKNOWLEDGMENTS
|
|---|
Thanks are extended to Dr. J. Reynolds-Vargas for helpful discussion and facilitation of site access. Dr. J. Warren's assistance is gratefully acknowledged. Funds were provided by the Canadian International Development Agency.
|
REFERENCES
|
|---|
- Babbar, L.I., and D.R. Zak. 1995. Nitrogen loss from coffee agroecosystems in Costa Rica: Leaching and denitrification in the presence and absence of shade trees. J. Environ. Qual. 24:227–233.[Abstract/Free Full Text]
- British Geological Survey/SENARA (Groundwater Bureau of Costa Rica). 1988. Continuation of the hydrogeological investigation in the zone north and east of the Valle Central, 1984–1987. Final rep. SENARA, San Jose, Costa Rica.
- Clark, C.J., and M.B. McBride. 1984. Cation and anion retention by natural and synthetic allophane and imogolite. Clays Clay Miner. 32:291–299.[Abstract]
- Dixon, J.B. 1989. Kaolin and serpentine group. p. 467–525. In J.B. Dixon and S.B. Weed (ed.) Minerals in soil environments. 2nd ed. SSSA Book Ser. 1. SSSA, Madison, WI.
- Dudas, M.J., and S. Pawluk. 1982. Reevaluation of the occurrence of interstratified clays and other phyllosilicates in southern Alberta soils. Can. J. Soil Sci. 62:61–69.
- Fields, M., and K.W. Perrot. 1966. The nature of allophane in soils: Part 3. Rapid field and laboratory tests for allophane. N.Z. J. Soil Sci. 9:623–629.
- Gonzalez-Pradas, E., M. Villafranca-Sanchez, and M. Socias-Viciana. 1993. Phosphate and nitrate sorption on calcareous soils from Spain. Arid Soil Res. Rehab. 7:181–190.
- Graham, G. 1996. Nitrate attenuation beneath two coffee plantations in the Central Valley of Costa Rica. M.Sc. thesis. Univ. of Waterloo, ON, Canada.
- Hendershot, W.H., H. Lalande, and M. Duquette. 1993. Ion exchange and exchangeable cations. p. 183–205. In R.C. Martin (ed.) Soil sampling and methods of analysis. Canadian Soc. of Soil Sci. Lewis Publ., Boca Raton, FL.
- Jongmans, A.G., P. Verburg, A. Nieuwenhuyse, and F. Van Oort. 1995. Allophane, imogolite and gibbsite in coatings in a Costa Rican Andisol. Geoderma 64:327–342.
- Kinjo, T., and P.F. Pratt. 1971. Nitrate adsorption: I. In some acid soils of Mexico and South America. Soil Sci. Soc. Am. Proc. 35:722–732.
- Matson, P.A., P.M. Vitousek, J.J. Ewel, M.J. Mazzarion, and G.P. Robertson. 1987. Nitrogen transformations following tropical forest felling and burning on a volcanic soil. Ecology 68:491–502.[ISI]
- McKeague, J.A. (ed.) 1978. Manual on soil sampling and methods of analysis. 2nd ed. Subcommittee on Methods of Analysis of the Canadian Soil Survey Committee, Ottawa, ON, Canada.
- Nieuwenhuyse, A., A.G. Jongmans, and N. Van Breemen. 1994. Mineralogy of a Holocene chronosequence on Andesitic beach sediments in Costa Rica. Soil Sci. Soc. Am. J. 58:485–494.[Abstract/Free Full Text]
- Peres, S.R., E. Ramirez, and E. Alvarado. 1978. Mapa preliminar de asociaciones de subgrupos de suelos de Costa Rica. OPSA, San José, Costa Rica.
- Reynolds-Vargas, J., L. Araguas, J. Fraile, L. Castro, and K. Rozanski. 1994a. Nitrate leaching through volcanic soils in the Central Valley of Costa Rica. p. 549–559. In Proc. of the Int. Symp. on Nuclear and Related Techniques in Soil–Plant Studies on Sustainable Agric. and Environ. Preservation. Int. Atomic Energy Agency and the U.N. Food and Agric. Organization, Vienna, Austria. 17–21 Oct. 1994. SM-334/29. IAEA, Vienna, Austria.
- Reynolds-Vargas, J., D.D. Richter, and E. Bornemisza. 1994b. Environmental impacts of nitrification and nitrate adsorption in fertilized Andisols in the Valle Central, Costa Rica. Soil Sci. 157:289–299.
- Rodriguez-Estrada, H., and H.A. Loaiciga. 1995. Planning urban growth in ground water recharge areas: Central Valley, Costa Rica. Ground Water Monit. Rem. 15:144–148.
- Schalscha, E.B., P.F. Pratt, and T.C. Domecq. 1973. Nitrate adsorption by some volcanic-ash soils of Southern Chile. Soil Sci. 117:44–45.
- Shoji, S., M. Nanzyo, and R.A. Dahlgren. 1993. Volcanic ash soils: Genesis, properties and utilization. Elsevier Publ., Amsterdam, the Netherlands.
- Singh, B.R., and Y. Kanehiro. 1969. Adsorption of nitrate in amorphous and kaolinitic Hawaiian soils. Soil Sci. Soc. Am. Proc. 31:681–683.
- Tan, K.H. 1993. Principles of soil chemistry. Marcel Dekker, New York.
- Theisen, A.A., and M.E. Harward. 1962. A paste method for preparation of slides for clay mineral identification by x-ray diffraction. Soil Sci. Soc. Am. Proc. 26:90–91.
- Theng, B.K.G., M. Russell, G.J. Churchman, and R.L. Parfitt. 1982. Surface properties of allophane, halloysite and imogolite. Clays Clay Miner. 30:143–149.[Abstract]
- Wada, K. 1989. Allophane and imogolite. p. 1051–1087. In J.B. Dixon and S.B. Weed (ed.) Minerals in soil environments. 2nd ed. SSSA Book Ser. 1. SSSA, Madison, WI.
- Wong, M.T.F., R. Hughes, and D.L. Rowell. 1990. The retention of nitrate in acid soils from the tropics. Soil Use Manage. 6:72–74.
This article has been cited by other articles:
|
|
|
|
 
M. Maeda, H. Ihara, and T. Ota
Deep-Soil Adsorption of Nitrate in a Japanese Andisol in Response to Different Nitrogen Sources
Soil Sci. Soc. Am. J.,
May 1, 2008;
72(3):
702 - 710.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. J. Allred, J. M. Bigham, and G. O. Brown
The Impact of Clay Mineralogy on Nitrate Mobility under Unsaturated Flow Conditions
Vadose Zone J.,
April 9, 2007;
6(2):
221 - 232.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. E. Clay, Z. Zheng, Z. Liu, S. A. Clay, and T. P. Trooien
Bromide and Nitrate Movement through Undisturbed Soil Columns
J. Environ. Qual.,
January 1, 2004;
33(1):
338 - 342.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
