Soil Phosphorus Forms Along a Strong Nutrient Gradient in a Tropical Ombrotrophic Wetland
Total Page:16
File Type:pdf, Size:1020Kb
Wetland Soils Soil Phosphorus Forms along a Strong Nutrient Gradient in a Tropical Ombrotrophic Wetland Phosphorus cycling influences productivity and diversity in tropical wetlands, Alexander W. Cheesman* yet little is known about the forms of P found in the accreting organic matter Wetland Biogeochemistry Laboratory of these ecosystems. We used alkaline (NaOH–ethylenediamine tetraacetic Soil and Water Science Dep. acid [EDTA]) extraction and solution 31P nuclear magnetic resonance (NMR) Univ. of Florida 106 Newell Hall spectroscopy to characterize P in surface soils across a strong nutrient gradi- Gainesville, FL 32611 ent within a tropical ombrotrophic peat dome. From the interior bog plain to the marginal Raphia taedigera swamp, total soil P increased from 14.6 and to 70.9 g m−3 and resin-extractable P from 0.1 to 30 mg kg−1. Phosphatase Smithsonian Tropical Research Institute activity declined across the same transect (364–46 mmol methylumbellifer- Apartado 0843-03092 one kg−1 min−1), indicating an increase in P availability toward the periph- Balboa, Ancón, Republic of Panama ery of the wetland. Organic P identified by solution 31P NMR spectroscopy Benjamin L. Turner included phosphomonoesters (12–17%), phosphodiesters (10–14%), and Smithsonian Tropical Research Institute phosphonates (up to 3.3% of total P). Inositol phosphates were not detected Apartado 0843-03092 in these acidic peats. Inorganic P forms included orthophosphate (9–25% of Balboa, Ancón total P), pyrophosphate (up to 3%), and long-chain polyphosphates; the latter Republic of Panama occurred in concentrations (up to 24% of total soil P) considerably higher than previously found in wetland soils. The concentration of residual (unex- K. Ramesh Reddy tractable) P was similar among sites (mean 280 mg kg−1), resulting in an Wetland Biogeochemistry Lab. increase in its proportion of the total soil P from 29% at the P-rich margins to Soil and Water Science Dep. 55% at the P-poor interior. This is the first information on the P composition Univ. of Florida of tropical wetland soils and provides a basis for further study of the cycling 106 Newell Hall and contribution of P forms to the nutrition of plants and microorganisms. Gainesville, FL 32611 Abbreviations: AEM, anion exchange membrane; EDTA, ethylenediamine tetraacetic acid; NMR, nuclear magnetic resonance. hosphorus is a key element limiting ecosystem processes in freshwater wetlands (Daniel et al., 1998; Rejmánková, 2001). Because much of the P in wetland soils occurs in organic forms (Davelaar, 1993; Newman and PRobinson, 1999; Reddy et al., 2005), P availability is dependent on the cycling of P from biological material. Biologically derived P in soils, however, represents a va- riety of compounds that differ markedly in their behavior and bioavailability (Celi and Barberis, 2005; Condron et al., 2005). The nature of these P forms is the result of the dynamic interplay among biological inputs (Makarov et al., 2005), abiotic stabilization (Celi and Barberis, 2005), and biological modification (Cheesman et al., 2010b), all of which are influenced by edaphic, environmental, and biological factors. Information on the role these factors have in determining P composition in wetland soils is therefore critical for understanding the cycling and productivity of wetland systems, as well as in explaining future responses to perturbations. Unlike quantitative extraction and determination of P pools based on chemi- cal stability, techniques such as solution 31P NMR spectroscopy allow the assess- ment of actual P forms present in environmental samples (Cade-Menun, 2005; McKelvie, 2005; Turner et al., 2005). This has yielded valuable insight into the na- Soil Sci. Soc. Am. J. doi:10.2136/sssaj2011.0365 Received 28 Oct. 2011. *Corresponding author ([email protected]). © Soil Science Society of America, 5585 Guilford Rd., Madison WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher. Soil Science Society of America Journal ture of P forms and their biogeochemical processing in wetlands. visible soil–vegetation catena, however, with current communi- For example, previous work has highlighted the diverse range of ties ranging from a monodominant Raphia taedigera Mart. palm biogenic P forms that can be present in wetland soils (Sundaresh- swamp at the periphery, through mixed and monodominant war et al., 2009), the importance of site conditions in determin- Campnosperma panamense Standl. swamp forest, to a central ing P composition (Cheesman et al., 2010b), and differences in “bog plain” community dominated by herbaceous species and P composition between wetland and terrestrial soils (Turner and stunted trees (Phillips et al., 1997; Sjögersten et al., 2011). Newman, 2005; Turner et al., 2006). Yet despite clear evidence that P availability influences both vegetation (Hagerthey et al., Sampling 2008; Sjögersten et al., 2011) and biogeochemical processes such We established nine sampling sites at 300-m intervals along as C fluxes (Wright et al., 2011) and N2 fixation (Šantrůčková a linear transect running perpendicularly from a bordering canal et al., 2010), little is known about P forms and their associated toward the geodesic center of the peat dome (Fig. 1; Table 1) (Co- cycling in tropical wetlands. hen et al., 1989; Sjögersten et al., 2011; Troxler, 2007). Sample Tropical peat domes are “self-emergent” organic wetlands Site 9 could not be located in the very center of the bog plain due within the humid tropics (Semeniuk and Semeniuk, 1997). to practical constraints but was considered to be just inside the Their upper surface shows a pronounced convex morphology central vegetation zone (Myrica–Cyrilla bog plain). In September leading to (or resulting from) their hydrologic isolation and an 2007, three replicate peat samples were collected from within 20 ombrotrophic hydrologic regime (Anderson, 1983; Andriesse, m of each site using a sharpened metal cutting head on a ridged 1988; Belyea and Baird, 2006). They are systems of environ- polycarbonate tube. Each replicate was an amalgamated sample mental, social, cultural, and economic importance, providing of three surface cores (7.5-cm diameter, 10-cm depth) collected numerous direct ecosystem services to local populations (Cen- from within 2 m of each other. Samples were immediately cooled tral American Commission for Environment and Development, (~10°C) and returned to the laboratory, where they were stored 2002; Ellison, 2004) as well as representing substantial and dy- at 4°C until processing within 72 h of collection. This processing namic pools in the global C cycle (Maltby and Immirzi, 1993; involved homogenization and the removal by hand of recogniz- Yu et al., 2010). Often associated with the swamps of maritime able roots (>1-mm diameter) and lignified structures (seeds, twigs, Southeast Asia (Anderson, 1983; Anderson and Muller, 1975), etc.). Due to the high concentration of fine roots from herbaceous significant peat deposits also occur throughout the Caribbean vegetation at Sites 8 and 9, however, a substantial amount of root coastal plain (Ellison, 2004; Phillips and Bustin, 1996). Al- biomass may have remained within these samples. The samples though these remain poorly studied in comparison with their were subsequently split, with half being air dried (~22°C for 10 Asian counterparts, there are strong similarities, including a vis- d) to constant weight and the remainder stored at 4°C in sealed ible soil–vegetation catena across the convex surface (Phillips et bags (subsequently referred to as fresh sample). Air-dried samples al., 1997) and distinct gradients in nutrient availability (Sjöger- were ground in a ball mill using tungsten carbide vessels and stored sten et al., 2011; Troxler, 2007). in airtight containers under ambient laboratory conditions until We used solution 31P NMR spectroscopy to assess the analysis. Additional soil was collected from Site 9 in November functional forms of P present within the surface soils of a tropi- 2007 for 31P NMR spectroscopy, for which the total P was not cal ombrotrophic wetland. By using a natural soil P gradient and significantly different from the September sample at this sitet ( - range of vegetation types while controlling for more general site test: P > 0.05). conditions, we aimed to derive novel information on the influ- ence of nutrient status on the chemical nature of soil P found in Soil Properties tropical wetlands. Such data are critical to our understanding of Soil moisture was determined by gravimetric loss after dry- P cycling in wetlands, particularly given the extent to which an- ing at 105°C for 24 h and used to estimate the soil bulk density. thropogenic P enrichment affects such ecosystems (Cheesman Soil pH was determined on fresh soil using a 1:20 soil (oven-dry et al., 2010b; McDowell, 2009; Turner et al., 2006). weight)/water ratio and a glass electrode. Total elemental concen- trations were determined on dried and ground samples. Total soil METHODS C and N were determined by combustion and gas chromatogra- Study Site phy using a Flash EA1112 (Thermo Scientific), while total P was 2 The Changuinola peat deposit is an approximately 80-km determined by H2O2 +