- Email: [email protected]
Dynamics of phosphorus and nitrogen through litter fall and decomposition in a tropical mangrove forest
MARINE ENVIRONMENTAL RESEARCH Marine Environmental Research 64 (2007) 524–534 www.elsevier.com/locate/marenvrev
Dynamics of phosphorus and nitrogen through litter fall and decomposition in a tropical mangrove forest Carlos A. Ramos e Silva a
a,*
, Se´rgio R. Oliveira a,b, Ronaldo D.P. Reˆgo Antonio A. Mozeto c
a,c
,
Departamento de Oceanografia e Limnologia (Laborato´rio de Biogeoquı´mica Ambiental), Universidade Federal do Rio Grande do Norte, Natal, RN, Cxa. Postal 1202, 59075-970, Brazil b Laborato´rio de Cieˆncias Ambientais, Universidade Estadual do Norte Fluminense, Campos do Goytacazes, RJ, Brazil c Departamento de Quı´mica, Universidade Federal de Sa˜o Carlos, Sa˜o Carlos, SP 13565-905, Brazil Received 14 November 2005; received in revised form 20 June 2006; accepted 20 April 2007
Abstract Distribution, dynamics and mass budget of phosphorus and nitrogen in a red mangrove forest were studied in the Potengi mangrove forest in northern Brazil (lat. 542 0 and 553 0 S, long. 355 0 and 3525 0 W). Tidal hydrology, net primary productivity, leaf litter decomposition rate and standing stock of leaf litter in a red mangrove forest were measured. The results showed that the main reservoir for total P and total N was the sediment with 309 kg ha1 and 4619 kg ha1 (77% and 95% of the total P and N content in the mangrove forest), respectively, for the two elements. Total P and total N in Rhizophora mangle trees accounted for 145 ± 14 kg ha1 and 216 ± 23 kg ha1 (23% and 5% of the total P and N in the mangrove forest). The estimated average export rates for P and N through leaf litter are 0.5 kg ha1 yr1 and 1.6 kg ha1 yr1 respectively. Our measurements support previous results in concluding that mangrove forests efficiently retain P and N. 2007 Elsevier Ltd. All rights reserved. Keywords: Rhizophora mangle; Phosphorus; Nitrogen; Mangrove; Litter fall
1. Introduction Mangrove forests may be considered as a ‘‘chemical reactors’’ not only because of their physiological and biochemical processes but also due to their active influence in the mobility of nutrients (Lacerda et al., 1993; Silva, 1996; Silva and Mozeto, 1997; Silva et al., 1998; Silva and Sampaio, 1998; Nielsen and Andersen, 2003; Bosire et al., 2005). This influence is affected by tidal hydrology, litter fall production, leaf decomposition rate, and standing stock and the physical–chemical composition of local sediment.
*
Corresponding author. Tel.: +55 84 32154433; fax: +55 84 32023004. E-mail address: [email protected] (C.A. Ramos e Silva).
0141-1136/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.marenvres.2007.04.007
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
525
The addition of sewage effluent with high nutrient concentrations has influenced estuaries in various tropical regions. Of these nutrients, phosphorus and nitrogen have received attention due to their effects on estuarine organisms (Kennish, 1991; Ferreira et al., 1996). Previous studies have proposed that mangrove forests not only resist the effect of excess nutrients from sewage, but also decrease their concentrations (Por, 1984; Tam and Wong, 1995). However, few quantitative studies have been carried out on the role of litter production, decomposition and export in P and N cycling in mangrove ecosystems. The cycling of organic matter through litter production, decomposition and tidal transport, may eventually export a fraction of the accumulated P and N to local estuarine waters and increase the load of these nutrients in these estuarine waters. The objective of the present study was to quantify the cycling of phosphorus and nitrogen through litter detritus in a red mangrove (Rhizophora mangle) forest in Natal, near the city of Rio Grande do Norte, a moderately polluted area. 2. Materials and methods 2.1. Study sites The study was carried out in the Potengi experimental forest (PEF), located in the Potengi estuary, State of Rio Grande do Norte, northern Brazil (lat. 542 0 and 5 53 0 S, long. 355 0 and 3525 0 W) (Fig. 1). The forest has an area of approximately 863 ha, and is limited by two tidal creeks running almost perpendicular to the shore. The average temperature ranges between 24.6 C in the rainy season (March–August) and 28.8 C in the dry season (September–February). In the rainy season rainfall is about 1390 mm, while in the dry season it is about 220 mm. The dominant species is the red mangrove R. mangle although isolated trees of the black mangrove (Avicennia schaueriana) and of the white mangrove (Laguncularia racemosa) occur throughout the forest representing less than 10% respectively of the total coverage (Souza, 1999). Average R. mangle tree density is 3200 trees ha1, with an average tree height of 5.4 m and an average trunk diameter of 6.2 cm. The total basal areas are 9.5 m2 ha1 (Silva and Rego, 1997). The waters of the Potengi, Jundiaı´ and Doce Rivers form the estuary. The Potengi River, which drains a catchment of 3180 km2, has a total course of 176 km and a discharge of approximately 5 m3 s1 during the rainy season (March–August). The Jundiaı´ River has an independent catchment with a discharge smaller than that of the Potengi River. With this smaller discharge the Jundiaı´ is more influenced by tides, the effects of which have been observed as far upstream as the city of Macaı´ba, 30 km from the mouth of the estuary. The Doce River comprises the smallest freshwater input to the estuary with a discharge of approximately 2 m3 s1 during most of the year. Water temperatures vary from 26.5 to 29.0 C during the year.
Fig. 1. Location of the Potengi estuary in Natal, Rio Grande do Norte State, NE Brazil.
526
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
The Potengi estuary extends for about 25 km with a width ranging between 400 and 600 m. The estuary is bordered by three cities with a total population of more than 1,000,000 inhabitants. One of the cities, Natal, discharges more than 60% of its untreated sewage directly into the Potengi river (15,000 m3 per day – 14 mg L1 of total phosphorus and 50 mg L1 of total nitrogen) and there are several other point sources of organic effluent along the estuary. The mean concentration of phosphorus ranges from 2.10 to 3.70 lM for PO4 2 3 ðH3 PO4 þ H2 PO 4 þ HPO4 þ PO4 Þ and from 4.20 to 7.90 lM for total dissolved phosphorus (TDP) (Silva, þ unpublished data). Total nitrogen ðNO 3 þ NO2 þ NH3 þ NH4 Þ are also found in high concentrations in the estuary studied, where the mean concentrations range from 23.1 to 108.7 lM (Silva, unpublished data). On the west side of the estuary the mangrove forests have been removed and replaced by shrimp ponds. 2.2. Collection and preparation of samples 2.2.1. Sediment samples Ten sediment cores were collected in a single time (February, 1997) to a depth of 25 cm (which represent the highest biomass of roots) from the mudflat, sliced at 5-cm intervals and each section placed in clean plastic bag and immediately brought to the laboratory in ice buckets. In the laboratory sediment samples were oven dried at 60 C for 72 h, sieved (<1 mm mesh) and ground to fine powder and placed in glass vials for analysis. To obtain the fraction of phosphate binding to iron and aluminum (Fe/Al-P) and binding to calcium (Ca-P), 1 g of dry sediment was washed in an alcoholic solution of 15 mL 1 N KCl to eliminate calcium ions in the sediment. The first phase (determination of Fe/Al-P) consisted of an extraction with 15 mL 1 N NaOH + 1 N Na2SO4 before the next phase (determination of Ca-P) which consisted of extractions of phosphorus with 15 mL 1 N H2SO4; the residual was washed in 4% Na2SO4 (Kurmies, 1972). The total inorganic phosphorus (TIP) concentrations were obtained by adding the Fe/Al-P and Ca-P concentrations. The concentrations of the phosphorus fraction were measured by the colorimetric method (Grasshoff et al., 1983). Leeg and Black’s method (1955) was used to determine the concentrations of phosphate binding to organic matter (OM-P) by colorimetry (Grasshoff et al., 1983). The OM-P fraction was calculated by subtracting the TIP concentrations from the total phosphorus concentration. Total nitrogen in sediment samples was determined only in a half of the cores by sulfuric acid (H2SO4) digestion employing salicylic acid in the indophenol blue formation at 350 C according to Nogueira et al. (1992). 2.3. Standing stocks of phosphorus and nitrogen in R. mangle trees and litterfall The working units were seven quadrates of 100 m2 each, established perpendicular to the sea, approximately 45 m from the Doce River and 18 m from the forest seaward edge. In each quadrate all trees were counted. Height and diameter were measured in half of the trees in each quadrate and the results were then extrapolated to 1 ha (Citron and Schaefer-Novelli, 1983). For the estimation of total aboveground biomass a regression curve was drawn using the total dry weight (trunks, branches, prop roots and leaves) and diameter of eight cut trees, with diameters ranging from 2.1 to 18.7 cm. From the curve, estimates of the weight of all trees in the quadrates could be obtained and then summed up. This was possible since the total aboveground biomass was significantly correlated with tree diameter. Different parts of the R. mangle trees (leaves, branches, trunks, prop roots) were analyzed to determine the P and N concentrations. Plant samples were collected from 10 trees representative of the dominant class of tree diameter in the forest. The field samples were packed in plastic bags and taken to the laboratory. They were washed with tap water, oven dried (60 C for 96 h) and ground for nutrient analysis. The standing stocks of P and N in R. mangle trees were calculated by multiplying nutrient concentrations in each compartment by its biomass in 1 ha. The biomass of belowground roots was estimated by digging in the sediment to 25 cm depth through eight squares of 0.25 m2 random distributed through the study area. In the field, samples (fine and gross roots together) were weighed on manual scales to obtain the wet biomass. In the laboratory samples were oven dried at 60 C for 96 h and weighed. A relationship between wet weight and dry weight was obtained and the average dry biomass was extrapolated to 1 ha (Silva et al., 1990).
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
527
Litter fall was collected in seventeen 1.0 m2 baskets (1 mm nylon mesh) supported approximately 1.5 m above the ground. Litter fall was collected at 30 day intervals from August 1997 to July 1998. The material from each basket was dried at 60 C up to constant weight and separated into leaves, reproductive parts and thin branches (lesser than 2 cm diameter). Only the leaves were used for analysis of P and N. 2.4. Reservoir of phosphorus and nitrogen in sediment In the sediment the P and N contents were determined by multiplying the P and N concentrations by the total dry weight of sediment to a depth of 25 cm (see data from Table 1). 2.5. Standing crop of phosphorus and nitrogen in the leaf litter The standing crop of leaf litter was measured over 6 months, always during neap low tide, by collecting each month 15 0.25 m2 (0.5 m · 0.5 m) samples of litter on the surface of the forest floor in the vicinity of the litter fall collectors. Surface leaf samples were processed in the same manner as litter fall samples. In the laboratory the sediment was removed from the leaves by gently rubbing the leaves in running tap water and the samples were dried at 60 C per 96 h. After these procedures, the samples were prepared for phosphorus and nitrogen analysis. The reservoir of nutrients in the leaf litter was calculated by multiplying the standing stock of leaves (0.4 ± 0.3 t ha1, n = 90) by the concentration of P and N in R. mangle leaves from 15 to 90 days of decomposition (1.53 ± 0.29 mg g1 and 2.87 ± 1.24 mg g1 respectively, for the two elements). 2.6. Transfer of phosphorus and nitrogen from R. mangle trees to the sediment The transfer of nutrients through R. mangle leaf fall to the sediment was calculated by multiplying the mean concentration (mg g1), of nutrients in leaf fall (P = 1.16 ± 0.28 and N = 3.08 ± 0.64) by the mean leaf fall rate (1000 kg ha1 month1). 2.7. Leaf decomposition A leaf decomposition experiment was carried out from March to May 1998. Senescent yellow leaves which were ready to fall were picked from R. mangle trees. The leaves were split into 24 samples, each one weighing 30 g. Each sample was placed in a separate nylon mesh bag of 16 cm · 21 cm and 1.0 mm mesh size, large enough to permit the entry of small invertebrates, but preventing entry of large consumers, such as crabs, which actually cut the leaves. Twenty one bags were attached to R. mangle roots in contact with surface mud and subjected to intertidal conditions.
Table 1 Biomass and concentrations of phosphorus and nitrogen from the sediment and the R. mangle structures in a mangrove forest in the Potengi estuary Structure
t ha1 [tones ha1]
Phosphorus [mg g1]
Nitrogen [mg g1]
Total biomass Below ground roots Prop roots Trunks Branches Leaves Total sediment (25 cm deep) Total nitrogen Fe/Al-P Ca-P OM-P
107.46 21.50 28.30 36.80 16.45 4.41 2061.94
1.02 ± 0.14 1.35 ± 0.41 1.10 ± 0.30 2.20 ± 0.59 1.76 ± 0.18
3.14 ± 0.62 2.31 ± 0.68 0.66 ± 0.13 1.23 ± 0.64 8.43 ± 1.49
(n = 15) (n = 10) (n = 15) (n = 15) (n = 15)
(n = 15) (n = 10) (n = 15) (n = 15) (n = 15)
2.24 ± 0.80 (n = 25) 0.09 ± 0.02 (n = 50) 0.03 ± 0.01 (n = 50) 0.03 ± 0.01 (n = 50)
528
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
Three bags were retrieved from the site 1, 2, 3, 7, 15, 30, 60 and 90 days after the beginning of the experiment. In the laboratory the sediment was removed from the leaves by gently rubbing the leaves under running tap water and the samples were dried to constant weight for 96 h at 60 C. After these procedures the samples were prepared for nutrient analysis. A regression model quadratic equation (Eq. (1)) with standard error of 0.48 for subsequent comparison with the dry weight of field results were evaluated from 26 fresh samples from 17 g to 37 g. The fresh leaf material were dried for 96 h until reaching a constant weight at 60 C and related to their respective dry weight: dw ¼ 0:3742fw 0:0019fw2 ;
dw ¼ dry weight; fw ¼ fresh weight:
ð1Þ
Percentage of dry weight loss was plotted against time in order to evaluate the half life (the time required by the leaves to lose one half of their dry weight) of decomposition of the R. mangle leaves. The rate at which dry weight was lost was calculated using a simple exponential model, W t ¼ W 0 ekt ; in which Wt is the remaining dry weight at time (t), W0 is the initial leaf dry weight, k is the decay constant and e is the base of the natural logarithm (Olson, 1963). This model has been frequently used in mangrove decomposition studies (Citron and Schaefer-Novelli, 1983). Independent duplicate analyses of nutrients from mixed leaf samples (three bags) were made on 1.0 g samples. The samples were oven dried (60 C, 96 h) and ashed (450 C, 24 h). 2.8. Export of P and N Through leaf litter The amount of R. mangle leaves exported in the form of macro-detritus was assumed to be 7% (0.42 t ha1 yr1) of the total leaf fall according to Silva et al. (1998) who evaluated the macro-detritus balance in a mangrove forest in Sepetiba Bay (Brazil). This estimate agrees with the direct measurements made by Silva et al. (1993) in Sepetiba Bay. These authors sampled macro-detritus using a floating net fixed in two points in the main tidal creek that drains the system. Sepetiba Bay has a similar inundation regime as the forest studied here, where the intertidal zone is inundated approximately 60–70% during the year (Aragon, 1997; Souza, 1999). By considering the concentrations of P (1.21 mg g1) and N (3.83 mg g1) in 15-day leaf detritus (the residence time of leaf litter on the surface of the forest floor, Silva et al., 2006) it was possible to calculate an average export of P and N through leaf litter by multiplying these nutrient concentrations by the mass of leaves exported in the form of macro-detritus. 2.9. Biological analysis Different parts of R. mangle trees (leaves, branches, trunks, prop roots and below ground roots) were analyzed to determine N and P concentrations. The field samples were placed in a plastic bag and taken to the laboratory. They were washed with tap water, oven dried at 60 C until constant weight and powdered. Total P concentrations were measured after ashing (450 C during 24 h) and digestion with persulphate (Grasshoff et al., 1983). Total nitrogen in sediment samples was determined only in a half of the cores by sulfuric acid (H2SO4) digestion employing salicylic acid in the indophenol blue formation at 350 C according to Nogueira et al. (1992). All samples were run in duplicate (differences among duplicates were always less than 5%), and all digestion procedures were tested on reference material (IEAE-336 ‘‘lichen’’). Differences between certified and measured results were always less than 20%. To calculate likely errors in the chemical analysis, the coefficient of variation (CV) was calculated for the concentrations measured from 10 replicates of the same sediment sampled and four replicates of the same biological samples. The results are given in Table 2 and showing a satisfactory precision of analyses. 2.10. Statistical analyses Two sets of statistical tests were employed: parametric tests – analysis of variance (ANOVA) and Tukey test (ANOVA), and a non-parametric test – Kruskal–Wallis analysis (K–W), according to whether or not the data were normally distributed. Statistica (Statsoft) was used for all statistical analyses.
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
529
Table 2 Measurement variability of different analyses made in 10 replicates at the same mangrove sediment sample and R. mangle leaves sample Measurements
Coefficient of variation found in this work
Methodology used
Sediment – Fe/Al-P Sediment – Ca-P Sediment – OM-P Leaves – P Leaves – N
1.64 9.78 16.83 2.38 2.56
Kurmies (1972) Kurmies (1972) Leeg and Black (1955) Grasshoff et al. (1983) Nogueira et al. (1992)
3. Results and discussion 3.1. Reservoir of P and N The main reservoir for total P and total N was the sediment containing 309 kg ha1 and 4619 kg ha1 respectively (77% and 95% of the total habitat contents-sediments and R. mangle trees) (Table 3). In the sediment Fe/Al-P accounted for 60% while OM-P and Ca-P both accounted for 20%. These proportions are not in agreement with those found in Sepetiba Bay by Silva et al. (1998) where Fe/Al-P and OM-P accounted for 21.2% while Ca-P accounted for 57.3%. Fabre et al. (1999) showed that percentage of FeP was larger than 75% and Ca-P fraction represented only 19% of the total P in the mangrove sediment in French Guiana. Silva and Sampaio (1998) found that almost 70% of the sum of fractions of phosphate in the low floodplain was on Fe/Al-P. These different proportions show different availability of P under the redox (Eh) and pH variations observed in the mangroves sediment (Nriagu, 1976; Aragon, 1997). Fe/Al-P is the principal fraction in this study it is also the main source of inorganic fraction of P to R. mangle under Eh changes (Caraco et al., 1989; Silva and Mozeto, 1997). Although OM-P accounted for only 20% in the sediment, it is another potential source of P to R. mangle trees through decomposition of organic matter over time. The reservoir of total P and total N in the R. mangle trees accounted for 145 ± 14 kg ha1 and 216 ± 23 kg ha1 representing 23% and 5% of the total reservoir (sediment + R. mangle trees). The below ground roots (68 ± 15 kg ha1) and prop roots (65 ± 20 kg ha1) of R. mangle retained the highest amounts of N (Tukey, p < 0.05). Tree trunks were the principal reservoir for P (41 ± 11 kg ha1) (Tukey, p < 0.05). Table 3 shows the relative contribution for each structure in R. mangle trees. Silva et al. (1998) argued that leaves are the main reservoir of P. This selective storage has not occurred in this work for P and N. The perennial tissues of the R. mangle trees in this work are the main reservoir for P and N. R. mangle leaves, yellowing during senescence, translocate nutrients before abscission which could explain the different P reservoirs found between the data from Silva et al. (1998) and this work possibly due to the different P availability of sediment source. Allocation of nutrients appears to be affected most heavily by the biomass of the tissues. Li (1997) studying nutrient dynamics in a mangrove forest dominated by Aegiceras corniculatum and Kandelia candel found the perennial tissues to be the principal stocks for N and P. Silva et al. (1990) found similar results for heavy metals. The mechanisms responsible for these different results are not clear and requires more investigation.
Table 3 Phosphorus and nitrogen percentages in the sediment and in R. mangle trees structures Compartment
P (%)
N (%)
Leaf Branch Trunk Prop root Below ground root Sediment
1.15 6.44 5.98 5.98 3.45 77
0.85 0.45 0.55 1.55 1.60 95
530
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
3.2. Litter fall from R. mangle to the sediment The average monthly rates of litter fall (leaves + branches + reproductive structures) are presented in Fig. 2. Significant seasonal variation (K–W, p < 0.05) in litter fall is attributed to both variations in leaf fall and inputs from other structures. The average litter fall value obtained (12.3 ± 1.8 t ha1 yr1) in this study is in the higher range of values recorded for some estuarine mangroves (average 9–12 t ha1 yr1; Amarasinghe and Balasubramaniam, 1992; Wafar et al., 1997). This high primary productivity can be attributed to high litter degradation rates and efficient recycling of nutrients which are supplied by allochthonous inputs from natural and anthropogenic sources (Bouillon et al., 2002). The annual transfer of P and N to the sediment by leaf fall was estimated to be 12 kg ha1 yr1 and 31 kg ha1 yr1 respectively. The highest rates of transfer of P and N from R. mangle trees to the sediment (1.5 ± 0.2 kg ha1 month1 and 3.7 ± 0.9 kg ha1 month1 respectively, for the two elements) were recorded in the late dry season (August, September, October, November and December), but did not correlate with strong winds and rains prevailing during this period (r = 0.02, p < 0.05,r = 0.27, p < 0.05 respectively). 3.3. Decomposition and nutrient release The leaves in the litterbags lost 50% of their original dry weight in 32 days. This degradation rate is high when compared with results from similar experiments (Table 4). The rates of weight loss found in this research are in the range reported by other studies in mangroves, which recorded losses in field studies ranging from 13% to 44% during the first 10 days of decomposition. Changes in original nutrient concentrations are due to the effects of leaching, decomposition, accumulative adsorption process, and microorganism growth (Nielsen and Andersen, 2003). Leaching is a natural process in which leaves on the mangrove sediment lose mass due to chemical and physical processes. When leaves fall to the sediment and are exposed to water, there appears to be a rapid loss of dissolved organic matter and phosphate. The R. mangle leaves lost 20% of their initial weight during the first 10 days. This initial fast loss was attributable to leaching of dissolved organic matter (Ashton et al., 1999) and in agreement with Bosire et al. (2005) and Middleton and McKee (2001). P concentrations in the R. mangle leaves decreased 26% in relation to original concentrations during the leaching phase. After 15 days concentrations increased substantially for P and N (Fig. 3). Nielsen and Andersen (2003) found the same behavior for P during decomposition
Leaf
Branch
Reproductive Parts
1
0.5
Jul
Jun
May
Apr
Mar
Feb
Jan
Dec
Nov
Oct
Sep
0
Ago
Litterfall (t ha-1 month-1)
1.5
Time 97
98
Fig. 2. Seasonal litter fall rate of the structures of R. mangle trees in the Potengi estuary.
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
531
Table 4 Loss of 50% dry weight in leaves of red mangrove Site
Species
Days
Author
Kenya Thailand USA USA USA Malaysia Belize Panama Brazil Brazil Brazil
R. R. R. R. R. R. R. R. R. R. R.
14a–35b 40 180 45 60 10 35 27 77 23a–45b 32
Bosire et al. (2005) Boonruang (1978) Davis et al. (2003) Heald (1971) Fell (1974) Ashton et al. (1999) Middleton and McKee (2001) D’croz et al. (1989) Silva (1988)
a b
mucronata apiculata mangle mangle mangle apiculata mangle mangle mangle mangle mangle
This work
Wet season. Dry season.
of Rhizophora apiculata leaves in sediments. Bosire et al. (2005) recorded an increase of N concentrations throughout the sampling period (6 weeks) in all mangrove species during their experiment. The high concentrations of P (2.0 mg g1) and N (10.4 mg g1) in 90-day leaf detritus found in our study in R. mangle leaves were higher than those values during the initial 15 days (Fig. 3). The increase in N concentration in red mangrove leaves is probably due to a progressive reduction in the amount of organic C and immobilization of N by fungi and bacteria (Davis et al., 2003; Bosire et al., 2005). 3.4. Standing stock of leaf litter and residence time of leaves on sediments There was a significant change (Tukey, p < 0.05) in the standing stock of leaves (dry weight) on the surface forest floor over the 6 months measured (0.2 ± 0.1–1.0 ± 0.4 t ha1). August (1997) presented the highest values 1.0 ± 0.4 t ha1. This highest biomass of leaves on the floor does not agree with the peak of leaf fall (December 1997, Fig. 3), nor with the smallest frequency of sediment flooding along the year (November 1997) (9.0%). Perhaps a relative lack of activity by crabs, that can break down and remove litter from the forest floor (Middleton and McKee, 2001; Ashton, 2002), is responsible for this highest biomass of leaves on the floor in August 1997. The mean standing stock of leaf litter was 0.4 t ha1. Therefore the estimated mean residence time was 15 days (Silva et al., 2006), assuming a mean annual leaf litter fall rate of 9.5 t ha1 yr1. 11
2.2
10 9 8
1.8
7 1.6
6
N (mgg-1)
P (mg g-1)
2.0
5
1.4
4 1.2 1.0
3 0
15
30
45 Days
60
75
90
2
Fig. 3. Temporal changes in the average concentrations of P and N during leaf decomposition of R. mangle. Intervals.
532
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
Mangrove litter residence time is quite variable depending mostly on the tidal range (Ong et al., 1982; Twilley, 1985), the elevation of the forest relative to inundation (Twilley et al., 1986) and the activity of litter-feeding organisms in particular of seasarmid crabs (Robertson, 1986). Twilley et al. (1986) reported a residence time of 2–6 months in a basin forest. This long residence time is due to infrequent flooding of a basin forest, while our forest is a fringe type which is flooded nearly twice every day. In riverine Rhizophora forests, Ong et al. (1982) found residence time of leaves of about 30 days. Robertson (1986) showed that crabs can bury over 75% of total litter fall at low tide, speeding the ‘‘export’’ of deposited leaves. In our forest, the activity of seasarmid crabs, evaluated visually, is very high and may contribute to the short residence time of leaves on the sediment. The turnover rate was obtained by a ratio of the N and P in the biomass of R. mangle trees to the average transfer of the element through yearly leaf fall. The turnover rates of P and N in this mangrove forest were calculated to be 12 years for P and 7 years for N. The more rapid cycling of N than P may be due to the higher availability of N (Prescott et al., 2000). Our data are similar to those of Li (1997) whose values were 10 years for P and 7 years for N. On the other hand our forest has a smaller (four times) N turnover rate than that found by Gong and Ong (1990) for R. apiculata. Turnover rate is influenced by the biomass of trees, litter fall, and the concentrations of nutrients in the leaves and in the sediment. 3.5. Export of P and N Associated with leaf litter The export of N and P from the mangrove forest to coastal marine systems through leaf litter is a function of the amount of leaf litter exported and its nutrient concentrations. Nutrient concentrations depend on the residence time of leaf litter on the forest floor and the enrichment of nutrients during leaf litter decomposition. The short residence time (15 days) of deposited leaf litter on the mud surface in our forest limits the enrichment of N and P concentrations in decomposing leaves exported, since the nutrient concentrations in this study increased only after at least 30 days of decomposition. Therefore, the concentration of P and N in leaves during export is equal to or lower than the concentration in recently fallen leaves. The estimated average export rates of P and N through leaf litter are 0.5 kg ha1 yr1 and 1.6 kg ha1 yr1 respectively, for the two elements (Fig. 4).
Fig. 4. Dynamics of N and P in the Potengi experimental forest (PEF).
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
533
4. Conclusions The dynamics of N and P in the red mangrove forest studied here is summarized in Fig. 4. Despite the main reservoir of P and N being the sediment, the R. mangle trees play an important role in the cycling of P and N in the Potengi estuary. Decomposing leaf litter does not contribute significantly with nutrient output to estuarine waters since the leaf litter only becomes enriched (P – 1.3 times and N – 4.0 times from initial concentrations) in P and N at the end of the leaf decomposition process (90 days). The short residence time (15 days) makes it clear that the budget of N and P in the leaves is retained in the mangrove forest. The export of nutrients through R. mangle leaf fall is small; these fluxes represent 3% for P and only 0.6% for N of the nutrient stock in the sediment, or 26 and 149 years of growth for P and N. The annual export of nutrients as leaf macrodetritus produced by R. mangle trees is equivalent to 0.1% and 0.03% of the total stock of P and N in the mangrove forest (biomass and sediment). On the other hand, nutrient export from this mangrove in suspended materials phase and the dynamics of groundwater fluxes into the estuary, which was not studied here, could contribute significantly to nutrient export out of the mangrove ecosystem into the estuary. Tropical mangrove forests are probably efficient biogeochemical barriers to the transport of P and N in tropical coastal areas. Hence mangroves could be an important and efficient ally against eutrophication in tropical coastal areas. Acknowledgements The authors wish to acknowledge the International Foundation of Science (IFS-D/2382-1) and Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico (CNPq, Process 100515/1994-0) of Brazil for grant support. The first author is grateful to Professor Philip S. Rainbow (Head of Department of Zoology – The Natural History Museum of London) and Professor Sternberg L.S.L. (Department of Biology, University of Miami) for English corrections. References Amarasinghe, M.D., Balasubramaniam, S., 1992. Net primary productivity of two mangrove forest stands on the northwestern coast of Sri Lanka. Hydrobiologia 247, 37–47. Aragon, G.T., 1997. Biogeoquı´mica sedimentar do ferro e do enxofre em um manguezal da Baı´a de Sepetiba. Ph.D. Thesis. Neoformac¸a˜o de sulfetos ferrosos, Universidade Federal Fluminense, Nitero´i, RJ, 119p. Ashton, E.C., Hogarth, P.J., Ormond, R., 1999. Breakdown of mangrove leaf litter in a managed mangrove forest in Peninsular Malaysia. Hydrobiologia 413, 77–88. Ashton, E.C., 2002. Mangrove sesarmid crab feeding experiments in Peninsular Malaysia. Journal of Experimental Marine Biology and Ecology 273, 97–119. Boonruang, G.P., 1978. The degradation rates of mangrove leaves of Rhizophora apiculata (Bl.) and Avicennia marina (Forsk.) Vierh. at Phuket Island, Thailand. Phuket Marine Biological Centre 26, 1–7. Bosire, J.O., Dahdouh-Guebas, F., Kairo, J.G., Kazungu, J., Dehairs, F., Koedam, N., 2005. Litter degradation and CN dynamics in reforested mangrove plantations at Gazi Bay, Kenya. Biological Conservation 126, 287–295. Bouillon, S., Koedam, N., Raman, A.V., Dehairs, F., 2002. Primary producers sustaining macro-invertebrate communities in intertidal mangrove forests. Oecologia 130, 441–448. Caraco, N.F., Cole, J.J., Likens, G.E., 1989. Evidence for sulphate controlled phosphorus release from sediments of aquatic systems. Nature 341, 316–318. Citron, G., Schaefer-Novelli, Y., 1983. Introduccio´n a la ecologı´a del manglar. UNESCO/ROSTLAC, Montevideo, p. 109. Davis III, S.E., Molina, C.C., Childers, D.L., Day Jr., J.W., 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 73, 199–215. D’croz, L., Delrosario, J., Holness, R., 1989. Mangrove (Rhizophora mangle L.) leaves in the Bay of Panama. Revista de Biologia Tropical 37, 101–104. Fabre, A., Fromard, Fr., Trichon, V., 1999. Fractionation of phosphate in sediments of four representative mangrove stages (French Guiana). Hydrobiologia 392, 13–19. Fell, J.W., 1974. Microbial activities in the mangrove (Rhizophora mangle) leaf detrital system. In: Proc. Int. Symp. Biol. Management of Mangrove, pp. 661–679. Ferreira, M.F., Chiu, W.S., Cheok, H.K., Cheang, F., Sun, W., 1996. Accumulation of nutrients and heavy metals in surface sediments near Macao. Marine Pollution Bulletin 32 (5), 420–426.
534
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
Gong, W.K., Ong, J.E., 1990. Plant biomass and nutrient flux in a managed mangrove forest in Malaysia. Estuarine Coastal and Shelf Science 31, 519–530. Grasshoff, K., Ehrhardt, M., Kremling, K., 1983. Methods of Seawater Analysis. Verlag Chimie, Nurnberg, 419 p. Heald, E.J., 1971. The production of organic detritus in a south Florida estuary. Sea Grant Technical Bulletin No. 6. University of Miami, Coral Gables, FL, USA. Kennish, M.J., 1991. Ecology of Estuaries: Anthropogenic Effects. CRC, FL, 494 p. Kurmies, B., 1972. Zur Fraktionierung der Bodenphosphate. Phosphosa¨ure 29, 118–149. Lacerda, L.D., Conde, J.E., Alarco´n, C., Bodero, A., Bacon, P., Alvarez-Leo´n, R., D’croz, L., Kjerfve, B., Olanı´a, J., Vannucci, M., 1993. Mangroves of Latin America and the Caribbean, 1–42. In: Lacerda, L.D. (Ed.), Conservation and Sustainable Utilization of Mangroves Forests in Latin America and Africa. Part I. Latin America. International Society for Mangrove Ecosystems, ISME Tech. Rep. no. 2 Okinawa. Leeg, J.O., Black, C.A., 1955. Determination of organic phosphorus in sediments: II ignition method. Sediment Science Society Proceedings 19, 139–143. Li, M.S., 1997. Nutrient dynamics of a Futian mangrove forest in Shenzhen, south China. Estuarine, Coastal and Shelf Science 45, 463– 472. Middleton, B.A., McKee, K.L., 2001. Degradation of mangrove tissues and implications for peat formation in Belizean island forests. Journal of Ecology 89, 818–828. Nielsen, T., Andersen, F.ø., 2003. Phosphorus dynamics during decomposition of mangrove (Rhizophora apiculata) leaves in sediments. Journal of Experimental Marine Biology and Ecology 293, 73–88. Nogueira, A.R.A., Souza, G.B., Batista, L.A.R., 1992. Determinac¸a˜o espectrofotome´trica de nitrogeˆnio em digeridos de plantas em sistema de ana´lise por injec¸a˜o em fluxo. Quı´mica Nova 19 (1), 33–36. Nriagu, J.O., 1976. Phosphate–clay mineral relations in soils and sediments. Canadian Journal of Earth Sciences 13, 717–736. Olson, J.S., 1963. Energy storage and the balance of producers and decomposers in ecological systems. Ecology 42 (2), 322–330. Ong, J.E., Gong, W.K., Wong, C.H., 1982. Productivity and nutrient status of litter in a managed mangrove forest in Malaysia. In: Kostermans, A.V., Sastroutomo, S.S. (Eds.), Proc. Symp. Bangor, Indonesia, Biotropica, vol. 17, pp. 33–41. Por, E.D., 1984. The ecosystem of the mangal: general considerationsHydrobiology of the Mangal. Junk Publishers, pp. 1–15. Prescott, C.E., Chappell, H.N., Vesterdal, L., 2000. Nitrogen turnover in forest floors of coastal Douglas-Fir at sites differing in soil nitrogen capital. Ecology 81 (7), 1878–1886. Robertson, A.I., 1986. Leaf-burying crabs: their influence on energy flow and transport from mixed mangrove forests (Rhizophora spp.) in northeastern Australia. Journal of Experimental Marine Biology and Ecology 102, 237–248. Silva, C.A.R., Lacerda, L.D., Rezende, C.E., 1990. Heavy metal reservoirs in a red mangrove forest. Biotropica 22 (4), 339–345. Silva, L.F.F., Lacerda, L.D., Ovalle, A.R.C., Rezende, C.E., Silva, C.A.R., Carvalho, C.E.V., 1993. Dinaˆmica de macrodetritos em um ecossistema de manguezal, Baı´a de Sepetiba, RJ. Ann III Symp. Ecossistemas da Costa Brasileira, Serra Negra. Academia de Cieˆncias do Estado de Sa˜o Paulo, vol. 1, pp. 204–211. Silva, C.A.R., 1996. Manguezal: Ecossistema egoı´sta ou benevolente? Cieˆncia Hoje 20 (120), 6–11. Silva, C.A.R., Mozeto, A.A., 1997. Release and retention of phosphorus in mangrove sediments: Sepetiba bay, Brazil. Mangrove Ecosystem Studies in Latin America and Africa, UNESCO. Silva, C.A.R., Rego, R.D.P., 1997. Aspectos estruturais e distribuic¸a˜o de biomassa em um mangue vermelho (Rhizophora mangle), Natal, RN. In: Anais do VIII Semina´rio Regional de Ecologia, VIII, pp. 285–293. Silva, C.A.R., 1988. Distribuic¸a˜o e ciclagem interna de metais pesados em um ecossitema de manguezal dominado por Rhizophora mangle, Baı´a de Sepetiba, Rio de Janeiro. M.Sc. Thesis, Universidade Federal Fluminense, Nitero´i, 67p. Silva, C.A.R., Sampaio, L.S., 1998. Speciation of phosphorus in a tidal floodplain forest in the Amazon estuary. Mangroves and Salt Marshes 02 (01), 59–62. Silva, C.A.R., Mozeto, A.A., Ovalle, A.R.C., 1998. Distribution and fluxes as macrodetritus of phosphorus in red mangroves, Sepetiba Bay, Brazil. Mangroves and Salt Marshes 02 (01), 37–42. Silva, C.A.R., Silva, A.P., Oliveira, S.R., 2006. Concentration, stock and transport rate of heavy metals in a tropical red mangrove, NatalBrazil. Marine Chemistry (99), 2–11. ´ reas de Manguezal do Estua´rio Potengi/RN, Nordeste, Brasil. M.Sc. Thesis. Dissertac¸a˜o, Souza, F.E.S., 1999. Estudo de Valorac¸a˜o das A Universidade Federal do Rio Grande do Norte, Natal, 87p. Tam, N.F.Y., Wong, Y.S., 1995. Mangrove sediments as sinks for wastewater-borne pollutants. Hydrobiologia 295, 231–241. Twilley, R.R., 1985. The exchange of organic carbon in basin mangrove forests in a southwest Florida estuary. Estuarine, Coastal and Shelf Science 20, 543–557. Twilley, R.R., Lugo, A.E., Patterson- Zucca, C., 1986. Litter production and turnover in basin mangrove forests in southwest Florida. Ecology 67, 670–683. Wafar, S., Untawale, A.G., Wafar, M., 1997. Litter fall and energy flux in mangrove ecosystem. Estuarine, Coastal and Shelf Science 44, 111–124.
Dynamics of phosphorus and nitrogen through litter fall and decomposition in a tropical mangrove forest Carlos A. Ramos e Silva a
a,*
, Se´rgio R. Oliveira a,b, Ronaldo D.P. Reˆgo Antonio A. Mozeto c
a,c
,
Departamento de Oceanografia e Limnologia (Laborato´rio de Biogeoquı´mica Ambiental), Universidade Federal do Rio Grande do Norte, Natal, RN, Cxa. Postal 1202, 59075-970, Brazil b Laborato´rio de Cieˆncias Ambientais, Universidade Estadual do Norte Fluminense, Campos do Goytacazes, RJ, Brazil c Departamento de Quı´mica, Universidade Federal de Sa˜o Carlos, Sa˜o Carlos, SP 13565-905, Brazil Received 14 November 2005; received in revised form 20 June 2006; accepted 20 April 2007
Abstract Distribution, dynamics and mass budget of phosphorus and nitrogen in a red mangrove forest were studied in the Potengi mangrove forest in northern Brazil (lat. 542 0 and 553 0 S, long. 355 0 and 3525 0 W). Tidal hydrology, net primary productivity, leaf litter decomposition rate and standing stock of leaf litter in a red mangrove forest were measured. The results showed that the main reservoir for total P and total N was the sediment with 309 kg ha1 and 4619 kg ha1 (77% and 95% of the total P and N content in the mangrove forest), respectively, for the two elements. Total P and total N in Rhizophora mangle trees accounted for 145 ± 14 kg ha1 and 216 ± 23 kg ha1 (23% and 5% of the total P and N in the mangrove forest). The estimated average export rates for P and N through leaf litter are 0.5 kg ha1 yr1 and 1.6 kg ha1 yr1 respectively. Our measurements support previous results in concluding that mangrove forests efficiently retain P and N. 2007 Elsevier Ltd. All rights reserved. Keywords: Rhizophora mangle; Phosphorus; Nitrogen; Mangrove; Litter fall
1. Introduction Mangrove forests may be considered as a ‘‘chemical reactors’’ not only because of their physiological and biochemical processes but also due to their active influence in the mobility of nutrients (Lacerda et al., 1993; Silva, 1996; Silva and Mozeto, 1997; Silva et al., 1998; Silva and Sampaio, 1998; Nielsen and Andersen, 2003; Bosire et al., 2005). This influence is affected by tidal hydrology, litter fall production, leaf decomposition rate, and standing stock and the physical–chemical composition of local sediment.
*
Corresponding author. Tel.: +55 84 32154433; fax: +55 84 32023004. E-mail address: [email protected] (C.A. Ramos e Silva).
0141-1136/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.marenvres.2007.04.007
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
525
The addition of sewage effluent with high nutrient concentrations has influenced estuaries in various tropical regions. Of these nutrients, phosphorus and nitrogen have received attention due to their effects on estuarine organisms (Kennish, 1991; Ferreira et al., 1996). Previous studies have proposed that mangrove forests not only resist the effect of excess nutrients from sewage, but also decrease their concentrations (Por, 1984; Tam and Wong, 1995). However, few quantitative studies have been carried out on the role of litter production, decomposition and export in P and N cycling in mangrove ecosystems. The cycling of organic matter through litter production, decomposition and tidal transport, may eventually export a fraction of the accumulated P and N to local estuarine waters and increase the load of these nutrients in these estuarine waters. The objective of the present study was to quantify the cycling of phosphorus and nitrogen through litter detritus in a red mangrove (Rhizophora mangle) forest in Natal, near the city of Rio Grande do Norte, a moderately polluted area. 2. Materials and methods 2.1. Study sites The study was carried out in the Potengi experimental forest (PEF), located in the Potengi estuary, State of Rio Grande do Norte, northern Brazil (lat. 542 0 and 5 53 0 S, long. 355 0 and 3525 0 W) (Fig. 1). The forest has an area of approximately 863 ha, and is limited by two tidal creeks running almost perpendicular to the shore. The average temperature ranges between 24.6 C in the rainy season (March–August) and 28.8 C in the dry season (September–February). In the rainy season rainfall is about 1390 mm, while in the dry season it is about 220 mm. The dominant species is the red mangrove R. mangle although isolated trees of the black mangrove (Avicennia schaueriana) and of the white mangrove (Laguncularia racemosa) occur throughout the forest representing less than 10% respectively of the total coverage (Souza, 1999). Average R. mangle tree density is 3200 trees ha1, with an average tree height of 5.4 m and an average trunk diameter of 6.2 cm. The total basal areas are 9.5 m2 ha1 (Silva and Rego, 1997). The waters of the Potengi, Jundiaı´ and Doce Rivers form the estuary. The Potengi River, which drains a catchment of 3180 km2, has a total course of 176 km and a discharge of approximately 5 m3 s1 during the rainy season (March–August). The Jundiaı´ River has an independent catchment with a discharge smaller than that of the Potengi River. With this smaller discharge the Jundiaı´ is more influenced by tides, the effects of which have been observed as far upstream as the city of Macaı´ba, 30 km from the mouth of the estuary. The Doce River comprises the smallest freshwater input to the estuary with a discharge of approximately 2 m3 s1 during most of the year. Water temperatures vary from 26.5 to 29.0 C during the year.
Fig. 1. Location of the Potengi estuary in Natal, Rio Grande do Norte State, NE Brazil.
526
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
The Potengi estuary extends for about 25 km with a width ranging between 400 and 600 m. The estuary is bordered by three cities with a total population of more than 1,000,000 inhabitants. One of the cities, Natal, discharges more than 60% of its untreated sewage directly into the Potengi river (15,000 m3 per day – 14 mg L1 of total phosphorus and 50 mg L1 of total nitrogen) and there are several other point sources of organic effluent along the estuary. The mean concentration of phosphorus ranges from 2.10 to 3.70 lM for PO4 2 3 ðH3 PO4 þ H2 PO 4 þ HPO4 þ PO4 Þ and from 4.20 to 7.90 lM for total dissolved phosphorus (TDP) (Silva, þ unpublished data). Total nitrogen ðNO 3 þ NO2 þ NH3 þ NH4 Þ are also found in high concentrations in the estuary studied, where the mean concentrations range from 23.1 to 108.7 lM (Silva, unpublished data). On the west side of the estuary the mangrove forests have been removed and replaced by shrimp ponds. 2.2. Collection and preparation of samples 2.2.1. Sediment samples Ten sediment cores were collected in a single time (February, 1997) to a depth of 25 cm (which represent the highest biomass of roots) from the mudflat, sliced at 5-cm intervals and each section placed in clean plastic bag and immediately brought to the laboratory in ice buckets. In the laboratory sediment samples were oven dried at 60 C for 72 h, sieved (<1 mm mesh) and ground to fine powder and placed in glass vials for analysis. To obtain the fraction of phosphate binding to iron and aluminum (Fe/Al-P) and binding to calcium (Ca-P), 1 g of dry sediment was washed in an alcoholic solution of 15 mL 1 N KCl to eliminate calcium ions in the sediment. The first phase (determination of Fe/Al-P) consisted of an extraction with 15 mL 1 N NaOH + 1 N Na2SO4 before the next phase (determination of Ca-P) which consisted of extractions of phosphorus with 15 mL 1 N H2SO4; the residual was washed in 4% Na2SO4 (Kurmies, 1972). The total inorganic phosphorus (TIP) concentrations were obtained by adding the Fe/Al-P and Ca-P concentrations. The concentrations of the phosphorus fraction were measured by the colorimetric method (Grasshoff et al., 1983). Leeg and Black’s method (1955) was used to determine the concentrations of phosphate binding to organic matter (OM-P) by colorimetry (Grasshoff et al., 1983). The OM-P fraction was calculated by subtracting the TIP concentrations from the total phosphorus concentration. Total nitrogen in sediment samples was determined only in a half of the cores by sulfuric acid (H2SO4) digestion employing salicylic acid in the indophenol blue formation at 350 C according to Nogueira et al. (1992). 2.3. Standing stocks of phosphorus and nitrogen in R. mangle trees and litterfall The working units were seven quadrates of 100 m2 each, established perpendicular to the sea, approximately 45 m from the Doce River and 18 m from the forest seaward edge. In each quadrate all trees were counted. Height and diameter were measured in half of the trees in each quadrate and the results were then extrapolated to 1 ha (Citron and Schaefer-Novelli, 1983). For the estimation of total aboveground biomass a regression curve was drawn using the total dry weight (trunks, branches, prop roots and leaves) and diameter of eight cut trees, with diameters ranging from 2.1 to 18.7 cm. From the curve, estimates of the weight of all trees in the quadrates could be obtained and then summed up. This was possible since the total aboveground biomass was significantly correlated with tree diameter. Different parts of the R. mangle trees (leaves, branches, trunks, prop roots) were analyzed to determine the P and N concentrations. Plant samples were collected from 10 trees representative of the dominant class of tree diameter in the forest. The field samples were packed in plastic bags and taken to the laboratory. They were washed with tap water, oven dried (60 C for 96 h) and ground for nutrient analysis. The standing stocks of P and N in R. mangle trees were calculated by multiplying nutrient concentrations in each compartment by its biomass in 1 ha. The biomass of belowground roots was estimated by digging in the sediment to 25 cm depth through eight squares of 0.25 m2 random distributed through the study area. In the field, samples (fine and gross roots together) were weighed on manual scales to obtain the wet biomass. In the laboratory samples were oven dried at 60 C for 96 h and weighed. A relationship between wet weight and dry weight was obtained and the average dry biomass was extrapolated to 1 ha (Silva et al., 1990).
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
527
Litter fall was collected in seventeen 1.0 m2 baskets (1 mm nylon mesh) supported approximately 1.5 m above the ground. Litter fall was collected at 30 day intervals from August 1997 to July 1998. The material from each basket was dried at 60 C up to constant weight and separated into leaves, reproductive parts and thin branches (lesser than 2 cm diameter). Only the leaves were used for analysis of P and N. 2.4. Reservoir of phosphorus and nitrogen in sediment In the sediment the P and N contents were determined by multiplying the P and N concentrations by the total dry weight of sediment to a depth of 25 cm (see data from Table 1). 2.5. Standing crop of phosphorus and nitrogen in the leaf litter The standing crop of leaf litter was measured over 6 months, always during neap low tide, by collecting each month 15 0.25 m2 (0.5 m · 0.5 m) samples of litter on the surface of the forest floor in the vicinity of the litter fall collectors. Surface leaf samples were processed in the same manner as litter fall samples. In the laboratory the sediment was removed from the leaves by gently rubbing the leaves in running tap water and the samples were dried at 60 C per 96 h. After these procedures, the samples were prepared for phosphorus and nitrogen analysis. The reservoir of nutrients in the leaf litter was calculated by multiplying the standing stock of leaves (0.4 ± 0.3 t ha1, n = 90) by the concentration of P and N in R. mangle leaves from 15 to 90 days of decomposition (1.53 ± 0.29 mg g1 and 2.87 ± 1.24 mg g1 respectively, for the two elements). 2.6. Transfer of phosphorus and nitrogen from R. mangle trees to the sediment The transfer of nutrients through R. mangle leaf fall to the sediment was calculated by multiplying the mean concentration (mg g1), of nutrients in leaf fall (P = 1.16 ± 0.28 and N = 3.08 ± 0.64) by the mean leaf fall rate (1000 kg ha1 month1). 2.7. Leaf decomposition A leaf decomposition experiment was carried out from March to May 1998. Senescent yellow leaves which were ready to fall were picked from R. mangle trees. The leaves were split into 24 samples, each one weighing 30 g. Each sample was placed in a separate nylon mesh bag of 16 cm · 21 cm and 1.0 mm mesh size, large enough to permit the entry of small invertebrates, but preventing entry of large consumers, such as crabs, which actually cut the leaves. Twenty one bags were attached to R. mangle roots in contact with surface mud and subjected to intertidal conditions.
Table 1 Biomass and concentrations of phosphorus and nitrogen from the sediment and the R. mangle structures in a mangrove forest in the Potengi estuary Structure
t ha1 [tones ha1]
Phosphorus [mg g1]
Nitrogen [mg g1]
Total biomass Below ground roots Prop roots Trunks Branches Leaves Total sediment (25 cm deep) Total nitrogen Fe/Al-P Ca-P OM-P
107.46 21.50 28.30 36.80 16.45 4.41 2061.94
1.02 ± 0.14 1.35 ± 0.41 1.10 ± 0.30 2.20 ± 0.59 1.76 ± 0.18
3.14 ± 0.62 2.31 ± 0.68 0.66 ± 0.13 1.23 ± 0.64 8.43 ± 1.49
(n = 15) (n = 10) (n = 15) (n = 15) (n = 15)
(n = 15) (n = 10) (n = 15) (n = 15) (n = 15)
2.24 ± 0.80 (n = 25) 0.09 ± 0.02 (n = 50) 0.03 ± 0.01 (n = 50) 0.03 ± 0.01 (n = 50)
528
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
Three bags were retrieved from the site 1, 2, 3, 7, 15, 30, 60 and 90 days after the beginning of the experiment. In the laboratory the sediment was removed from the leaves by gently rubbing the leaves under running tap water and the samples were dried to constant weight for 96 h at 60 C. After these procedures the samples were prepared for nutrient analysis. A regression model quadratic equation (Eq. (1)) with standard error of 0.48 for subsequent comparison with the dry weight of field results were evaluated from 26 fresh samples from 17 g to 37 g. The fresh leaf material were dried for 96 h until reaching a constant weight at 60 C and related to their respective dry weight: dw ¼ 0:3742fw 0:0019fw2 ;
dw ¼ dry weight; fw ¼ fresh weight:
ð1Þ
Percentage of dry weight loss was plotted against time in order to evaluate the half life (the time required by the leaves to lose one half of their dry weight) of decomposition of the R. mangle leaves. The rate at which dry weight was lost was calculated using a simple exponential model, W t ¼ W 0 ekt ; in which Wt is the remaining dry weight at time (t), W0 is the initial leaf dry weight, k is the decay constant and e is the base of the natural logarithm (Olson, 1963). This model has been frequently used in mangrove decomposition studies (Citron and Schaefer-Novelli, 1983). Independent duplicate analyses of nutrients from mixed leaf samples (three bags) were made on 1.0 g samples. The samples were oven dried (60 C, 96 h) and ashed (450 C, 24 h). 2.8. Export of P and N Through leaf litter The amount of R. mangle leaves exported in the form of macro-detritus was assumed to be 7% (0.42 t ha1 yr1) of the total leaf fall according to Silva et al. (1998) who evaluated the macro-detritus balance in a mangrove forest in Sepetiba Bay (Brazil). This estimate agrees with the direct measurements made by Silva et al. (1993) in Sepetiba Bay. These authors sampled macro-detritus using a floating net fixed in two points in the main tidal creek that drains the system. Sepetiba Bay has a similar inundation regime as the forest studied here, where the intertidal zone is inundated approximately 60–70% during the year (Aragon, 1997; Souza, 1999). By considering the concentrations of P (1.21 mg g1) and N (3.83 mg g1) in 15-day leaf detritus (the residence time of leaf litter on the surface of the forest floor, Silva et al., 2006) it was possible to calculate an average export of P and N through leaf litter by multiplying these nutrient concentrations by the mass of leaves exported in the form of macro-detritus. 2.9. Biological analysis Different parts of R. mangle trees (leaves, branches, trunks, prop roots and below ground roots) were analyzed to determine N and P concentrations. The field samples were placed in a plastic bag and taken to the laboratory. They were washed with tap water, oven dried at 60 C until constant weight and powdered. Total P concentrations were measured after ashing (450 C during 24 h) and digestion with persulphate (Grasshoff et al., 1983). Total nitrogen in sediment samples was determined only in a half of the cores by sulfuric acid (H2SO4) digestion employing salicylic acid in the indophenol blue formation at 350 C according to Nogueira et al. (1992). All samples were run in duplicate (differences among duplicates were always less than 5%), and all digestion procedures were tested on reference material (IEAE-336 ‘‘lichen’’). Differences between certified and measured results were always less than 20%. To calculate likely errors in the chemical analysis, the coefficient of variation (CV) was calculated for the concentrations measured from 10 replicates of the same sediment sampled and four replicates of the same biological samples. The results are given in Table 2 and showing a satisfactory precision of analyses. 2.10. Statistical analyses Two sets of statistical tests were employed: parametric tests – analysis of variance (ANOVA) and Tukey test (ANOVA), and a non-parametric test – Kruskal–Wallis analysis (K–W), according to whether or not the data were normally distributed. Statistica (Statsoft) was used for all statistical analyses.
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
529
Table 2 Measurement variability of different analyses made in 10 replicates at the same mangrove sediment sample and R. mangle leaves sample Measurements
Coefficient of variation found in this work
Methodology used
Sediment – Fe/Al-P Sediment – Ca-P Sediment – OM-P Leaves – P Leaves – N
1.64 9.78 16.83 2.38 2.56
Kurmies (1972) Kurmies (1972) Leeg and Black (1955) Grasshoff et al. (1983) Nogueira et al. (1992)
3. Results and discussion 3.1. Reservoir of P and N The main reservoir for total P and total N was the sediment containing 309 kg ha1 and 4619 kg ha1 respectively (77% and 95% of the total habitat contents-sediments and R. mangle trees) (Table 3). In the sediment Fe/Al-P accounted for 60% while OM-P and Ca-P both accounted for 20%. These proportions are not in agreement with those found in Sepetiba Bay by Silva et al. (1998) where Fe/Al-P and OM-P accounted for 21.2% while Ca-P accounted for 57.3%. Fabre et al. (1999) showed that percentage of FeP was larger than 75% and Ca-P fraction represented only 19% of the total P in the mangrove sediment in French Guiana. Silva and Sampaio (1998) found that almost 70% of the sum of fractions of phosphate in the low floodplain was on Fe/Al-P. These different proportions show different availability of P under the redox (Eh) and pH variations observed in the mangroves sediment (Nriagu, 1976; Aragon, 1997). Fe/Al-P is the principal fraction in this study it is also the main source of inorganic fraction of P to R. mangle under Eh changes (Caraco et al., 1989; Silva and Mozeto, 1997). Although OM-P accounted for only 20% in the sediment, it is another potential source of P to R. mangle trees through decomposition of organic matter over time. The reservoir of total P and total N in the R. mangle trees accounted for 145 ± 14 kg ha1 and 216 ± 23 kg ha1 representing 23% and 5% of the total reservoir (sediment + R. mangle trees). The below ground roots (68 ± 15 kg ha1) and prop roots (65 ± 20 kg ha1) of R. mangle retained the highest amounts of N (Tukey, p < 0.05). Tree trunks were the principal reservoir for P (41 ± 11 kg ha1) (Tukey, p < 0.05). Table 3 shows the relative contribution for each structure in R. mangle trees. Silva et al. (1998) argued that leaves are the main reservoir of P. This selective storage has not occurred in this work for P and N. The perennial tissues of the R. mangle trees in this work are the main reservoir for P and N. R. mangle leaves, yellowing during senescence, translocate nutrients before abscission which could explain the different P reservoirs found between the data from Silva et al. (1998) and this work possibly due to the different P availability of sediment source. Allocation of nutrients appears to be affected most heavily by the biomass of the tissues. Li (1997) studying nutrient dynamics in a mangrove forest dominated by Aegiceras corniculatum and Kandelia candel found the perennial tissues to be the principal stocks for N and P. Silva et al. (1990) found similar results for heavy metals. The mechanisms responsible for these different results are not clear and requires more investigation.
Table 3 Phosphorus and nitrogen percentages in the sediment and in R. mangle trees structures Compartment
P (%)
N (%)
Leaf Branch Trunk Prop root Below ground root Sediment
1.15 6.44 5.98 5.98 3.45 77
0.85 0.45 0.55 1.55 1.60 95
530
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
3.2. Litter fall from R. mangle to the sediment The average monthly rates of litter fall (leaves + branches + reproductive structures) are presented in Fig. 2. Significant seasonal variation (K–W, p < 0.05) in litter fall is attributed to both variations in leaf fall and inputs from other structures. The average litter fall value obtained (12.3 ± 1.8 t ha1 yr1) in this study is in the higher range of values recorded for some estuarine mangroves (average 9–12 t ha1 yr1; Amarasinghe and Balasubramaniam, 1992; Wafar et al., 1997). This high primary productivity can be attributed to high litter degradation rates and efficient recycling of nutrients which are supplied by allochthonous inputs from natural and anthropogenic sources (Bouillon et al., 2002). The annual transfer of P and N to the sediment by leaf fall was estimated to be 12 kg ha1 yr1 and 31 kg ha1 yr1 respectively. The highest rates of transfer of P and N from R. mangle trees to the sediment (1.5 ± 0.2 kg ha1 month1 and 3.7 ± 0.9 kg ha1 month1 respectively, for the two elements) were recorded in the late dry season (August, September, October, November and December), but did not correlate with strong winds and rains prevailing during this period (r = 0.02, p < 0.05,r = 0.27, p < 0.05 respectively). 3.3. Decomposition and nutrient release The leaves in the litterbags lost 50% of their original dry weight in 32 days. This degradation rate is high when compared with results from similar experiments (Table 4). The rates of weight loss found in this research are in the range reported by other studies in mangroves, which recorded losses in field studies ranging from 13% to 44% during the first 10 days of decomposition. Changes in original nutrient concentrations are due to the effects of leaching, decomposition, accumulative adsorption process, and microorganism growth (Nielsen and Andersen, 2003). Leaching is a natural process in which leaves on the mangrove sediment lose mass due to chemical and physical processes. When leaves fall to the sediment and are exposed to water, there appears to be a rapid loss of dissolved organic matter and phosphate. The R. mangle leaves lost 20% of their initial weight during the first 10 days. This initial fast loss was attributable to leaching of dissolved organic matter (Ashton et al., 1999) and in agreement with Bosire et al. (2005) and Middleton and McKee (2001). P concentrations in the R. mangle leaves decreased 26% in relation to original concentrations during the leaching phase. After 15 days concentrations increased substantially for P and N (Fig. 3). Nielsen and Andersen (2003) found the same behavior for P during decomposition
Leaf
Branch
Reproductive Parts
1
0.5
Jul
Jun
May
Apr
Mar
Feb
Jan
Dec
Nov
Oct
Sep
0
Ago
Litterfall (t ha-1 month-1)
1.5
Time 97
98
Fig. 2. Seasonal litter fall rate of the structures of R. mangle trees in the Potengi estuary.
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
531
Table 4 Loss of 50% dry weight in leaves of red mangrove Site
Species
Days
Author
Kenya Thailand USA USA USA Malaysia Belize Panama Brazil Brazil Brazil
R. R. R. R. R. R. R. R. R. R. R.
14a–35b 40 180 45 60 10 35 27 77 23a–45b 32
Bosire et al. (2005) Boonruang (1978) Davis et al. (2003) Heald (1971) Fell (1974) Ashton et al. (1999) Middleton and McKee (2001) D’croz et al. (1989) Silva (1988)
a b
mucronata apiculata mangle mangle mangle apiculata mangle mangle mangle mangle mangle
This work
Wet season. Dry season.
of Rhizophora apiculata leaves in sediments. Bosire et al. (2005) recorded an increase of N concentrations throughout the sampling period (6 weeks) in all mangrove species during their experiment. The high concentrations of P (2.0 mg g1) and N (10.4 mg g1) in 90-day leaf detritus found in our study in R. mangle leaves were higher than those values during the initial 15 days (Fig. 3). The increase in N concentration in red mangrove leaves is probably due to a progressive reduction in the amount of organic C and immobilization of N by fungi and bacteria (Davis et al., 2003; Bosire et al., 2005). 3.4. Standing stock of leaf litter and residence time of leaves on sediments There was a significant change (Tukey, p < 0.05) in the standing stock of leaves (dry weight) on the surface forest floor over the 6 months measured (0.2 ± 0.1–1.0 ± 0.4 t ha1). August (1997) presented the highest values 1.0 ± 0.4 t ha1. This highest biomass of leaves on the floor does not agree with the peak of leaf fall (December 1997, Fig. 3), nor with the smallest frequency of sediment flooding along the year (November 1997) (9.0%). Perhaps a relative lack of activity by crabs, that can break down and remove litter from the forest floor (Middleton and McKee, 2001; Ashton, 2002), is responsible for this highest biomass of leaves on the floor in August 1997. The mean standing stock of leaf litter was 0.4 t ha1. Therefore the estimated mean residence time was 15 days (Silva et al., 2006), assuming a mean annual leaf litter fall rate of 9.5 t ha1 yr1. 11
2.2
10 9 8
1.8
7 1.6
6
N (mgg-1)
P (mg g-1)
2.0
5
1.4
4 1.2 1.0
3 0
15
30
45 Days
60
75
90
2
Fig. 3. Temporal changes in the average concentrations of P and N during leaf decomposition of R. mangle. Intervals.
532
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
Mangrove litter residence time is quite variable depending mostly on the tidal range (Ong et al., 1982; Twilley, 1985), the elevation of the forest relative to inundation (Twilley et al., 1986) and the activity of litter-feeding organisms in particular of seasarmid crabs (Robertson, 1986). Twilley et al. (1986) reported a residence time of 2–6 months in a basin forest. This long residence time is due to infrequent flooding of a basin forest, while our forest is a fringe type which is flooded nearly twice every day. In riverine Rhizophora forests, Ong et al. (1982) found residence time of leaves of about 30 days. Robertson (1986) showed that crabs can bury over 75% of total litter fall at low tide, speeding the ‘‘export’’ of deposited leaves. In our forest, the activity of seasarmid crabs, evaluated visually, is very high and may contribute to the short residence time of leaves on the sediment. The turnover rate was obtained by a ratio of the N and P in the biomass of R. mangle trees to the average transfer of the element through yearly leaf fall. The turnover rates of P and N in this mangrove forest were calculated to be 12 years for P and 7 years for N. The more rapid cycling of N than P may be due to the higher availability of N (Prescott et al., 2000). Our data are similar to those of Li (1997) whose values were 10 years for P and 7 years for N. On the other hand our forest has a smaller (four times) N turnover rate than that found by Gong and Ong (1990) for R. apiculata. Turnover rate is influenced by the biomass of trees, litter fall, and the concentrations of nutrients in the leaves and in the sediment. 3.5. Export of P and N Associated with leaf litter The export of N and P from the mangrove forest to coastal marine systems through leaf litter is a function of the amount of leaf litter exported and its nutrient concentrations. Nutrient concentrations depend on the residence time of leaf litter on the forest floor and the enrichment of nutrients during leaf litter decomposition. The short residence time (15 days) of deposited leaf litter on the mud surface in our forest limits the enrichment of N and P concentrations in decomposing leaves exported, since the nutrient concentrations in this study increased only after at least 30 days of decomposition. Therefore, the concentration of P and N in leaves during export is equal to or lower than the concentration in recently fallen leaves. The estimated average export rates of P and N through leaf litter are 0.5 kg ha1 yr1 and 1.6 kg ha1 yr1 respectively, for the two elements (Fig. 4).
Fig. 4. Dynamics of N and P in the Potengi experimental forest (PEF).
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
533
4. Conclusions The dynamics of N and P in the red mangrove forest studied here is summarized in Fig. 4. Despite the main reservoir of P and N being the sediment, the R. mangle trees play an important role in the cycling of P and N in the Potengi estuary. Decomposing leaf litter does not contribute significantly with nutrient output to estuarine waters since the leaf litter only becomes enriched (P – 1.3 times and N – 4.0 times from initial concentrations) in P and N at the end of the leaf decomposition process (90 days). The short residence time (15 days) makes it clear that the budget of N and P in the leaves is retained in the mangrove forest. The export of nutrients through R. mangle leaf fall is small; these fluxes represent 3% for P and only 0.6% for N of the nutrient stock in the sediment, or 26 and 149 years of growth for P and N. The annual export of nutrients as leaf macrodetritus produced by R. mangle trees is equivalent to 0.1% and 0.03% of the total stock of P and N in the mangrove forest (biomass and sediment). On the other hand, nutrient export from this mangrove in suspended materials phase and the dynamics of groundwater fluxes into the estuary, which was not studied here, could contribute significantly to nutrient export out of the mangrove ecosystem into the estuary. Tropical mangrove forests are probably efficient biogeochemical barriers to the transport of P and N in tropical coastal areas. Hence mangroves could be an important and efficient ally against eutrophication in tropical coastal areas. Acknowledgements The authors wish to acknowledge the International Foundation of Science (IFS-D/2382-1) and Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico (CNPq, Process 100515/1994-0) of Brazil for grant support. The first author is grateful to Professor Philip S. Rainbow (Head of Department of Zoology – The Natural History Museum of London) and Professor Sternberg L.S.L. (Department of Biology, University of Miami) for English corrections. References Amarasinghe, M.D., Balasubramaniam, S., 1992. Net primary productivity of two mangrove forest stands on the northwestern coast of Sri Lanka. Hydrobiologia 247, 37–47. Aragon, G.T., 1997. Biogeoquı´mica sedimentar do ferro e do enxofre em um manguezal da Baı´a de Sepetiba. Ph.D. Thesis. Neoformac¸a˜o de sulfetos ferrosos, Universidade Federal Fluminense, Nitero´i, RJ, 119p. Ashton, E.C., Hogarth, P.J., Ormond, R., 1999. Breakdown of mangrove leaf litter in a managed mangrove forest in Peninsular Malaysia. Hydrobiologia 413, 77–88. Ashton, E.C., 2002. Mangrove sesarmid crab feeding experiments in Peninsular Malaysia. Journal of Experimental Marine Biology and Ecology 273, 97–119. Boonruang, G.P., 1978. The degradation rates of mangrove leaves of Rhizophora apiculata (Bl.) and Avicennia marina (Forsk.) Vierh. at Phuket Island, Thailand. Phuket Marine Biological Centre 26, 1–7. Bosire, J.O., Dahdouh-Guebas, F., Kairo, J.G., Kazungu, J., Dehairs, F., Koedam, N., 2005. Litter degradation and CN dynamics in reforested mangrove plantations at Gazi Bay, Kenya. Biological Conservation 126, 287–295. Bouillon, S., Koedam, N., Raman, A.V., Dehairs, F., 2002. Primary producers sustaining macro-invertebrate communities in intertidal mangrove forests. Oecologia 130, 441–448. Caraco, N.F., Cole, J.J., Likens, G.E., 1989. Evidence for sulphate controlled phosphorus release from sediments of aquatic systems. Nature 341, 316–318. Citron, G., Schaefer-Novelli, Y., 1983. Introduccio´n a la ecologı´a del manglar. UNESCO/ROSTLAC, Montevideo, p. 109. Davis III, S.E., Molina, C.C., Childers, D.L., Day Jr., J.W., 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 73, 199–215. D’croz, L., Delrosario, J., Holness, R., 1989. Mangrove (Rhizophora mangle L.) leaves in the Bay of Panama. Revista de Biologia Tropical 37, 101–104. Fabre, A., Fromard, Fr., Trichon, V., 1999. Fractionation of phosphate in sediments of four representative mangrove stages (French Guiana). Hydrobiologia 392, 13–19. Fell, J.W., 1974. Microbial activities in the mangrove (Rhizophora mangle) leaf detrital system. In: Proc. Int. Symp. Biol. Management of Mangrove, pp. 661–679. Ferreira, M.F., Chiu, W.S., Cheok, H.K., Cheang, F., Sun, W., 1996. Accumulation of nutrients and heavy metals in surface sediments near Macao. Marine Pollution Bulletin 32 (5), 420–426.
534
C.A. Ramos e Silva et al. / Marine Environmental Research 64 (2007) 524–534
Gong, W.K., Ong, J.E., 1990. Plant biomass and nutrient flux in a managed mangrove forest in Malaysia. Estuarine Coastal and Shelf Science 31, 519–530. Grasshoff, K., Ehrhardt, M., Kremling, K., 1983. Methods of Seawater Analysis. Verlag Chimie, Nurnberg, 419 p. Heald, E.J., 1971. The production of organic detritus in a south Florida estuary. Sea Grant Technical Bulletin No. 6. University of Miami, Coral Gables, FL, USA. Kennish, M.J., 1991. Ecology of Estuaries: Anthropogenic Effects. CRC, FL, 494 p. Kurmies, B., 1972. Zur Fraktionierung der Bodenphosphate. Phosphosa¨ure 29, 118–149. Lacerda, L.D., Conde, J.E., Alarco´n, C., Bodero, A., Bacon, P., Alvarez-Leo´n, R., D’croz, L., Kjerfve, B., Olanı´a, J., Vannucci, M., 1993. Mangroves of Latin America and the Caribbean, 1–42. In: Lacerda, L.D. (Ed.), Conservation and Sustainable Utilization of Mangroves Forests in Latin America and Africa. Part I. Latin America. International Society for Mangrove Ecosystems, ISME Tech. Rep. no. 2 Okinawa. Leeg, J.O., Black, C.A., 1955. Determination of organic phosphorus in sediments: II ignition method. Sediment Science Society Proceedings 19, 139–143. Li, M.S., 1997. Nutrient dynamics of a Futian mangrove forest in Shenzhen, south China. Estuarine, Coastal and Shelf Science 45, 463– 472. Middleton, B.A., McKee, K.L., 2001. Degradation of mangrove tissues and implications for peat formation in Belizean island forests. Journal of Ecology 89, 818–828. Nielsen, T., Andersen, F.ø., 2003. Phosphorus dynamics during decomposition of mangrove (Rhizophora apiculata) leaves in sediments. Journal of Experimental Marine Biology and Ecology 293, 73–88. Nogueira, A.R.A., Souza, G.B., Batista, L.A.R., 1992. Determinac¸a˜o espectrofotome´trica de nitrogeˆnio em digeridos de plantas em sistema de ana´lise por injec¸a˜o em fluxo. Quı´mica Nova 19 (1), 33–36. Nriagu, J.O., 1976. Phosphate–clay mineral relations in soils and sediments. Canadian Journal of Earth Sciences 13, 717–736. Olson, J.S., 1963. Energy storage and the balance of producers and decomposers in ecological systems. Ecology 42 (2), 322–330. Ong, J.E., Gong, W.K., Wong, C.H., 1982. Productivity and nutrient status of litter in a managed mangrove forest in Malaysia. In: Kostermans, A.V., Sastroutomo, S.S. (Eds.), Proc. Symp. Bangor, Indonesia, Biotropica, vol. 17, pp. 33–41. Por, E.D., 1984. The ecosystem of the mangal: general considerationsHydrobiology of the Mangal. Junk Publishers, pp. 1–15. Prescott, C.E., Chappell, H.N., Vesterdal, L., 2000. Nitrogen turnover in forest floors of coastal Douglas-Fir at sites differing in soil nitrogen capital. Ecology 81 (7), 1878–1886. Robertson, A.I., 1986. Leaf-burying crabs: their influence on energy flow and transport from mixed mangrove forests (Rhizophora spp.) in northeastern Australia. Journal of Experimental Marine Biology and Ecology 102, 237–248. Silva, C.A.R., Lacerda, L.D., Rezende, C.E., 1990. Heavy metal reservoirs in a red mangrove forest. Biotropica 22 (4), 339–345. Silva, L.F.F., Lacerda, L.D., Ovalle, A.R.C., Rezende, C.E., Silva, C.A.R., Carvalho, C.E.V., 1993. Dinaˆmica de macrodetritos em um ecossistema de manguezal, Baı´a de Sepetiba, RJ. Ann III Symp. Ecossistemas da Costa Brasileira, Serra Negra. Academia de Cieˆncias do Estado de Sa˜o Paulo, vol. 1, pp. 204–211. Silva, C.A.R., 1996. Manguezal: Ecossistema egoı´sta ou benevolente? Cieˆncia Hoje 20 (120), 6–11. Silva, C.A.R., Mozeto, A.A., 1997. Release and retention of phosphorus in mangrove sediments: Sepetiba bay, Brazil. Mangrove Ecosystem Studies in Latin America and Africa, UNESCO. Silva, C.A.R., Rego, R.D.P., 1997. Aspectos estruturais e distribuic¸a˜o de biomassa em um mangue vermelho (Rhizophora mangle), Natal, RN. In: Anais do VIII Semina´rio Regional de Ecologia, VIII, pp. 285–293. Silva, C.A.R., 1988. Distribuic¸a˜o e ciclagem interna de metais pesados em um ecossitema de manguezal dominado por Rhizophora mangle, Baı´a de Sepetiba, Rio de Janeiro. M.Sc. Thesis, Universidade Federal Fluminense, Nitero´i, 67p. Silva, C.A.R., Sampaio, L.S., 1998. Speciation of phosphorus in a tidal floodplain forest in the Amazon estuary. Mangroves and Salt Marshes 02 (01), 59–62. Silva, C.A.R., Mozeto, A.A., Ovalle, A.R.C., 1998. Distribution and fluxes as macrodetritus of phosphorus in red mangroves, Sepetiba Bay, Brazil. Mangroves and Salt Marshes 02 (01), 37–42. Silva, C.A.R., Silva, A.P., Oliveira, S.R., 2006. Concentration, stock and transport rate of heavy metals in a tropical red mangrove, NatalBrazil. Marine Chemistry (99), 2–11. ´ reas de Manguezal do Estua´rio Potengi/RN, Nordeste, Brasil. M.Sc. Thesis. Dissertac¸a˜o, Souza, F.E.S., 1999. Estudo de Valorac¸a˜o das A Universidade Federal do Rio Grande do Norte, Natal, 87p. Tam, N.F.Y., Wong, Y.S., 1995. Mangrove sediments as sinks for wastewater-borne pollutants. Hydrobiologia 295, 231–241. Twilley, R.R., 1985. The exchange of organic carbon in basin mangrove forests in a southwest Florida estuary. Estuarine, Coastal and Shelf Science 20, 543–557. Twilley, R.R., Lugo, A.E., Patterson- Zucca, C., 1986. Litter production and turnover in basin mangrove forests in southwest Florida. Ecology 67, 670–683. Wafar, S., Untawale, A.G., Wafar, M., 1997. Litter fall and energy flux in mangrove ecosystem. Estuarine, Coastal and Shelf Science 44, 111–124.