Variability in the origin of carbon substrates for bacterial communities in mangrove sediments

FEMS Microbiology Ecology 49 (2004) 171–179 www.fems-microbiology.org Variability in the origin of carbon substrates for bacterial communities in mangrove sediments a c a,* , Tom Moens b, Nico Koedam c, Farid Dahdouh-Guebas c, Willy Baeyens a, Frank Dehairs a Department of Analytical and Environmental Chemistry, Mangrove Management Group, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium b Department of Biology, Marine Biology Section, Ghent University, Krijgslaan 281/Sterre S8, B-9000 Gent, Belgium Department of General Botany and Nature Management, Mangrove Management Group, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium Received 3 December 2003; received in revised form 24 February 2004; accepted 4 March 2004 First published online 23 March 2004 Abstract Organic carbon in mangrove sediments originates from both local sources (mangroves, microphytobenthos) and tidal inputs (e.g. phytoplankton, seagrass-derived material). The relative inputs of these sources may vary strongly, both within and between different mangrove sites. We combined elemental (TOC/TN) and bulk d13 C analysis on sediment cores from various mangrove sites with d13 C data of bacteria-specific phospholipid fatty acids (PLFA) in order to identify the dominant carbon substrates used by in situ bacterial communities. d13 C values of each of these markers showed a range of 10‰ or more across the different sites and sampling depths, but generally followed the d13 C trend observed in bulk organic carbon. Several sediment cores show a strong vertical gradient in PLFA d13 C, suggesting a selectivity for algal-derived carbon in the surface layers. Our data demonstrate that the origin of bacterial carbon substrates varies widely across different mangrove sites, and imply that data on mineralization of organic matter cannot be directly incorporated in ecosystem carbon budgets without an estimation of the contribution of various sources. Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. Keywords: Intertidal sediments; Microbial carbon dynamics; GC-IRMS; Mangrove; Phospholipid fatty acids; Mineralization 1. Introduction Mangroves are known as highly productive ecosystems (global litterfall of 100 Tg C y
1 [1]) and recent estimates that as much as 11% of the total organic C inputs across the land–ocean interface (i.e. through riverine transport) is of mangrove origin show that the carbon fixed by mangroves is potentially significant in the carbon budget of the coastal zone [1]. Such estimates are not well constrained, however, and our understanding of the ecological fate of these large quantities of organic matter is far from complete. It has become clear that a strong interaction exists between the inter* Corresponding author. Tel.: +32-62912665; fax: +32-6293274. E-mail address: (S. Bouillon). tidal zone and the adjacent aquatic environment, i.e. both import (of terrestrial material, phytoplankton, seagrasses, etc.) and export (of mangrove-derived organic matter) which are expected to have major consequences for the carbon dynamics in both compartments [2–4]. The imported carbon sources have an important trophic role in sustaining macro-invertebrate communities in the intertidal zone [5] – at least in estuarine systems where tidal import is significant. Mineralization, however, could represent a major fate for exported organic matter [6,7], but its role remains insufficiently quantified. Similarly, the few available estimates [8–11] indicate that intense mineralization takes place in intertidal mangrove sediments. However, no data are currently available that directly identify the carbon sources sustaining these microbial communities. In 0168-6496/$22.00 Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.femsec.2004.03.004 Steven Bouillon 172 S. Bouillon et al. / FEMS Microbiology Ecology 49 (2004) 171–179 2. Materials and methods 2.1. Study sites and sample collection Sediment cores were collected in November and December 2002 in three mangrove systems: (1) Pambala– Chilaw lagoon, western Sri Lanka (07°350 N–79°470 E), (2) the estuarine mangroves of Pichavaram, south-east India (11°270 N, 79°170 E), and (3) a small fringing riverine mangrove stand along the Chunnambar river (11°530 N, 79°480 E), south-east India. Pambala lagoon can be characterized as a low-tidal amplitude system (tidal variations rarely exceeds 0.2 m and much of the forest is rarely inundated [19]), with 3.5 km2 of fringing mangroves. Surface water salinity varies strongly both seasonally and spatially, between 0 and 55‰. At this site, one sediment core was taken in each of five different forest zones, each characterized by a different dominant mangrove species, i.e. Rhizophora mucronata, Bruguiera gymnorrhiza, Avicennia officinalis, Excoecaria agallocha, and Lumnitzera racemosa. One additional core was taken at a recently converted site, which was formerly an unvegetated mudflat but is now dominated by (10 yr old) R. mucronata. The Pichavaram mangroves are located 250 km south of Chennai (Tamil Nadu, India) in the estuaries of the Vellar and Coleroon. Tidal range at this site is reported to be 0.5– 1.0 m. One sediment core was collected at a site dominated by A. officinalis, and one at a mixed species stand close to a creek border. A sediment core was also collected at a small stand of fringing mangroves (<10 m width) along the Chunnambar river (between Pondicherry and Cuddalore, Tamil Nadu), with mostly A. officinalis and R. mucronata. We have no precise data on the tidal range at this site, but expect it to be similar to that in the nearby Pichavaram mangroves (i.e. 0.5–1 m). Unfortunately, data on mangrove biomass or productivity are lacking for all the mangrove sites studied here. All cores from Pambala and Pichavaram were subsectioned at 0–1, 1–2, 2–4, and 4–10 cm. For the site in Chunnambar, only three intervals were taken: 0–1, 1–3, and 3–5 cm. Samples were stored frozen and transported on dry ice back to the home laboratory where they were freeze–dried and stored frozen until further processing. 2.2. Analytical procedures Sediments were ground to a fine powder using a mortar and pestle. Subsamples for elemental (TOC/TN, atom) and bulk stable isotope composition were acidified with dilute (5%) HCl before analysis to remove carbonates. Concentrations of organic carbon and total nitrogen, and TOC/TN ratios were determined by combusting preweighed samples in a ThermoFinnigan Flash1112 elemental analyzer; acetanilide (Merck) was used for calibration. d13 C analysis of TOC was performed with the aforementioned elemental analyzer, linked with a ThermoFinnigan delta +XL via a Conflo III interface, with a typical reproducibility of 0.15‰. Extraction and derivatization of PLFA for compound-specific d13 C analysis followed the general outline of [12]. Briefly, 2–6 g of sediment (dry weight) (...truncated)