46: Biogeochemistry 4565, 1999. 1999 Kluwer Academic Publishers. Printed in the Netherlands.
Nitrogen stable isotopic composition of leaves and soil: Tropical versus temperate forests
L.A. MARTINELLI1, M.C. PICCOLO1, A.R. TOWNSEND2, P.M. VITOUSEK3, E. CUEVAS4, W. MCDOWELL5, G.P. ROBERTSON6, O.C. SANTOS7 & K. TRESEDER3
1 Cena Av. Centenrio 303, Piracicaba-SP. 13416-000, Brazil; 2 INSTAAR and Department
of EPO-Biology, Campus Box 450, University of Colorado, Boulder, CO 80309, U.S.A.;
3 Department of Biological Sciences, Stanford University, Stanford, CA 94305-5020, U.S.A.; 4 IVIC, BAMCO CCS-199-00, PO Box 025322, Miami, Fl 33102-5322, U.S.A.; 5 Department
of Natural Resources, University of New Hampshire, Durham, NH, 03824, U.S.A.;
6 W.K. Kellogg Biological Station, Michigan State University, Hickory Corners, MI 9060-9516, U.S.A.; 7 Meteorological Institute, Ministry of Science Technology and
Environment, Aptdo 17032, Habana 17, CP 11700, Havana, Cuba Received 10 December 1998 Key words: N15, nitrogen, nutrient cycling, plants, stable isotopes, soil, temperate forest, tropical forest Abstract. Several lines of evidence suggest that nitrogen in most tropical forests is relatively more available than N in most temperate forests, and even that it may function as an excess nutrient in many tropical forests. If this is correct, tropical forests should have more open N cycles than temperate forests, with both inputs and outputs of N large relative to N cycling within systems. Consequent differences in both the magnitude and the pathways of N loss imply that tropical forests should in general be more 15 N enriched than are most temperate forests. In order to test this hypothesis, we compared the nitrogen stable isotopic composition of tree leaves and soils from a variety of tropical and temperate forests. Foliar 15 N values from tropical forests averaged 6.5 higher than from temperate forests. Within the tropics, ecosystems with relatively low N availability (montane forests, forests on sandy soils) were signicantly more depleted in 15 N than other tropical forests. The average 15 N values for tropical forest soils, either for surface or for depth samples, were almost 8 higher than temperate forest soils. These results provide another line of evidence that N is relatively abundant in many tropical forest ecosystems.
Introduction A number of lines of evidence suggest that N in most tropical forests is relatively more available than is N in most temperate forests. On average, more N circulates annually through lowland tropical forests, and does so at higher concentrations, than through temperate forests (Proctor et al. 1983;
46 Vitousek 1984; Vitousek & Sanford 1986; Vogt et al. 1986). Emissions of N-containing trace gases are also higher, both absolutely and as a fraction of N circulating through forests (Keller et al. 1986, 1993; Matson & Vitousek 1987, 1990). Comparable data on rates of N mineralization and leaching losses are sparser, but they generally show greater rates of N cycling and loss in many lowland tropical forests (Vitousek & Denslow 1986; Lewis 1986; Matson et al. 1987; Neill et al. 1995). Overall, these observations suggest that N functions as an excess nutrient in most tropical forests, but not in the majority of temperate forests. The major exceptions to this generalization in the tropics are forests on white-sand soils and montane tropical forests; by the measures above, N is in relatively short supply in these ecosystems (Salati et al. 1982; Cuevas & Medina 1988; Tanner et al., in press). In the temperate zone, the major exceptions are forests dominated by symbiotic N xing trees (usually monocultures), and forests that receive substantial anthropogenic N deposition (Binkley et al. 1992; Aber et al. 1995; Berendse et al. 1993). If N functions as an excess nutrient in tropical forests, the N cycle in such systems should be more open than in temperate forests, with both inputs and outputs of N large relative to internal N cycling. Moreover, the pathways of N loss should differ: losses from low-N systems may be predominantly in the form of DON (Hedin et al. 1995), while leaching losses of nitrate and nitrication/denitrication driven trace gas uxes should predominate where N is in excess (Matson et al., this volume). These differences in both the magnitude and the pathways of loss imply that tropical forests should in general contain N that is more enriched in 15 N than most temperate forests. This relative enrichment should occur because pathways of loss in N-rich systems are more likely to be fractionating, and because losses by fractionating pathways leave the N remaining within the system enriched (Hogberg 1997). In order to test this hypothesis, we compared the nitrogen stable isotopic composition of tree leaves and soils from a variety of tropical and temperate forests. Methods We surveyed N stable isotopic composition and N content of leaves from adult trees of non-leguminous species from temperate and tropical forests. Where available, the same data for soil organic matter also were evaluated. As there is signicant variation with depth in both concentration and stable isotopic composition of nitrogen in soils, data were grouped according to depth. Surface samples were those collected from no more than 10 cm deep, and samples collected below this depth were called depth samples. The nitrogen stable isotopic composition was expressed as () notation: 15 N = (Rsample /Rstd 1) 1000,
47 where R is the ratio of 15 N/14 N of the sample and standard (std). The isotopic standard for nitrogen is atmospheric air. We predicted that tropical forests would have higher 15 N values in relation to temperate forests; statistical differences were tested with a t-test for unequal variance.
Results and discussion Tropical vs temperate systems The average 15 N value for tropical foliage was 3.73.5 (n = 73), which is signicantly greater (p < 0.01) than the temperate forest value of 2.82.0 (n = 90) [Table 1, Figure 1]. This 6.5 difference occurred despite the fact that trees from low-N montane and white-sand tropical sites were included in the analysis. If these are excluded, the average 15 N value for tropical trees increases to 4.72.1 (n = 65). The average concentration of nitrogen in leaves from tropical forests was 1.90.8 (n = 78), which was not signicantly different from the 1.60.5 (n = 28) average value found for temperate forests. When data are grouped by site, there is a signicant positive correlation between 15 N and N concentration (p < 0.007); sites with higher nitrogen concentration in their leaves tend to have higher 15 N values (Figure 2). Tropical forest soils were also much more 15 N-enriched than temperate forest soils (Table 2, Figure 3a), with tropical 15 N values averaging almost 8 higher, both at the surface and at depth. Variation with depth followed the classical pattern discussed by Nadelhoffer and Fry (1988); the average 15 N value for depth samples was approximately 2 higher than for surface samples in both tropical and temperate forests (Figure 3a). The comparison of nitrogen concentration of soil organic matter between tropical and temperate forests was difcult to make due to the small number of samples from temperate forest soils. With the available data, the average N concentration was smaller in the tropical soils than in the temperate forests for both surface and depth samples (Figure 3b). The results clearly showed that the 15 N values of tree leaves and soil organic matter were signicantly higher in tropical forests than in temperate forests. Therefore, the initial working hypothesis was conrmed, in that the results are consistent with tropical forests having a more open nitrogen cycle, with greater losses via fractionating pathways, suggesting that N is in relative excess in many moist tropical forests.
48
Table 1. 15 N () values of plant species. %N is the nitrogen concentration (%).
Species Metrosideros polymorpha Metrosideros polymorpha Metrosideros polymorpha Metrosideros polymorpha Metrosideros polymorpha Metrosideros polymorpha Cibotium glaucum Cibotium glaucum Cibotium glaucum Cibotium glaucum Cibotium glaucum Cibotium glaucum Amphirrox latifolia Licania hispidula Maquira guianensis Naucleopsis sp Naucleopsis sp Neea sp Protium carnosum Protium sp Protium sp Tachigalia cavipes Undetermined Undetermined Mixture Rinorea racemosa Inga sp Oxandra polyantha Nectandra amazonum Leonia racemosa Undetermined Sacoglotis sp Ficus glabrata Tichila sp Laetia crynbulosa Undetermined Undetermined Undetermined G.thophilurum Sapotaceae Lecitidaceae Virola surinasuensis Carapa guianensis Theobroma cacau Site Thurston Olaa Laupahoehoe Kohala Malokai Kauai Thurston Olaa Laupahoehoe Kohala Malokai Kauai Samuel Samuel Samuel Samuel Samuel Samuel Samuel Samuel Samuel Samuel Samuel Samuel Faz. Nova Vida Varzea Varzea Varzea Varzea Varzea Varzea Varzea Varzea Varzea Varzea Campina Campina Campina Campina Campina R.Ducke R. Ducke R. Ducke R. Ducke Region Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia rio Amazonas rio Amazonas rio Amazonas rio Amazonas rio Amazonas rio Amazonas rio Amazonas rio Amazonas rio Amazonas rio Amazonas Manaus Manaus Manaus Manaus Manaus Manaus Manaus Manaus Manaus Country U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil
15 N
%N 0.87 1.12 1.42 1.14 1.06 0.86 1.79 1.53 1.82 1.79 1.79 1.47 3.16 1.15 1.06 3.51 0.85 3.76 0.98 1.12 1.10 3.02 3.96 1.04 1.43 1.00 2.22 2.74 2.76 2.80 1.22 1.22 1.77 0.29 2.76 1.86 2.49 1.30
Ref 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 4 4 4 4 4 5
6.8 4.9 +0.9 2.2 2.3 0.5 9.3 6.0 +0.7 3.7 1.3 1.3 +6.8 +5.7 +6.4 +8.3 +6.5 +9.6 +5.7 +4.4 +5.7 +7.4 +6.4 +6.3 +8.0 +2.0 +1.2 +2.5 +3.7 +4.7 +4.0 +1.8 +1.3 +1.2 +2.4 2.2 0.2 +0.6 7.0 7.0 +5.8 +5.9 +7.0 +6.2
1.32
49
Table 1. Continued.
Species Melastoma bellucia Melastoma bellucia Melastoma bellucia Cecropia eucomona Cecropia eucomona Piper piperaceae Aegiphula scandens Rolinea exsucca Verbenaceae Alchornea schomburgkii Myrtaceae Compositae Rubiaceae Melastomataceae Flacourtia rukan Anacardium occidentale Zizyplus mauritana Terminalia catappa Lansium domesticum Ficus glberrima Crataeva erythrocarpa Kerangas scrub Kerangas scrub Kerangas forest Kerangas forest 8 ECM species 10 VAM species Quercus kelloggii Pinus sabiniana Pinus contorta ssp. Pinus albicaulis Pinus contorta ssp. Tsuga mertensiana Abies concolor Pinus ponderosa Prunus emarginata Salix scouleriana Picea abies Picea abies Picea abies Picea abies Picea abies Picea abies Site R. Ducke R. Ducke R. Ducke R. Ducke R. Ducke R. Ducke R. Ducke R. Ducke R. Ducke R. Ducke Cerrado Cerrado Cerrado Cerrado Toongfax Toongfax Toongfax Kwae Nakhon Pathom Kwae Kwae Region Manaus Manaus Manaus Manaus Manaus Manaus Manaus Manaus Manaus Manaus Bras`lia Bras`lia Bras`lia Bras` ilia Country Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Thailand Thailand Thailand Thailand Thailand Thailand Thailand Sarawak Sarawak Sarawak Sarawak Africa Africa U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. Germany Germany Germany Germany Germany Germany
15 N
%N 2.38 1.68 1.81 2.20 1.86 2.58 3.09 1.48 2.12 2.10 0.94 1.31 1.09 1.30 1.70 1.30 1.83 1.88 2.18 1.78 2.44
Ref 5 5 5 5 5 5 5 5 5 5 6 6 6 6 5 5 5 5 5 5 5 7 7 7 7 20 20 8 8 8 8 8 8 8 8 8 8 9 9 9 9 9 9
Korup Korup Mix Canyon Mix Canyon Grass Lake Carson Pass-3 Carson Pass-4 Carson Pass-5 Rice canyon Rice canyon Rice canyon Rice canyon Fichtelgebirge Fichtelgebirge Fichtelgebirge Fichtelgebirge Fichtelgebirge Fichtelgebirge
Cameroon Cameroon California California California California California California California California California California Bavaria Bavaria Bavaria Bavaria Bavaria Bavaria
+3.3 +4.6 +4.7 +5.9 +4.9 +4.6 +5.6 +2.0 +6.0 +1.7 +1.8 +0.1 0.6 +1.3 +4.5 +3.6 +2.5 +7.5 +6.0 +9.4 +8.0 2.3 7.4 2.4 3.5 +4.9 +4.6 +0.4 +0.8 +0.2 0.3 +0.4 +1.0 0.8 0.5 0.5 0.6 3.1 3.5 3.6 3.7 3.3 3.8
2.10 2.11
1.75 1.70 1.70 1.50 1.50 1.45
50
Table 1. Continued.
Species Picea abies Picea abies Picea abies Picea abies Picea abies Picea abies Picea abies Picea abies Picea abies Salix purpurea Betula verrucosa Salix purpurea Acer pseucoplatanus Acer pseudoplatanus Corylus avellana Fraxinus excelsior Abies alba Rubus sp Spruce-r Beech Cove hardwood Mixed hardwood Floodplain poplar Yellow poplar Pine Xeric oak Picea glauca Picea mariana Acer rubrum Cornus orida Liliodendron tulipifera Acer rubrum Cornus orida Liliodendron tulipifera Facus grandifolia Acer spp. Betula alleghaniensis Picea rubens Picea abis Juniperus communis Juniperus communis Betula nana Betula nana Pinus sylvestris Site Fichtelgebirge Fichtelgebirge Fichtelgebirge Fichtelgebirge Fichtelgebirge Fichtelgebirge Fichtelgebirge Fichtelgebirge Fichtelgebirge Col dOrnon Col dOrnon Col dOrnon Col dOrnon Col dOrnon Col dOrnon Col dOrnon Col dOrnon Col dOrnon Great Smoky Mt. Great Smoky Mt. Great Smoky Mt. Great Smoky Mt. Great Smoky Mt. Great Smoky Mt. Great Smoky Mt. Great Smoky Mt. Dalton Highway Dalton Highway Walker Branch Walker Branch Walker Branch Walker Branch Walker Branch Walker Branch Bear Brook Bear Brook Bear Brook Bear Brook Southern Sweden Subsite 1&2 Subsite 3 (boggy) Region Bavaria Bavaria Bavaria Bavaria Bavaria Bavaria Bavaria Bavaria Bavaria Northern Alps Northern Alps Northern Alps Northern Alps Northern Alps Northern Alps Northern Alps Northern Alps Northern Alps Tennessee Tennessee Tennessee Tennessee Tennessee Tennessee Tennessee Tennessee Alaska Alaska Tennessee Tennessee Tennessee Tennessee Tennessee Tennessee Maine Maine Maine Maine Sweden Scotland Scotland Northern Sweden Northern Sweden Northern Sweden Country Germany Germany Germany Germany Germany Germany Germany Germany Germany France France France France France France France France France U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. Sweden Scotland Scotland Sweden Sweden Sweden
15 N
%N 1.50 1.45 1.40 1.40 1.20 1.48 1.52 1.54 1.48
Ref 9 9 9 9 9 9 9 9 9 10 10 10 10 10 10 10 10 10 11 11 11 11 11 11 11 11 12 12 13 13 13 13 13 13 14 14 14 14 15 16 16 17 17 17
3.8 4.0 3.9 3.9 3.9 3.4 3.8 3.9 4.0 3.8 6.0 3.4 4.3 3.6 4.4 3.8 3.6 4.3 2.2 0.8 1.9 1.0 2.4 3.5 1.7 1.8 6.2 10.1 3.5 3.9 3.4 1.7 1.0 2.4 0.7 1.7 1.1 0.6 2.2 0.6 5.3 7.4 3.9 3.9
0.97 0.91
2.30 2.30 2.20 1.10 1.30 1.70 1.00 1.90 3.30
51
Table 1. Continued.
Species Picea abies Pseudotsuga menziesii Acer rubrum Quercus rubra Tsuga canadensis Acer rubrum Pinus strobus Betula papyrifera Facus grandifolia Prunus pensylvanica Betula lutea Pinus spp Pinus resinosa Quercus rubra Betula papyrifera Picea glauca Populus tremuliodes Fraxinus spp Betula papyrifera Acer rubrum Quercus macrocarpa Quercus rubra Cornus orida Acer rubrum Liliodendron tulipifera Site Region Central Sweden Oregon Massachusetts Massachusetts New Hampshire New Hampshire New Hampshire New Hampshire New Hampshire New Hampshire New Hampshire South Carolina Wiscosin Wiscosin Wiscosin Alaska Alaska Minnesota Minnesota Minnesota Minnesota Minnesota North Carolina North Carolina North Carolina Country Sweden U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A.
15 N
%N
Ref 18 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19
Andrews Harvard Forest Harvard Forest Harvard Forest Harvard Forest Harvard Forest Harvard Forest Harvard Forest Harvard Forest Harvard Forest North Inlet North Lakes North Lakes North Lakes Bonanza creek Bonanza creek Cedar Creek Cedar Creek Cedar Creek Cedar Creek Cedar Creek Coweeta Coweeta Coweeta
1.3 3.2 3.6 2.7 3.7 5.9 1.4 0.6 1.2 0.4 2.7 0.9 2.0 3.0 2.5 3.3 1.4 4.2 4.0 5.2 4.5 3.3 1.8 5.2 5.3
1 Vitousek PM (nonpublished data); 2 Almeida S (1995); 3 McClain M (nonpublished data); 4 Salati et al. (1982) 5 Yoneyama et al (1993); 6 Sprent et al. (1996); 7 Treseder K (nonpublished data); 8 Virginia and Delwiche (1982); 9 Gebauer and Schulze (1991); 10 Domenach et al. (1989); 11 Garten Jr and Miegroet (1994); 12 Schulze et al. (1994); 13 Garten Jr (1993); 14 Adelhoffer et al. (1995); 15 Nasholm et al. (1997); 16 Hill et al. (1996); 17 Michelsen et al. (1996); 18 Hogberg et al. (1996); 19 Fry (1991); 20 Hogberg and Alexander (1995)
Nutrient rich vs nutrient poor systems in the tropics Forests on white-sand soils in the tropics are considered to be nitrogenpoor systems (Vitousek & Sanford 1986; Cuevas & Medina 1988). Table 3 summarizes the elemental composition of leaves in contrasting vegetation types in the Amazon. Higher N concentrations were found in Varzea forest, followed by Terra-rme forests in Samuel and Manaus. The same trend was also found by Furch and Klinge (1989). As expected, the lowest concentrations were in the Campina forests. Following our initial hypothesis,
52
Table 2. 15 N values () for soil samples. Z is the soil depth (cm), %N is the nitrogen concentration (%) and Re stands for references.
Country Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Region Par a Par a Par a Par a Par a Par a Par a Par a Par a Par a Par a Par a Par a Par a Par a Par a a Par Par a Par a Par a Amazonas Amazonas Amazonas Amazonas Amazonas Amazonas Amazonas Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Soil type Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Red-yellow latosol (Hapludox) Red-yellow latosol (Hapludox) Red-yellow latosol (Hapludox) Red-yellow latosol (Hapludox) Red-yellow podzolic latosol (Kandiudult) Red-yellow podzolic latosol (Kandiudult) Red-yellow podzolic latosol (Kandiudult) Red-yellow podzolic latosol (Kandiudult) Red-yellow podzolic (Paleudult) Red-yellow podzolic (Paleudult) Red-yellow podzolic (Paleudult) Red-yellow podzolic (Paleudult) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) z 05 510 1020 2030 3040 4050 5060 6070 100110 140150 05 510 1020 2030 3040 4050 5060 6070 8090 120130 03 312 1236 3651 5180 80140 140 05 510 1020 2030 05 510 1020 2030 05 510 1020 2030 05 510 1020
15 N
%N 0.26 0.17 0.11 0.09 0.07 0.05 0.04 0.03 0.02 0.01 0.13 0.05 0.04 0.02 0.01 0.01 0.00 0.01 0.00 0.00 0.38 0.19 0.09 0.02 0.01 0.00 0.00 0.24 0.22 0.16 0.13 0.18 0.14 0.08 0.08 0.21 0.14 0.11 0.09 0.28 0.19 0.12
Re 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
+9.8 +10.8 +12.0 +12.5 +12.6 +13.2 +12.8 +12.9 +12.4 +11.9 +8.4 +9.2 +9.4 +9.9 +9.8 +9.3 +9.3 +8.5 +9.8 +8.6 +7.7 +9.8 +11.9 +14.9 +15.6 +20.0 +21.7 +9.8 +10.1 +10.5 +10.8 +8.6 +9.2 +9.9 +10.3 +10.7 +11.4 +11.9 +11.9 +9.3 +9.8 +11.3
53
Table 2. Continued.
Country Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil Brazil U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. Thailand Thailand Thailand Thailand Thailand U.S.A. U.S.A. U.S.A. U.S.A. Region Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Rondnia Par a Amazonas Paran a Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Hawaii Chachoengsao Chantaburi Rayong Chantaburi Chainat Wiscosin Wisconsin Maine Maine Soil type Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Yellow latosol (Hapludox) Red yellow podzolic latosol (Kandiudult) Red yellow podzolic latosol (Kandiudult) Red yellow podzolic latosol (Kandiudult) Red yellow podzolic latosol (Kandiudult) Red yellow podzolic latosol (Kandiudult) Red yellow podzolic latosol (Kandiudult) Red yellow podzolic latosol (Kandiudult) Red yellow podzolic latosol (Kandiudult) Red-yellow podzolic (Paleudult) Red-yellow podzolic (Paleudult) Red-yellow podzolic (Paleudult) Red-yellow podzolic (Paleudult) Red-yellow podzolic (Paleudult) Red-yellow podzolic (Paleudult) Red-yellow podzolic (Paleudult) Xanthic Ferralsols Xanthic Ferralsols Rhodic Ferralsols Volcanic tephra Volcanic tephra Volcanic tephra Volcanic tephra Volcanic tephra Volcanic tephra Volcanic tephra Volcanic tephra Volcanic tephra Volcanic tephra Volcanic tephra Volcanic tephra Gray podzolic Red yellow podzolic Red yellow podzolic Regosols Gray lowland Typic Hapludalfs Typic Hapludalfs Caribou Caribou z 2030 05 510 1020 2030 05 510 1020 2030 3040 4050 8090 120130 010 1020 2030 3040 4050 7080 90100
15 N
%N Re 0.08 0.09 0.07 0.05 0.04 0.12 0.10 0.06 0.04 0.04 0.04 0.06 0.04 0.07 0.04 0.02 0.02 0.01 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 5 5 6 6
+11.7 +6.4 +7.6 +9.2 +10.6 +11.2 +12.2 +13.3 +13.6 +13.1 +12.9 +12.4 +12.7 +10.5 +11.7 +11.9 +12.1 +12.2 +11.4 +9.8 +10.3 +7.4 010 +10.9 Surface 2.2 Surface 2.0 Surface +1.4 Surface 1.0 Surface 0.8 Surface +0.2 3040 +1.4 3040 0.4 3040 +6.2 3040 +4.7 3040 +4.6 3040 +5.1 015 +7.7 030 +7.1 010 +8.4 +3.7 010 +9.5 010 +2.6 +5.3 1020 +4.9 011.4 11.412.7 +8.5
0.02 0.03 0.06 0.04 0.08 0.25 0.08
54
Table 2. Continued.
Country U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. Japan Japan Belgium Belgium Belgium Belgium Belgium Belgium Belgium Belgium Belgium Belgium Belgium Belgium France France France France France France France U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. Germany Germany Germany Region Maine Maine Maine Maine Maine New Hampshire Central Illinois Oregon Oregon Tokyo Nagano Ardennes Ardennes Ardennes Ardennes Ardennes Ardennes Ardennes Ardennes Ardennes Ardennes Ardennes Ardennes Northern Alps Northern Alps Northern Alps Northern Northern Alps Alps Northern Alps Northern Alps Tennessee Tennessee Tennessee Tennessee Tennessee Tennessee Tennessee Tennessee Tennessee Tennesse Maine Bavaria Bavaria Bavaria Soil type Caribou Caribou Caribou Caribou Caribou Drummer silty clay loam 060 060 090 090 01 14 422 2243 4370 70100 01.5 1.510.5 1025 2551 5180 80110 01 26 611 1112 1214 1435 80 z 12.720.3 20.333.0 33.043.2 43.255.9 55.9101
15 N
%N
Re 6 6 6 6 6 7 7 8 8 9 9 10 10 10 10 10 10 10 10 10 10 10 10 11 11 11 11 11 11 11 12 12 12 12 12 12 12 12 13 13 14 15 15 15
Podzol Brown Forest Acid Brown Acid Brown Acid Brown Acid Brown Acid Brown Acid Brown Acid Brown Acid Brown Acid Brown Acid Brown
Ochreous podzolic Ochreous podzolic Ochreous podzolic Ochreous podzolic Ochreous podzolic Ochreous podzolic Ochreous podzolic
Olf Oh A0-5
+8.3 +8.0 +5.7 +3.7 +3.0 +9.6 +10.4 +2.1 +2.5 +5.0 +5.0 3.9 0.7 +0.7 +2.1 +2.6 +1.0 5.1 0.3 0.1 +1.3 +3.6 +1.0 4.0 2.8 +0.3 +2.8 +4.7 +5.1 +5.0 +3.6 +5.4 +3.9 +4.8 +5.1 +4.3 +5.5 +5.3 2.4 1.0 +2.3 3.5 +0.1 +2.9
1.57 1.39 1.21 0.84 0.20 0.17 0.21
55
Table 2. Continued.
Country Germany Sweden Sweden Sweden Sweden Sweden Sweden Sweden U.S.A. U.S.A. U.S.A. U.S.A. U.S.A. Region Bavaria Southern Sweden Northern Sweden Northern Sweden Northern Sweden Northern Sweden Northern Sweden Northern Sweden Massachusetts Massachusetts Massachusetts Minnesota Minnesota Soil type z A5-15 Oh
15 N
%N
Re 15 16 17 17 18 18 18 18 19 19 19 19 19
Oi Oa Oe A0-5 F H 1015 Oa/A 2025
+4.0 1.1 0.7 +0.5 2.0 0.4 +1.1 +5.0 +0.4 +4.0 +6.3 1.3 +5.3
1 Piccolo et al. (1996); 2 Yoneyama et al. (1993) 3 Vitousek et al. (1989); Vitousek PM (nonpublished data); 4 Yoneyama et al. (1990); 5 Nadelhoffer KJ and Fry B (1988); 6 Shearer et al. (1978); 7 Shearer et al. (1974); 8 Binkley et al. (1992); 9 Wada et al. (1984); 10 Riga et al. (1971); 11 Mariotti et al. (1980); 12 Garten Jr & Miegroet (1994); 13 Garten Jr CT (1993); 14 Nadelhoffer et al. (1995); 15 Gebauer and Schulze (1991); 16 Nasholm et al. (1997); 17 Michelsen et al. (1996); 18 Hogberg et al. (1996); 19 Fry (1991).
we expected that in forests on white sand soils, such as the Campina site, 15 N values would be lower than those from the relatively N-rich Terrarme and Varzea forests. Indeed, the only negative 15 N values among all lowland tropical forests occurred in the two white sand sites: the Brazilian Campina and the Kerangas site in Sarawak. The average value of these sites was signicantly lower (P < 0.001) than for any of the other tropical forests types we surveyed. Within the Terra-rme forests, the site which had higher elemental concentrations in leaves (Samuel) also showed an average 15 N value signicantly higher (P < 0.001) than the Terra-rme forest at Reserva Ducke in Manaus (Table 3). In contrast, the Varzea forest had the highest elemental concentration in leaves, but a signicantly lower 15 N in relation to the two Terra-rme forests. Isotopic values for surface soil samples (010 cm) were also signicantly lower in Varzea (15 N = +4.10.5, n = 17) than in Terrarme ( 15 N = +9.21.5, n = 10) sites. One possible cause for the lower 15 N values in the Varzea is the very high rate of nitrogen xation by legumes (and probably by Paspalum grasses) that occurs in this system (Martinelli et al. 1992); xed N has a 15 N value close to zero. In addition, Varzea soils are formed and renewed each year by sediments brought by the white-water rivers
56
Figure 1. Histogram of 15 N () for tree leaves collected in tropical and temperate forests. Solid bars are tropical sites and hatched bars are temperate sites. (b) Plot of 15 N () for tree leaves collected in tropical and temperate forests. Error bars represent one standard-deviation.
of the Amazon, and therefore substrate age differences between Varzea and Terra-rme sites may also contribute to the differences in 15 N. It is possible that the Varzea site has not been in place long enough for 15 N enrichment to occur to the same extent as in Terra-rme. This age effect is discussed in the next section.
57
Figure 2. Relationship between average 15 N values () and average nitrogen concentration (%) of foliar samples. Temperate sites are labeled TE, the unlabeled points are tropical sites. Bars are standard errors.
The Hawaii substrate age gradient and 15 N of plants and soils We used a developmental sequence of montane forests in the Hawaiian Islands to evaluate how losses of N by fractionating pathways could shape
Table 3. Average 15 N value () and average elemental composition of leaves (%) from distinct forest types in the Amazon Basin and averaged 15 N () of Kerangas site in Sarawak. Vegetation Varzea1 Terra-Firme2 Terra-Firme3 Campina4 15 N +2.5 +6.7 +4.5 3.2 N 2.19 1.90 1.84 1.11 P 0.19 0.08 0.05 0.05 K 1.23 0.77 0.43 0.37 Ca 1.34 0.77 0.43 0.37 Mg 0.36 0.33 0.29 0.26
1 Inundation forest Senna (1996); 2 Terra-rme forest at Samuel, Rondnia Almeida (1995); 3 Terra-rme forest at Manaus Klinge et al. (1984); 4 White-
sand soil forest (campina) near Manaus Klinge et al. (1984).
58
Figure 3. Box-whisker plot of (a) 15 N () and (b) nitrogen concentration (systems. The circles are average values, boxes are standard-errors, and bars represent one standarddeviation.
59 N values in forest ecosystems over time. The Hawaiian Islands result from the movement of the Pacic tectonic plate over a stationary convective plume or hot-spot in Earths mantle. The hot-spot is now located under the active volcanoes at the southeast edge of the Hawaiian chain; the age of the islands increases progressively to the northwest. Six sites with substrate ages of 300, 2100, 20,000, 150,000, 1.4 million, and 4.1 million years, located across the Islands, have been used in several soil and biogeochemical studies (Crews et al. 1995; Vitousek et al. 1997). All of the sites have basaltic parent material, are on constructional shield volcanic surfaces, at 1200 m elevation, and with 2500 mm annual precipitation. All are dominated by the native myrtaceous tree Metrosideros polymorpha; none have been cleared by humans. No symbiotic N xers are important components of any site. Fertilization studies show that productivity in the youngest sites is profoundly limited by low levels of N availability (Vitousek et al. 1993); N concentrations in plant tissues, inorganic N pools, rates of soil N transformations, and gaseous losses of N are also low in these sites (Riley & Vitousek 1995; Vitousek et al. 1995). By 20,000 years, more N has accumulated and N alone no longer limits production (Vitousek & Farrington 1997). Nitrogen concentration, transformations, and trace gas losses are all much greater than in the younger sites (Vitousek et al. 1995; Crews et al. 1995). Finally, plant production in the oldest site is limited by P (Herbert & Fownes 1995), while rates of N transformations and gaseous losses remain high (Riley & Vitousek 1995). We anticipated that if losses of N by fractionating pathways drive 15 N enrichment, then the youngest sites should have the lowest 15 N; they should be accumulating N from the atmosphere, with little N loss. As nitrication and losses of nitrate through leaching and/or denitrication increase through soil and ecosystem development, 15N in the systems as a whole should become enriched. We measured 15 N in the foliage of the dominant tree Metrosideros polymorpha, in the subcanopy tree fern Cibotium glaucum, and in surface and subsurface soils across the developmental sequence. In plants, we found the expected pattern of strongly depleted 15 N (to levels comparable to those in many temperate forests) in the youngest sites. Foliar 15 N was enriched by 810 by the 20,000 year-old site, where foliar N concentrations and several other measures of N transformation and loss peaked. Thereafter, 15 N became 13 more depleted, never approaching the very low levels observed in the youngest sites (Figure 4). 15 N values in soils yielded a similar pattern, although variation in soils across the sequence was less than that in plants. All soils were enriched relative to plants, and enriched at depth relative to the surface (Figure 4).
15
60
Figure 4. 15 N () variation for surface (open squares) and deeper (lled squares) soil samples, and for leaves of Metrosideros polymorpha (open circles) and Cibotium glaucum (lled circles) along a gradient of substrate age in Hawaii.
Overall, the observed pattern is consistent with the hypothesis that N losses by fractionating pathways cause plant and soil 15 N to become relatively enriched in sites where N supply is less- or nonlimiting to biological processes. In contrast, plant and soil 15 N is relatively depleted in N-limited areas.
15
N enrichment with soil depth
A common feature in soil proles is 15 N enrichment with soil depth. In order to compare different soil proles, we dened the value 15 N as the difference between the 15 N values of a soil depth and the 15 N values of the soil surface. Table 4 summarizes the largest 15 N values found for soil proles in tropical and temperate regions. In the tropical region, with the notable exception of the Manaus site, the 15 N values varied from 1.1 to 4.2, averaging 2.21.0 (n = 9). When the 15 N value of the Manaus site is included (14%), the average enrichment increases to 3.33.9 (n = 10). In the temperate sites, 15 N values varied from 2.7 to 9.1 with an average of 6.42.4 (n = 7). These differ signicantly at the 0.2 level, if the Manaus site is excluded. With the Manaus value included, the 15 N is still marginally higher in the temperate zone (p = 0.06). The causes for 15 N enrichment with depth are relatively well known (Wada et al. 1984; Nadelhoffer & Fry 1988; Piccolo et al. 1994; Piccolo et al. 1996). More intriguing is the different behavior for soil proles in the tropics versus temperate zones. Even among tropical soils,
61
Table 4. Isotopic enrichment with soil depth ( 15 N) for tropical and temperate regions. Depth of largest 15 N is the soil depth where the maximum isotopic enrichment occurred and Maximum depth is the deepest soil layer sampled. Region Site Depth of largest 15 N (cm) 4050 2030 140 2030 2030 2030 2030 2030 2030 3040 1112 4370 5180 1435 Horizon A (515 cm) Horizon A (05 cm) 1020 Maximum depth (cm) 150 130 140 30 30 30 30 30 130 100 100 100 110 80 Horizon Oh Horizon A 20
15 N
() 3.4 1.5 14.0 1.1 1.8 1.2 2.5 4.2 2.4 1.6 3.6 6.6 8.6 9.1 7.5 7.0 2.7
Tropical
Agua Parada Piqui Manaus Porto Velho Jamari Cacaulandia Vilhena-1 Vilhena-2 Nova Vida Benjamin Presque Isle Saint-Hubert Willerzie Chablais Bavaria Northern Sweden Madison
Temperate
15 N is dened as 15 N of depth i 15 N of surface .
the reason for the large 15 N value found in the Manaus prole is unclear; Piccolo et al. (1996) suggested that the very short dry season in the Manaus region makes losses of N higher in relation to other sites in the Amazon, where the dry season is more prolonged. However, the same explanation is unlikely when comparing tropical and temperate soils, since losses of nitrogen appear to be higher in tropical sites (Matson & Vitousek 1990; Keller et al. 1993; Neill et al. 1995). An alternative explanation could be based upon the effects of a higher cation exchange capacity and longer residence time of nitrogen in temperate soils. Higher exchange cation capacity would enhance the capability of the exchange complex to discriminate against 14 N, which would be preferentially lost from the prole, though the few measurements that exist suggest that ion exchange isotope effects may be minimal (B. Fry, pers. comm.). Finally, the observed differences between temperate and tropical soil proles may be partly a function of soil age.
62 Conclusions The 15 N values of plants and soils are readily measured, but not always readily interpreted. We hypothesized that if N functions as an excess nutrient in many lowland tropical forests, then losses of N by fractionating pathways would cause 15 N to become enriched in those forests. The results of our survey conrm this prediction: tropical forests in general are 15 N-enriched compared to temperate forests, and tropical forests in which N is abundant are 15 N-enriched compared to tropical forests in which N appears to be limiting. These 15 N results do not establish unequivocally that N is in excess in tropical forests; mechanisms other than N loss by fractionating pathways could cause part of the observed signal. For example, if N xation is relatively more important than atmospheric deposition in tropical forests, that could account for part although not all of the observed difference. However, the pattern of 15 N enrichment in tropical forests does provide an additional, time integrated line of evidence supporting the conclusion that many tropical forests are N-rich and have open N cycles in comparison to most temperate forests.
Acknowledgements We would like to thank Michael McClain for providing leaf samples from the Campina site, and Brian Fry for a critical review that improved an earlier draft of this manuscript.
References
Aber JD, Magill A, McNulty SG, Boone RD, Nadelhoffer RJ, Downs M & Hallett R (1995) Forest biogeochemistry and primary production altered by nitrogen saturation. Water Air and Soil Pollut. 85: 16651670 Almeida S (1995) Dinmica de nutrientes e variao natural de 13 C e 15 N de uma oresta tropical Hmida de terra-rme, Estao Ecolgica de Samuel, RO (BR). PhD Thesis. University of So Paulo, Piracicaba, Brazil Berendse F, Aerts R & Bobbink R (1993) Atmospheric nitrogen deposition and its impact on terrestrial ecosystems. In: Vos CC & Opdam P (Ed.) Landscape Ecology of a Stressed Environment (pp 104121). Chapman & Hall, London, England Binkley D, Sollins P, Bell R, Sachs D & Myrold D (1992) Biogeochemistry of adjacent alder and alder-conifer stands. Ecology 73: 20222033 Crews ...