57
(eDNA). Among these methods, isotope-based techniques, particularly those using
δ
13
C, are the easiest to perform and most straightforward and reliable methods. The
major groups of primary producers have specific and distinct δ
13
C ranges (Table 2.5),
and it has been empirically supported that the δ
13
C of organic matter is conserved
during diagenetic processes, even when the OC is converted into secondary organic
products (except lipids and hydrocarbons) by microorganisms. The δ
13
C signature is
particularly convenient for tracing seagrass-derived OC, because seagrasses have
the highest δ
13
C range among major marine and terrestrial primary producers
(Table 2.5). When sediment receives OC from multiple sources, the δ
13
C may be
used to evaluate the relative importance of each source using mixing models and
stochastic simulation (e.g., Minagawa 1992; Phillips et al. 2005; Parnell et al. 2010).
Typical applications of this approach have been reported by Kennedy et al. (2010)
and Miyajima et al. (2015). It should be noted that the reliability of estimates using
isotope mixing models may be reduced when the sources are too great in number or
not clearly distinguished from each other by the δ
13
C. Although the isotope ratios of
other biogenic elements, such as nitrogen (δ
15
N) and sulfur (δ
34
S), have been used
in combination with δ
13
C for source evaluation in food webs (Kharlamenko et al.
2001; Tewfik et al. 2005), these signatures are not conservative enough to be used
for source determination of sediment organic matter, as they are readily altered by
selective remineralization, immobilization, oxidation–reduction, and isotope
exchange reactions during early diagenetic processes (e.g., Francois 1987; Kohzu
et al. 2011; Möbius 2013).
Molecular biomarker approaches are generally not as convenient as isotope signatures for source estimation of OC stored in sediments long term, because many
molecular biomarkers are relatively unstable in an oxygenated aquatic environment
or easily degraded by the natural bacterial assemblage. Nonetheless, a few potential
biomarkers specific to seagrasses have been proposed, including dihydroxy fatty
acids (de Leeuw et  al. 1995), n-alkane-2-ones (Hernandez et  al. 2001; Xu et  al.
2006), and p-hydroxybenzoic acid (as a pyrolysis product; Kaal et al. 2016). Lignin
and lignin phenols contained in sediment provide a good proxy for OC derived from
vascular plants. Seagrasses also contain lignin, especially in their belowground
parts (Klap et  al. 2000). A combination of molecular biomarkers and
compound- specific isotope analysis would have much greater potential for
provenance analysis of sediment OC, because the use of isotopic signatures such as
δ
13
C allows a wider range of organic molecules to be used as seagrass-specific
tracers, including common fatty acids, sterols, and hydrocarbons. However,
application of this technique in seagrass meadow sediments is currently limited to
microbial trophic studies (Boschker et al. 2000; Bouillon and Boschker 2006).
The use of eDNA is more promising (Parducci et al. 2015), because short fragments of DNA can persist over a long period in sediments due to the sorptive preservation mechanism (Cai et  al. 2006). Recently, eDNA techniques were applied
successfully in the source evaluation of OC in seagrass meadows (Reef et al. 2017;
Hamaguchi et al. 2018). Also, Hamaguchi et al. (unpublished data) demonstrated
that their eDNA technique can be used to trace the fate of seagrass-derived OC
exported from seagrass habitats to the open ocean and deposited in pelagic
sediment.
2 Carbon Sequestration in Sediment as an Ecosystem Function of Seagrass Meadows
Précédent

- 67/378

Suivant