Chapter 9 Nutrients and Seagrasses
241
processes accounting for such losses are basically
three: (i) denitrification (only affecting N), (ii) sequestration in non-soluble minerals (only affecting
P), and (iii) burial in organic refractory compounds
(affecting both N and P). Seagrass beds are therefore
potential nutrient sinks, and this is relevant not only
for the seagrass ecosystem itself but also for the biogeochemical cycles of the elements and/or for the
nutrient dynamics of coastal ecosystems (see also
Marb` a et al., Chapter 6).
2. Denitrification
In denitrification, nitrate, nitrite, and nitrous oxide
are reduced by bacteria to N 2 (g) that fluxes off the
sediments to the atmosphere. The process takes place
in oxygen-depleted sediment layers, and requires an
adequate supply of nitrate, etc, and organic matter. Nitrifying aerobic bacteria oxidizing ammonium
(typically derived from biotic activity, i.e. ammonification) are recognized as the main agents providing these oxidised nitrogen sources in littoral marine
sediments (Seitzinger, 1988). Since denitrification
requires a close coupling of aerobic and anaerobic
processes, it would seem that the seagrass rhizosphere, where oxic microlayers caused by oxygen
release from roots intermix with hypoxic or anoxic
zones (Kuo and den Hartog, Chapter 3), constitutes
an optimal environment where it can take place
(Hemminga and Duarte, 2000). Indeed, some evidence indicates that denitrification is higher in seagrass beds than in neighboring bare sediments (e.g.
Caffrey and Kemp, 1990; Miyajima et al., 2001),
but this is not always the case (Boon et al., 1986b).
However, reported denitrification rates in seagrass
sediments are highly variable (Caffrey and Kemp,
1990; Morell and Corredo, 1993; Blackburn et al.,
1994; Rysgaard et al., 1996; Shieh and Yang, 1997;
Welsh et al., 2000; species concerned: Syringodium
sp., Halodule sp., and Thalassia sp., Thalassia
hemprichii, Halodule beaudettei, Halodule uninervis, Zostera marina, Zostera noltii), with values
ranging, generally, from 0.4 to 6 µmol N m
−2 h
−1 ,
but up to 8–30 µmol N m
−2 h
−1 , in eutrophicated lagoons with Zostera capricorni (Eyre and Ferguson,
2002) and even to 150 µmol N m
−2 h
−1 in some
cases (Hemminga and Duarte, 2000). Such rates are,
on average, lower than those found in coastal sediments without seagrass cover, where nitrogen efflux
is in the range 50–250 µmol N m
−2 h
−1 (Seitzinger,
1988; Kim et al., 1997; Valiela and Cole, 2002).
Estimation of denitrification rates faces several
problems, among which the choice of the methodology (acetylene-block vs. isotope pairing techniques,
see Middleburg et al., 1996, Welsh et al., 2001),
and the variability (in space and time) of the processes involved, making it difficult to produce accurate annual estimates. This has to be considered
when interpreting results, or when attempting comparisons. However, controls on denitrification are
probably much more complex in seagrass sediments
than elsewhere. Here, nitrate will most certainly be
the limiting resource for denitrification, and nitrate
supply will be determined by two factors: ammonium availability and nitrification, which in turn will
depend on oxygen concentration. Competition for
pore water ammonium among bacteria, microalgae,
and seagrass roots (Welsh et al., 2000) can limit ammonium supply. Moreover, it is known that oxygen
release from roots strongly varies among species and
seasons. Finally, a part (sometimes substantial, Caffrey and Kemp, 1990) of the nitrate being reduced
ends as ammonium, instead of as N 2 . In any case,
it would seem reasonable to hypothesize that high
denitrification will only occur where (and when)
ammonification exceeds ammonium primary producers’ demand and where (and when) oxygen efflux from roots can sustain a significant nitrifying
activity.
It is clear that the role of denitrification in seagrass beds needs to be further elucidated in the near
future. Tentatively, and compared to other important
N fluxes reviewed in this chapter (export, storage
in the sediment, etc), it would seem that, in general terms, the relative importance of denitrification
in the N budget of seagrass beds is from modest to
moderate. However, much caution should be applied
in this respect.
3. Phosphorus
A large amount of P occurs in inorganic form (Jensen
et al., 1998). This inorganic P can be found in two
forms: bound to oxidized Fe and bound to Ca. In
turn, Ca-bound P can be chemisorbed to CaCO 3 or
in the form of apatites. Fe-bound phosphorus is often
called exchangeable P, as it can redissolve following
chemical changes (e.g. decrease in redox potential
and increase in acidity) in the sediment, while P is
almost irreversibly bound to Ca when apatites have
been formed. Some dissolution of the carbonate matrix takes place in the root layer (Jensen et al., 1998).
241
processes accounting for such losses are basically
three: (i) denitrification (only affecting N), (ii) sequestration in non-soluble minerals (only affecting
P), and (iii) burial in organic refractory compounds
(affecting both N and P). Seagrass beds are therefore
potential nutrient sinks, and this is relevant not only
for the seagrass ecosystem itself but also for the biogeochemical cycles of the elements and/or for the
nutrient dynamics of coastal ecosystems (see also
Marb` a et al., Chapter 6).
2. Denitrification
In denitrification, nitrate, nitrite, and nitrous oxide
are reduced by bacteria to N 2 (g) that fluxes off the
sediments to the atmosphere. The process takes place
in oxygen-depleted sediment layers, and requires an
adequate supply of nitrate, etc, and organic matter. Nitrifying aerobic bacteria oxidizing ammonium
(typically derived from biotic activity, i.e. ammonification) are recognized as the main agents providing these oxidised nitrogen sources in littoral marine
sediments (Seitzinger, 1988). Since denitrification
requires a close coupling of aerobic and anaerobic
processes, it would seem that the seagrass rhizosphere, where oxic microlayers caused by oxygen
release from roots intermix with hypoxic or anoxic
zones (Kuo and den Hartog, Chapter 3), constitutes
an optimal environment where it can take place
(Hemminga and Duarte, 2000). Indeed, some evidence indicates that denitrification is higher in seagrass beds than in neighboring bare sediments (e.g.
Caffrey and Kemp, 1990; Miyajima et al., 2001),
but this is not always the case (Boon et al., 1986b).
However, reported denitrification rates in seagrass
sediments are highly variable (Caffrey and Kemp,
1990; Morell and Corredo, 1993; Blackburn et al.,
1994; Rysgaard et al., 1996; Shieh and Yang, 1997;
Welsh et al., 2000; species concerned: Syringodium
sp., Halodule sp., and Thalassia sp., Thalassia
hemprichii, Halodule beaudettei, Halodule uninervis, Zostera marina, Zostera noltii), with values
ranging, generally, from 0.4 to 6 µmol N m
−2 h
−1 ,
but up to 8–30 µmol N m
−2 h
−1 , in eutrophicated lagoons with Zostera capricorni (Eyre and Ferguson,
2002) and even to 150 µmol N m
−2 h
−1 in some
cases (Hemminga and Duarte, 2000). Such rates are,
on average, lower than those found in coastal sediments without seagrass cover, where nitrogen efflux
is in the range 50–250 µmol N m
−2 h
−1 (Seitzinger,
1988; Kim et al., 1997; Valiela and Cole, 2002).
Estimation of denitrification rates faces several
problems, among which the choice of the methodology (acetylene-block vs. isotope pairing techniques,
see Middleburg et al., 1996, Welsh et al., 2001),
and the variability (in space and time) of the processes involved, making it difficult to produce accurate annual estimates. This has to be considered
when interpreting results, or when attempting comparisons. However, controls on denitrification are
probably much more complex in seagrass sediments
than elsewhere. Here, nitrate will most certainly be
the limiting resource for denitrification, and nitrate
supply will be determined by two factors: ammonium availability and nitrification, which in turn will
depend on oxygen concentration. Competition for
pore water ammonium among bacteria, microalgae,
and seagrass roots (Welsh et al., 2000) can limit ammonium supply. Moreover, it is known that oxygen
release from roots strongly varies among species and
seasons. Finally, a part (sometimes substantial, Caffrey and Kemp, 1990) of the nitrate being reduced
ends as ammonium, instead of as N 2 . In any case,
it would seem reasonable to hypothesize that high
denitrification will only occur where (and when)
ammonification exceeds ammonium primary producers’ demand and where (and when) oxygen efflux from roots can sustain a significant nitrifying
activity.
It is clear that the role of denitrification in seagrass beds needs to be further elucidated in the near
future. Tentatively, and compared to other important
N fluxes reviewed in this chapter (export, storage
in the sediment, etc), it would seem that, in general terms, the relative importance of denitrification
in the N budget of seagrass beds is from modest to
moderate. However, much caution should be applied
in this respect.
3. Phosphorus
A large amount of P occurs in inorganic form (Jensen
et al., 1998). This inorganic P can be found in two
forms: bound to oxidized Fe and bound to Ca. In
turn, Ca-bound P can be chemisorbed to CaCO 3 or
in the form of apatites. Fe-bound phosphorus is often
called exchangeable P, as it can redissolve following
chemical changes (e.g. decrease in redox potential
and increase in acidity) in the sediment, while P is
almost irreversibly bound to Ca when apatites have
been formed. Some dissolution of the carbonate matrix takes place in the root layer (Jensen et al., 1998).