This article was downloaded by: [Niwa] On: 30 April 2014, At: 21:14 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
New Zealand Journal of Marine and Freshwater Research Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tnzm20
Mangrove (Avicennia marina subsp. australasica) litter production and decomposition in a temperate estuary a
RV Gladstone-Gallagher , CJ Lundquist
bc
& CA Pilditch
a
a
Department of Biological Sciences, University of Waikato, Hamilton, New Zealand b
National Institute of Water and Atmospheric Research Ltd (NIWA), Hamilton, New Zealand c
Institute of Marine Science, University of Auckland, Auckland, New Zealand Published online: 16 Sep 2013.
To cite this article: RV Gladstone-Gallagher, CJ Lundquist & CA Pilditch (2014) Mangrove (Avicennia marina subsp. australasica) litter production and decomposition in a temperate estuary, New Zealand Journal of Marine and Freshwater Research, 48:1, 24-37, DOI: 10.1080/00288330.2013.827124 To link to this article: http://dx.doi.org/10.1080/00288330.2013.827124
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &
Downloaded by [Niwa] at 21:14 30 April 2014
Conditions of access and use can be found at http://www.tandfonline.com/page/termsand-conditions
New Zealand Journal of Marine and Freshwater Research, 2014 Vol. 48, No. 1, 2437, http://dx.doi.org/10.1080/00288330.2013.827124
RESEARCH ARTICLE Mangrove (Avicennia marina subsp. australasica) litter production and decomposition in a temperate estuary RV Gladstone-Gallaghera*, CJ Lundquistb,c and CA Pilditcha a Department of Biological Sciences, University of Waikato, Hamilton, New Zealand; bNational Institute of Water and Atmospheric Research Ltd (NIWA), Hamilton, New Zealand; cInstitute of Marine Science, University of Auckland, Auckland, New Zealand
Downloaded by [Niwa] at 21:14 30 April 2014
(Received 15 May 2013; accepted 5 July 2013) Mangrove forests can provide important cross-boundary subsidies of organic matter to adjacent habitats through the production, export, decomposition and assimilation of litter. We quantified two of these components in a temperate mangrove forest in Whangamata Harbour, New Zealand: 1) litter production; and 2) decomposition rates as a function of tidal elevation, sediment type and burial depth. Litter traps sampled monthly for a year measured an annual detrital input of 3.24 5.38 t DW ha 1, of which 77% occurred in summer. Decomposition rates depended on litter type, with leaves decomposing faster (63 d to decay by 50%) than pneumatophore and wood material (316 and 460 d, respectively). Buried leaf and wood litter decomposed 1.31.4 times slower than litter on the sediment surface; however, tidal elevation and sediment type (mud vs. sand) had no effect. The slow decay of litter (an order of magnitude slower than tropical mangrove litter) suggests that its incorporation into temperate marine food webs may be relatively slow. Keywords: detritus; New Zealand; organic matter; litterfall; degradation; temperate mangroves; productivity; decay
Introduction Mangroves can represent an important crossboundary subsidy of organic matter to coastal and continental shelf food webs, through the production of litterfall, its export and subsequent decomposition (Lugo & Snedaker 1974; Saenger & Snedaker 1993; Polis et al. 1997). The magnitude of this subsidy is, in part, a function of litterfall production rates, as well as the rate of degradation into biologically available detritus (Lugo & Snedaker 1974; Rice & Tenore 1981; Robertson 1988). Litterfall production rates decrease with increasing latitude (alongside reduced tree height and tree species diversity; Twilley et al. 1992; Saenger & Snedaker 1993; Harty 2009), but are also influenced by site-specific environmental conditions *Corresponding author. Email:
[email protected] # 2013 The Royal Society of New Zealand
(Clarke 1994). Similarly, litter decomposition rates decrease with increasing latitude (Mackey & Smail 1996), but can vary within a system due to tidal submersion period, sediment properties and macrofaunal community (Robertson 1988; Dick & Osunkoya 2000; Holmer & Olsen 2002; On˜ate-Pacalioga 2005; Proffitt & Devlin 2005). Assessing the potential role of mangroves in cross-boundary subsidies therefore requires site-specific data. To date, most studies have focused on tropical mangroves and there is a paucity of data from temperate systems (latitudes308 N and S; reviewed by Morrisey et al. 2010). New Zealand has a single mangrove species, Avicennia marina subsp. australasica, which can be found in estuaries ranging from Ohiwa Harbour (38803?S;
Downloaded by [Niwa] at 21:14 30 April 2014
Temperate mangrove litter production and decomposition 25 southern limit) to Northland (34827?S; Harty 2009). However, litterfall production and decomposition studies have been restricted largely to the Auckland region (Albright 1976; Woodroffe 1982, 1985; On˜ate-Pacalioga 2005; Burns et al. unpublished data, as cited in Morrisey et al. 2010), with only one study from outside this area (Northland; May 1999). In many estuaries, forest area is increasing (Morrisey et al. 2010), and there is a need to assess rates at which litter enters local ecosystems. In response to forest area increases, consents for large-scale mangrove removals (e.g. c. 110 ha in Tauranga Harbour) have been granted in New Zealand. Estimates of decomposition rates are needed to help predict recovery rates following mangrove removal measures that leave mulched and sometimes buried vegetation in situ to decompose (Harty 2009; Lundquist et al. 2012). The goal of this study was to measure seasonal variations in litterfall and, for the first time in a single study, the effects of multiple environmental variables on its degradation. We also examined decomposition rates of mangrove wood and pneumatophores, since removal methodologies may leave behind significant quantities of this material. In the Whangamata estuary, we measured litterfall accumulation rates every month for a year in mature and newly established forest to estimate the potential contribution to crossboundary subsidies. Decomposition rates of leaves, pneumatophores and woody material were measured as a function of tidal elevation, because submergence is known to accelerate decay (e.g. Robertson 1988; Dick & Osunkoya 2000). We also buried material (which occurs when litter is retained within the forest) to test whether anoxic conditions slowed decomposition rates, as has been demonstrated elsewhere (Albright 1976; Van der Valk & Attiwill 1984; Fourqurean & Schrlau 2003). While the importance of litter burial in temperate forests has not been investigated, in tropical mangrove systems, the burial and incorporation of litter in the sediments is an important pathway of organic matter recycling within the forest (e.g. Bouillon
et al. 2004; Kristensen et al. 2008). Furthermore, some mangrove removal methods result in the burial of mulched decomposing litter (Lundquist et al. 2012). The carbon and nitrogen content of leaf litter during decomposition was measured to see how conditions influenced nutritional value to consumers (Fell et al. 1984; Nordhaus et al. 2011). Finally, because decomposition rates may be altered by sediment properties and associated changes in faunal composition (Bosire et al. 2005; On˜ate-Pacalioga 2005; Proffitt & Devlin 2005), we compared decay in permeable sandy sediment and nearby cohesive muddysand. Overall, this study makes a contribution to the relatively limited information available on the production of mangrove detritus and the potential environmental factors affecting its decomposition in a temperate estuary. Materials and methods Study site The study was conducted at two sites in Whangamata Harbour, North Island, New Zealand. Mangroves cover an area of 101 ha (approximately 22% of the harbour area), which has increased from 31 ha since the 1940s (Singleton 2007). The mangrove forest where our study sites were located covered an area of 27 ha and extended approximately 200 m out from the shoreline onto the intertidal flat. On the intertidal flat seaward of the forest, site 1 (sand; 37810?43.4ƒS, 175851?37.4ƒE) was located approximately 50 m from site 2 (muddy sand; 37810?39.9ƒS, 175851?36.8ƒE). Macrofaunal community structure was different between the two sites (measured at the mid-tidal position), driven primarily by differences in the relative abundances of some polychaete species (GladstoneGallagher 2012). Sediment at site 1 had a lower mud (particles B63 mm) content (14.4% vs. 29.9%) and a greater median grain size (197.6 mm vs. 130.8 mm) than the adjacent site 2. Landward of the sites, 40 m into the forest, tree heights ranged from 1.53 m, compared to 1.2 1.9 m at the forest edge. Mean tree density (trees 1 m height) within the mangrove forest
26 RV Gladstone-Gallagher et al.
Downloaded by [Niwa] at 21:14 30 April 2014
and on the edge of the forest was 14 (93; 9 SEM, n 6) and 7 (91; n6) trees per 100 m2, respectively.
Litterfall production Litter traps were constructed using nylon shade cloth (2 mm mesh), and placed under the mangrove canopy to quantify litterfall (as in Woodroffe 1985; May 1999). Traps were conical (0.5 m depth), with a 0.25 m2 opening, designed to minimise litter loss. In total there were 24 traps: 12 replicate traps (3 m2 sample area) were randomly positioned under the mangrove trees at the edge of the forest (herein referred to as ‘edge’); and another 12 were placed 40 m within the forest (herein referred to as ‘within’). At both the edge and within locations, traps were randomly distributed in a line parallel to the shore, to encompass both sites 1 and 2. Trap openings were placed above the high tide water level to minimise litter loss during tidal inundation. Traps were sampled monthly from February 2011 to January 2012. Samples were rinsed with freshwater to remove sediment and salt, and then dried to constant dry weight (DW) at 60 8C. Litter was separated into leaf, wood, fruit/seed and inflorescences, and the DW of each determined.
Decomposition rates Decomposition rates of mangrove wood, pneumatophores and leaf litter were measured using litter bags (16 16 cm) made from 2 mm mesh nylon shade cloth (as in Woodroffe 1982). The 2 mm mesh size allowed small invertebrates access to the decomposing litter, whilst excluding larger macrofauna. The litter bag method has been criticised for excluding some macroinvertebrates, which may aid in the breakdown of litter into detritus (Fell et al. 1984), though a temperate study found similar decay rates between bags that allowed and bags that excluded entry of invertebrates (Goulter & Allaway 1979). Although litter bags may underestimate the
conditions of natural decay, they are appropriate for determining relative decay rates. The litter decomposition experiment began in summer (January 2011), to coincide with the period when the majority of mangrove litter is produced (see Results). Yellow senescent leaves (ready to abscise), wood (branch diameter 510 mm) and pneumatophores were collected from mangrove trees at Whangamata Harbour. Fibrous root material was removed from pneumatophores. Leaves, wood and pneumatophores were rinsed under freshwater to remove sediment and salt, air dried for 48 h (at constant temperature and humidity), weighed into subsamples (5 g leaves, 4 g pneumatophores, 7 g wood) and placed separately into the decomposition bags. A similar volume of litter was placed in each bag but, because of differences in water content and density, the initial weight varied slightly among litter types. To determine initial DW, 20 subsamples of each litter type were dried at 60 8C to constant DW. At each site the decomposition bags were placed at four tidal positions: low-tide (T-L); mid-tide (T-M); mangrove edge (pneumatophore zone with no canopy cover; M-E); and under the mangrove canopy (M-C). Additionally, at the M-E and M-C positions, bags were buried (1015 cm depth) in order to test the effect of burial on decomposition rates (b buried, s sediment surface). During periods of tidal inundation, bags positioned at the T-L position were submerged for 12.5 h longer than bags at the M-C position. The effect of tidal position on pneumatophore decomposition was not tested (bags only placed at M-C and M-E). At each tidal position, decomposition bags were tied to a central stake and then pegged down (four replicate stakes at each tidal position). Four replicate bags of each litter type (n4; one bag of each litter type per stake) were collected from each position on d 11, 24, 38, 51, 81, 169 and 357 after deployment. Only leaf samples were collected on d 11 and 38, and wood samples were also collected on d 262. Following each collection, samples were rinsed
Temperate mangrove litter production and decomposition 27
Downloaded by [Niwa] at 21:14 30 April 2014
with freshwater over a 500 mm sieve, placed in foil dishes and oven dried to constant DW at 60 8C. Decomposition was characterised by DW loss over time. Dried leaf samples for M-C-s, M-C-b, M-E-s, M-E-b and T-L-s were analysed for total carbon (C) and total nitrogen (N) by grinding to a fine powder using a ball mill (Retsch manual mixer mill type MM2000, Retsch GmbH & Co.KG, Germany) and then analysed using an Elementar Vario EL cube C and N analyser (Elementar Analysensyteme GmbH, Germany). C and N content were analysed in leaf samples up to d 169, as after this time there was insufficient material for analysis.
Data analysis Raw litterfall data were analysed using a single t-test to determine significant differences in total annual litterfall between mangroves on the edge of the forest (i.e. newly established trees) and mangroves 40 m within the forest (i.e. established trees). Decomposition (% weight loss after 357 d) was analysed using different fixed factor multi-way analyses of variance (ANOVA) to test the effect of: firstly, site (1, 2), tidal position (M-C-s, M-E-s, T-M-s, T-L-s) and litter type (wood, leaves); and, secondly, burial state (b, s), site (1, 2), forest position (M-C, M-E) and litter type (leaf, wood, pneumatophore). Percentage weight loss data were arcsine transformed and the distribution of residuals analysed after transformation, to ensure data met the assumptions of normality and homogeneity of variances. An additional fixed factor multi-way ANOVA tested the effect of tidal position and burial state (M-C-s, M-C-b, M-E-s, M-E-b, T-L-s) on leaf litter C and N content (raw data conformed to statistical assumptions) after 169 d. NewmanKeuls post hoc tests were used to determine where significant differences occurred. To characterise the decomposition process (as opposed to the final result tested above) and for comparison with previous studies (e.g. Robertson 1988; Mackey & Smail 1996; On˜atePacalioga 2005), we fitted decay models to the
time series data to estimate t50 (time taken for litter to decay to 50% of original weight). A single exponential decay model of X(t) e kt , where X(t) is the proportion of mangrove material remaining after time t (days) and k (day 1) is the decay constant, was used to describe the decay of surficial mangrove leaf litter. However, the decomposition of buried leaves was more suitably described by the asymptotic model (single r2 B0.3, asymptotic r2 0.8), X(t) C(1C)e kt, which assumes there is a fast initial decay of easily broken down labile material (k), followed by a completely decay-resistant recalcitrant fraction (C) (Wieder & Lang 1982). This asymptotic model assumes that litter will never decay completely, and therefore is unrealistic in nature; however, the model can be useful to describe decomposition rates during the study period (Wieder & Lang 1982). Wood and pneumatophore weight loss was small and data were highly variable. As a result, exponential decay models did not provide a good fit (r2 0.30.6), so a linear decay rate (a) was assumed between the initial (0 d) and final (357 d) sample date. Multi-way ANOVAs found no significant differences in weight loss (after 357 d) among tidal positions and sites (see Results), therefore decay constants and t50 values presented in the Results section were estimated from means pooled across sites and tidal positions. Results Litterfall production Mean annual litterfall 40 m within the forest (538974 g DW m 2 yr 1;9SEM) was significantly higher (t-test, P 0.015, n 12) compared to the younger trees on the edge of the forest (324943 g DW m 2 yr 1). Annual litterfall consisted of 60%65% leaf, 9%11% wood, 25%26% fruit and 1%3% inflorescences, both on the edge and within the mangrove forest. At both locations the majority of litterfall (77%) occurred during the warmer months of November to February (Fig. 1). Leaf fall occurred all year round, but was minimal in the colder months (MarchOctober). Wood fall remained relatively constant throughout the year,
28 RV Gladstone-Gallagher et al. 25
180
20
Litterfall (g DW m–2)
140 120
15 100 80
10 60 40
Mean monthly air temperature (°C)
160
5
Feb
Mar
Apr Leaf
May
Jun
Jul
Wood
Fruit
Oct
Nov
Dec
Edge
Within
Edge
Within
Edge
Within
Edge
Sep
Aug
Within
Edge
Within
Edge
Within
Edge
Within
E d ge
Within
Edge
Within
Edge
Within
Edge
Within
Edge
0
Within
Downloaded by [Niwa] at 21:14 30 April 2014
20
0
Jan
Inflorescences
Mean monthly air temperature
Figure 1 Mean (1 SEM, n 12) monthly (20112012) mangrove litterfall rates and composition, on the edge and 40 m within the forest. Secondary y-axis shows mean monthly air temperature (data obtained from temperature loggers deployed in Whangamata Harbour during the study).
and so comprised a greater fraction (up to 60%) of the monthly litterfall in winter. Fruit fall was largest during summer months (12%44% of monthly litterfall), but low during the remainder
of the year (0%29% of monthly litterfall). Inflorescences were collected throughout the year, although they represented a very small proportion of the annual litterfall (1%3%).
Table 1 Summary of multi-way ANOVA comparing mean percentage weight loss after 357 d (arcsine transformed) between sites (1, 2), litter types (wood, leaf), and tidal positions (M-C, M-E, T-M, T-L). Significant results (PB0.05) are indicated in bold. Results of Newman-Keuls post hoc tests are shown as footnotes. Source of variation
d.f.
Mean-square
F-ratio
P
Site Litter type Tidal position Site*Litter type Site*Tidal position Litter type*Tidal position Site*Litter type*Tidal position Error
1 1 3 1 3 3 3 47
0.01 7.49 0.03 0.01 0.01 0.05 0.01 0.02
0.62 420.31 1.66 0.69 0.30 2.71 0.29
0.436 B0.0011 0.189 0.411 0.824 0.056 0.829
1
Leaf wood.
Temperate mangrove litter production and decomposition 29 Decomposition rates A multi-way ANOVA revealed that, for wood and leaf litter on the sediment surface, neither site nor tidal position affected weight loss after 357 d of decomposition. There was, however, a significant difference in weight loss with litter type, where leaves lost on average 95% of their weight after 357 d, while wood lost only 40% (Table 1, Fig. 2). In a second multi-way ANOVA, burial state (surface, buried), litter type (leaf,
Leaf Weight remaining (%)
Site 1
Site 2
B
100 80 60 40 20 0
Wood Weight remaining (%)
C
E
Pneumatophore Weight remaining (%)
Downloaded by [Niwa] at 21:14 30 April 2014
A
wood, pneumatophore) and their interaction were significant; however, other factors (site, forest position) were not significant (Table 2, Fig. 2). Post hoc testing revealed leaf litter decomposition on the sediment surface (at both M-C and M-E positions) lost the most weight (95%), followed by pneumatophores (59%) and wood (40%). Buried leaf and pneumatophore material lost the same weight over 357 d (61% and 58%, respectively), which was about 2 times
D
100 80 60 40 20 0
F
100 80 60 40 20 0 0
50
100
150
200
250
300
350
400
0
50
100
150
200
250
300
350
400
Time (d)
Time (d) M-C-s M-C-b
M-E-s M-E-b
T-M-s T-L-s
Figure 2 Mangrove litter decomposition (expressed as the percentage of original weight remaining through time) at varying tidal positions and burial states. A, Leaves, site 1. B, Leaves, site 2. C, Wood, site 1. D, Wood, site 2. E, Pneumatophores, site 1. F, Pneumatophores, site 2. For clarity, each figure includes the maximum and minimum standard error (SEM).
30 RV Gladstone-Gallagher et al.
Downloaded by [Niwa] at 21:14 30 April 2014
Table 2 Summary of multi-way ANOVA comparing mean percentage weight loss after 357 d (arcsine transformed) between sites (1, 2), litter types (wood, leaf, pneumatophore), forest positions (M-C or M-E) and burial states (buried or surface). Significant results (PB0.05) are indicated in bold. Results of NewmanKeuls post hoc tests are shown as footnotes. Source of variation
d.f.
Mean-square
F-ratio
P
Site Litter type Forest position Burial state Site*Litter type Site*Forest position Litter type*Forest position Site*Burial state Litter type*Burial state Position*Burial state Site*Litter type*Forest position Site*Litter type*Burial state Site* Forest position*Burial state Litter type* Forest position*Burial state Site*Litter type* Forest position*Burial state Error
1 2 1 1 2 1 2 1 2 1 2 2 1 2 2 60
0.01 1.86 0.01 0.91 0.04 0.04 0.02 0.07 0.48 0.06 0.03 0.02 0.00 0.04 0.01 0.02
0.24 88.21 0.31 42.93 1.68 1.98 1.18 3.32 22.94 2.78 1.48 1.04 0.07 2.10 0.26
0.626 B0.001 0.580 B0.001 0.195 0.165 0.316 0.073 B0.0011 0.101 0.235 0.359 0.794 0.131 0.774
1 Buried: leaf pneum. wood; Surface: leafpneum. wood; Leaf: surface buried; Wood: surface buried; Pneum.: buriedsurface.
more weight than buried wood (30% loss). Pneumatophore weight loss was not affected by burial state, with a buried and surface weight loss of 58% and 59%, respectively. However, buried wood and leaves lost significantly less weight (30% and 61%, respectively) than their surficial counterparts (40% and 95%, respectively) (Table 2, Fig. 2). The fitted decay models show differences in decomposition rates that are reflected in weight loss differences after 357 d between litter types and burial treatments (Table 3). The t50 of wood (460613 d) was about 1.5 and 7 times longer than pneumatophores (317 d) and leaves (6388 d), respectively. The mean t50 of buried leaves and wood was approximately 1.3 and 1.4 times greater (an additional 25 d and 153 d, respectively) than for surface equivalents, respectively (Table 3). After 169 d of decomposition, average total N increased in all leaf litter from 0.96% to 1.31% (Fig. 3A, 3B), but there was no significant difference in final N content with site, tidal
position and burial state (Table 4). In contrast, total C content was dependent on site and burial state, but was unaffected by tidal position. C content of leaf litter decreased in leaves decomposing on the sediment surface from 45.1% (initial) to 42.7% and 39.0% (at sites 1 and 2, respectively; Fig. 3C, 3D). However, C content of buried leaves (47.6% site 1, 45.8% site 2, after 357 d) remained the same, or increased slightly, during the decomposition process (Fig. 3C, 3D). Consequently, the final C content of buried leaves was significantly higher than for leaves on the surface (Table 4). Leaves decomposing at site 2 had a significantly lower final C content (45.83% buried, 38.95% surface) than leaves at site 1 (47.58% buried, 42.71% surface; Table 4). As a result of N enrichment, the C:N ratio decreased in all leaves from a mean of 47 on d 0 to 31 (surface) and 37 (buried) after 169 d (Fig. 3E, 3F). Buried leaves had a significantly higher C:N ratio (37) after 169 d compared to leaves on the sediment surface (31). No significant
5868 7897 409511 539686 263373 X(t) e kt; 2X(t) C(1C)e kt; 3YaX1.
Discussion
1
t509SEM 95% CI
site effect was detected in leaf C:N ratios after 169 d (Table 4).
6393 8896 460928 613943 317930 Single 0.011190.0005 0.01020.0120 0.9090.01 Asymptotic2 0.032790.0042 0.02560.0399 0.463290.0134 0.44050.4860 0.8390.06 Linear3 0.001190.0001 0.0010.0012 0.8690.04 0.000890.0001 0.00070.0009 0.8290.09 Linear3 Linear3 0.001690.0001 0.00140.0018 0.9190.05 Leaves- surface Leaves- buried Wood- surface Wood- buried Pneum.
1
r29SEM 95% CI a9SEM 95% CI C9SEM 95% CI k9SEM Model
Table 3 Fitted decay models used to describe litter decomposition as a function of type and burial position. t50 values represent the time (days) for litter to decay to 50% of original weight (calculated from decay model equations and rate constants). Also shown are the decay constants (and 95% confidence intervals) and an estimate of the model fit (r2 values). Means are calculated from means pooled across sites and tidal positions (where ANOVA found no significant difference in total weight loss). Footnotes indicate the model equations that were fitted.
Downloaded by [Niwa] at 21:14 30 April 2014
Temperate mangrove litter production and decomposition 31
Litterfall production Annual litterfall estimates in Whangamata Harbour are equivalent to 3.245.38 t DW ha 1 yr 1 (324538 g DW m 2 yr 1), and are within the range reported in the literature for other sites in New Zealand (1.308.10 t DW ha 1 yr 1; Woodroffe 1985; May 1999; On˜ate-Pacalioga 2005). They are also similar to mangrove forests at other temperate latitudes (0.1111.68 t DW ha 1 yr 1), but are at the lower end of those observed for tropical forests (3.7418.7 t DW ha 1 yr 1; reviewed by Morrisey et al. 2010). Litter composition was consistent with other studies, where leaves constitute the largest proportion of the litter (60%65%; e.g. Woodroffe 1985; May 1999; Imgraben & Dittmann 2008). Litter production of younger trees on the edge of the Whangamata mangrove forest was roughly half that of the established trees within the forest, consistent with previous studies that have shown productivity to be spatially variable within a forest and can vary with tree height (reviewed by Morrisey et al. 2010). The seasonal variation in litterfall, characterised by a pulse in summer months and minimal production during the rest of the year, has been reported elsewhere and is attributed to seasonal changes in temperature, rainfall and evapotranspiration (Woodroffe 1985; Clarke 1994; May 1999; On˜ate-Pacalioga 2005; Sa´nche´zAndres et al. 2010). Conversely, in some tropical regions there are multiple peaks in litter production throughout the year (Duke 1990). While litterfall measurements do not measure the increase in plant biomass, they are regarded as an important component of primary productivity in determining organic matter contribution to the estuary (Woodroffe 1982; Clarke 1994). In temperate systems, cross-boundary subsidies are highly seasonal compared to tropical systems, and therefore are likely to function differently
32 RV Gladstone-Gallagher et al. Site 1
A 1.7
Site 2
B
Total nitrogen (%)
1.6 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0
C
D
55
Total carbon (%)
51 49 47 45 43 41 39 37 0
E
F
55 50 45
C:N ratio
Downloaded by [Niwa] at 21:14 30 April 2014
53
40 35 30 25 0
0
20
40
60
80 100 120 140 160 180 200
0
20
Time (d) M-C-s
40
60
80 100 120 140 160 180 200
Time (d) M-C-b
M-E-s
M-C-b
T-L-s
Figure 3 Carbon and nitrogen content of leaf litter during 169 d of decomposition. A, Total nitrogen content of leaves at site 1. B, Total nitrogen content of leaves at site 2. C, Total carbon content of leaves at site 1. D, Total carbon content of leaves at site 2. E, C:N ratio of leaves at site 1. F, C:N ratio of leaves at site 2. For clarity, each figure includes the maximum and minimum standard error (SEM).
Temperate mangrove litter production and decomposition 33 Table 4 Summary of ANOVA for %C, %N and C:N ratios in leaves after 169 d of decomposition, comparing differences between sites (1, 2) and decomposition bag positions (M-C-s, M-C-b, M-E-s, M-E-b, T-L-s). Significant results (PB0.05) are indicated in bold. Results of Newman-Keuls post hoc tests are shown as footnotes. %N Source of variation
d.f.
Bag position 4 Site 1 Bag position*Site 4 Error 26
Meansquare 0.02 0.06 0.00 0.03
Fratio
%C P
0.74 0.571 2.08 0.161 0.14 0.967
C:N ratio
d.f.
Meansquare
Fratio
4 1 4 26
78.12 77.27 16.03 13.70
5.70 0.0021 5.64 0.0252 1.17 0.347
P
d.f.
Meansquare
Fratio
4 1 4 26
78.97 0.75 5.78 6.42
12.30 B0.0013 0.12 0.735 0.90 0.478
P
M-E-b M-C-b M-C-s M-E-s T-L-s; 2site 1 site 2; 3M-C-b M-E-b M-C-s M-E-s T-L-s.
Downloaded by [Niwa] at 21:14 30 April 2014
1
with regard to consumer response and nutrient cycling. Although temperate mangrove forests are less productive than those in the tropics, our results suggest that mangrove detrital production is similar to that produced by seagrass (Zostera muelleri). Above-ground seagrass production in the Whangamata Harbour has been estimated at 1.11.6 g DW m 2 d 1 (Turner 2007) which, assuming a 71 ha area (Singleton 2007), suggests an annual production rate for the harbour of 370 t DW. Scaling up mangrove litter production rates measured in this study to the 101 ha of forest (Singleton 2007) gives a comparable estimate of 327543 t DW yr 1. Seagrass beds in both tropical and temperate coastal systems are an important detrital subsidy at the base of coastal food webs (e.g. Doi et al. 2009). Our simple comparison suggests mangroves could potentially provide an important source of detritus to temperate coastal systems; however, this will be dependent on how much of the productivity is exported from the forest, as well as its rate of decay and assimilation.
Decomposition rates Burial significantly reduced the rate and amount of leaf weight and carbon loss, and differences in the fitted decay models indicate distinct differences in the decay processes. A large proportion of leaf carbon is locked up in the recalcitrant
decay resistant fraction of the leaf (Davis et al. 2003), which resisted decay to a greater extent in buried leaves (asymptotic vs. single decay models). The asymptotic decay model (used to describe buried litter decay) has been associated with litter decomposition that excludes faunal activity (Wieder & Lang 1982), as macrofauna may aid in degradation of the recalcitrant portion of the leaf (On˜ate-Pacalioga 2005; Proffitt & Devlin 2005). Accordingly, anoxic litter decomposition is likely to be primarily through bacterial breakdown, as biotic and abiotic variables that accelerate litter fragmentation (macrofauna, climate and tidal inundation; Robertson 1988; Mackey & Smail 1996; Woitchik et al. 1997; Davis et al. 2003; Proffitt & Devlin 2005) will be minimal at depths of 1015 cm in the sediment. Our findings are consistent with results from subtropical Florida, where carbon loss was greater in surficial compared to buried leaves, although no differences in mass loss were observed (Fourqurean & Schrlau 2003). Leaf litter that is retained within the mangrove forest and buried due to sedimentation could result in nutrient recycling within the forest. However, the recycling of buried litter is likely to be slow in New Zealand forests due to the slow rate of weight loss and carbon decay under anoxic conditions. Fresh mangrove leaves are a relatively low quality food resource to marine consumers (high tannins and C:N ratios), and decomposition turns
Downloaded by [Niwa] at 21:14 30 April 2014
34 RV Gladstone-Gallagher et al. unpalatable leaf litter into biologically available detritus (via the microbial pathway; Robertson 1988; Nordhaus et al. 2011). Leaf palatability to consumers increases as a function of a reduced C:N ratio and enhanced bacterial colonisation (Fell et al. 1984; Nordhaus et al. 2011). Here, the C:N ratio in all leaves decreased from approximately 45 to 35 after 3 months of decomposition as a result of nitrogen enrichment and carbon loss; however, this occurred to a greater extent in surficial leaves, as buried leaves resisted carbon decay. Similar C:N decreases have been measured in subtropical and tropical studies (Rice & Tenore 1981; Dick & Osunkoya 2000). However, in other tropical mangrove studies, C:N ratios decreased by half within 4 months (C:N decrease from 110 to c. 50, Fourqurean & Schrlau 2003; and from 75 toB37.5, Robertson 1988). Although wood comprises 9%18% of annual mangrove litterfall (this study; May 1999), wood decomposition studies are lacking in temperate regions. As mangrove wood and root material is often left in situ following mangrove removals in New Zealand estuaries (Lundquist et al. 2012), it is useful to describe the decay of such material to determine site recovery. Model predictions (linear decay) suggest that mangrove wood and pneumatophores could take between 317 and 613 d to decompose to half their original weight. Even at lower latitudes (subtropical and tropical), mangrove wood decay is expected to take years (e.g. t50 179421 d; Steinke et al. 1983; Robertson & Daniel 1989; Mackey & Smail 1996). Wood decay was significantly slowed by burial, which can occur due to sedimentation or macroalgal blooms on cleared mangrove patches (Lundquist et al. 2012). In contrast, pneumatophore decay rate was unaffected by burial, though this study considered only the decay of pneumatophores, whilst excluding the fibrous root material. Just two temperate studies have investigated mangrove root decay; one found that pneumatophores and fibrous roots decomposed at similar rates (Albright 1976), while the other showed that pneumatophores decomposed significantly faster
(Van der Valk & Attiwill 1984). Thus, our results may provide an underestimate of complete root degradation rates. Previous investigations have linked litter decay rates with differences in tidal submersion (Robertson 1988; Mackey & Smail 1996; Woitchik et al. 1997; Dick & Osunkoya 2000) and site characteristics, such as local macrofaunal assemblages (Bosire et al. 2005; Proffitt & Devlin 2005) and sediment properties (Holmer & Olsen 2002). However, expectations that mangrove litter decay rates would differ between sites and tidal positions were not supported by our results. Differences in tidal submergence periods tested in previous studies were larger (e.g. submerged daily vs. submerged only in some tidal cycles; Robertson 1988; Mackey & Smail 1996; Woitchik et al. 1997; Dick & Osunkoya 2000) than those measured by this experiment (12.5 h difference in submergence period between high and low tidal elevations). The relatively small difference in the submergence times compared in this study may have been too small to influence mangrove litter decay. In addition, site-specific differences in decay rates have been observed in distinctly different macrofaunal communities (Bosire et al. 2005), but this study measured only differences in the relative abundances of species rather than differences in species compositions (Gladstone-Gallagher 2012). We anticipated that litter decay rates would differ with sediment type, because in tropical forests rates were faster in sandy nutrient-poor sediments than in muddy nutrient-rich sediments (k 0.013 vs. 0.0075 day 1; Holmer & Olsen 2002). Although, we detected no site effect in litter weight loss, the two sites exhibited differences in leaf carbon content. Leaf litter at the muddy sand site (site 2) lost significantly more carbon (after 169 d), which could indicate increased degradation of the recalcitrant components of the leaf (Davis et al. 2003), though this was not mirrored in mass loss. Mangrove leaf degradation is slow in temperate New Zealand forests (t50 4288 d; Albright
Downloaded by [Niwa] at 21:14 30 April 2014
Temperate mangrove litter production and decomposition 35 1976; Woodroffe 1982; On˜ate-Pacalioga 2005; this study), compared to tropical counterparts, where t50 is often reached within one week (e.g. Bosire et al. 2005; Sa´nche´z-Andres et al. 2010). Consequently, it is expected that organic matter cycling and detrital production is faster in tropical forests than in temperate New Zealand. However, cross-boundary subsidies of organic matter from mangrove forests are not only controlled by the production and decomposition of litter, but also depend on whether the litter is exported from the forest (Lugo & Snedaker 1974; Polis et al. 1997). Some authors have hypothesised that rapid decomposition rates could be associated with a lower incidence of litter export and a greater chance of nutrient recycling within the mangrove forest, with slow decomposition rates exhibiting the reverse (Imgraben & Dittmann 2008; Adame & Lovelock 2011). Other studies, utilising stable isotopes, have suggested that mangrove litter in temperate regions is more likely to be consumed within the forest rather than exported to adjacent food webs (Guest & Connolly 2004; Alfaro et al. 2006). Therefore, the fate of mangrove litter in temperate forests requires further investigation through the quantification of litter export to determine the magnitude of the potential cross-boundary subsidy. Acknowledgements Funding was provided by the New Zealand Ministry for Science and Innovation Project No. CO1X1002, the Waikato Regional Council and University of Waikato Masters Research Scholarship. We thank Sarah Hailes, Dudley Bell, Nikki Webb, Gillian Gladstone, Maxine Gillard, Ray Foster, Martin Gallagher, Simon Brown, Barry Greenfield, Lisa McCartain, Pauline Robert, Daniel Pratt, Jamie Armstrong, Arie Spyksma, Stacey Buchanan, Nicholas Wu, Rachel Harris, Clarisse Niemand, Dorothea Kohlmeier, Kelly Carter, Phil Ross, Tim Rogers, Anya Gladstone-Gallagher and Marenka Weis for assistance with laboratory and field work. Also, we thank three anonymous reviewers for their constructive comments that improved the manuscript.
References Adame MF, Lovelock CE 2011. Carbon and nutrient exchange of mangrove forests with the coastal ocean. Hydrobiologia 663: 2350. Albright LJ 1976. In situ degradation of mangrove tissues (note). New Zealand Journal of Marine and Freshwater Research 10: 385389. Alfaro AC, Thomas F, Sergent L, Duxbury M 2006. Identification of trophic interactions within an estuarine food web (northern New Zealand) using fatty acid biomarkers and stable isotopes. Estuarine, Coastal and Shelf Science 70: 271286. Bosire JO, Dahdouh-Guebas F, Kairo JG, Kazungu J, Dehairs F, Koedam N 2005. Litter degradation and CN dynamics in reforested mangrove plantations at Gazi Bay, Kenya. Biological Conservation 126: 287295. Bouillon S, Moens T, Koedam N, Dahdouh-Guebas F, Baeyens W, Dehairs F 2004. Variability in the origin of carbon substrates for bacterial communities in mangrove sediments. FEMS Microbiology Ecology 49: 171179. Clarke PJ 1994. Baseline studies of temperate mangrove growth and reproduction: demographic and litterfall measures of leafing and flowering. Australian Journal of Botany 42: 3748. Davis SE, Corronado-Molina C, Childers DL, Day Jr. JW 2003. Temporally dependent C, N and P dynamics associated with the decay of Rhizophora mangle L. leaf litter in oligotrophic mangrove wetlands of the southern Everglades. Aquatic Botany 75: 199215. Dick TM, Osunkoya OO 2000. Influence of tidal restriction floodgates on decomposition of mangrove litter. Aquatic Botany 68: 273280. Doi H, Matsumasa M, Fujikawa M, Kanou K, Suzuki T, Kikuchi E 2009. Macroalgae and seagrass contribution to gastropods in subtropical and temperate tidal flats. Journal of the Marine Biological Association of the United Kingdom 89: 399404. Duke NC 1990. Phenological trends with latitude in the mangrove tree Avicennia marina. Journal of Ecology 78: 113133. Fell JW, Master IM, Wiegert RG 1984. Litter decomposition and nutrient enrichment. In: Snedaker SC, Snedaker JG eds. The mangrove ecosystem: research methods. Paris, UNESCO. Pp. 239251. Fourqurean JW, Schrlau JE 2003. Changes in nutrient content and stable isotope ratios of C and N during decomposition of seagrasses and mangrove leaves along a nutrient availability gradient in Florida Bay, USA. Chemistry and Ecology 19: 373390.
Downloaded by [Niwa] at 21:14 30 April 2014
36 RV Gladstone-Gallagher et al. Gladstone-Gallagher RV 2012. The production and decay of mangrove (Avicennia marina subsp. australasica) detritus and its effects on coastal benthic communities. Unpublished MSc thesis. University of Waikato, New Zealand. 86 p. Goulter PFE, Allaway WG 1979. Litterfall and decomposition in a mangrove stand, Avicennia marina (Forsk.) Vierh., in Middle Harbour, Sydney. Australian Journal of Marine and Freshwater Research 30: 541546. Guest MA, Connolly RM 2004. Fine-scale movement and assimilation of carbon in saltmarsh and mangrove habitat by resident animals. Aquatic Ecology 38: 599609. Harty C 2009. Mangrove planning and management in New Zealand and South East Australia a reflection on approaches. Ocean & Coastal Management 52: 278286. Holmer M, Olsen AB 2002. Role of decomposition of mangrove and seagrass detritus in sediment carbon and nitrogen cycling in a tropical mangrove forest. Marine Ecology Progress Series 230: 87101. Imgraben S, Dittmann S 2008. Leaf litter dynamics and litter consumption in two temperate South Australian mangrove forests. Journal of Sea Research 59: 8393. Kristensen E, Bouillon S, Dittmar T, Marchand C 2008. Organic carbon dynamics in mangrove ecosystems: a review. Aquatic Botany 89: 201 219. Lugo AE, Snedaker SC 1974. The ecology of mangroves. Annual Review of Ecology and Systematics 5: 3964. Lundquist C, Hailes S, Cartner K, Carter K, Gibbs M 2012. Physical and ecological impacts associated with mangrove removals using in situ mechanical mulching in Tauranga Harbour. NIWA Technical Report 137. 106 p. Mackey AP, Smail G 1996. The decomposition of mangrove litter in a subtropical mangrove forest. Hydrobiologia 332: 9398. May JD 1999. Spatial variation in litter production by the mangrove Avicennia marina var. australasica in Rangaunu Harbour, Northland, New Zealand. New Zealand Journal of Marine and Freshwater Research 33: 163172. Morrisey DJ, Swales A, Dittmann S, Morrison MA, Lovelock CE, Beard CM 2010. The ecology and management of temperate mangroves. Oceanography and Marine Biology: An Annual Review 48: 43160. Nordhaus I, Salewski T, Jennerjahn TC 2011. Food preferences of mangrove crabs related to leaf nitrogen compounds in the Segara Anakan
Lagoon, Java, Indonesia. Journal of Sea Research 65: 414426. On˜ate-Pacalioga JA 2005. Leaf litter production, retention, and decomposition of Avicennia marina var. australasica at Whangateau Estuary, Northland, New Zealand. Unpublished MSc thesis. University of Auckland, New Zealand. 108 p. Polis GA, Anderson WB, Holt RD 1997. Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annual Review of Ecology and Systematics 28: 289316. Proffitt CE, Devlin DJ 2005. Grazing by the intertidal gastropod Melampus coffeus greatly increases mangrove litter degradation rates. Marine Ecology Progress Series 296: 209218. Rice DL, Tenore KR 1981. Dynamics of carbon and nitrogen during the decomposition of detritus derived from estuarine macrophytes. Estuarine, Coastal and Shelf Science 13: 681690. Robertson AI 1988. Decomposition of mangrove leaf litter in tropical Australia. Journal of Experimental Marine Biology and Ecology 116: 235 247. Robertson AI, Daniel PA 1989. Decomposition and the annual flux of detritus from fallen timber in tropical mangrove forests. Limnology and Oceanography 34: 640646. Saenger P, Snedaker SC 1993. Pantropical trends in above-ground biomass and annual litterfall. Oecologia 96: 293299. Sa´nche´z-Andres R, Sa´nchez-Carrillo S, Alatorre LC, Cirujano S, A´lvarez-Cobelas M 2010. Litterfall dynamics and nutrient decomposition of arid mangroves in the Gulf of California: their role in sustaining ecosystem heterotrophy. Estuarine, Coastal and Shelf Science 89: 101199. Singleton P 2007. Draft Whangamata Harbour plan. Environment Waikato Internal Series 2007/14. 84 p. Steinke TD, Naidoo G, Charles LM 1983. Degradation of mangrove leaf and stem tissues in situ in Mgeni Estuary, South Africa. In: Teas HJ ed. Biology and ecology of mangroves. The Hague, The Netherlands, Dr W Junk Publishers. Pp. 141149. Turner SJ 2007. Growth and productivity of intertidal Zostera capricorni in New Zealand estuaries. New Zealand Journal of Marine and Freshwater Research 41: 7790. Twilley RR, Chen RH, Hargis T 1992. Carbon sinks in mangroves and their implications to carbon budget of tropical coastal ecosystems. Water, Air and Soil Pollution 64: 265288. Van der Valk AG, Attiwill PM 1984. Decomposition of leaf and root litter of Avicennia marina at
Temperate mangrove litter production and decomposition 37
Downloaded by [Niwa] at 21:14 30 April 2014
Westernport Bay, Victoria, Australia. Aquatic Botany 18: 205221. Wieder RK, Lang GE 1982. A critique of the analytical methods used in examining decomposition data obtained from litter bags. Ecology 63: 16361642. Woitchik AF, Ohowa B, Kazungu JM, Rao RG, Goeyens L, Dehairs F 1997. Nitrogen enrichment during decomposition of mangrove leaf litter in an eastern African coastal lagoon
(Kenya): relative importance of biological nitrogen fixation. Biogeochemistry 39: 1535. Woodroffe CD 1982. Litter production and decomposition in the New Zealand mangrove, Avicennia marina var. resinifera. New Zealand Journal of Marine and Freshwater Research 16: 179188. Woodroffe CD 1985. Studies of a mangrove basin, Tuff Crater, New Zealand: I. Mangrove biomass and production of detritus. Estuarine, Coastal and Shelf Science 20: 265280.
