Introduction

The high rate of litter production and decomposition combined with the remarkable efficiency in nutrient recycling make mangrove forests one of the most productive ecosystems on the earth (Qasim and Wafar 1990; Kalio 1992; Harrison et al. 1994; Day et al. 1996; Duarte and Cebrian 1996; Kathiresan and Bingham 2001; Bosire et al. 2005; Morrisey et al. 2007; Alongi 2009). Litter constitutes up to one third of the primary production in mangrove forests (Alogi et al. 1992). Nutrients released from litter through decomposition are either retained in the ecosystem to support secondary production or exported to nearby coastal waters by the tides (Alongi et al. 1998). This process is, therefore, the most important source of carbon and nutrients in biogeochemical cycles (Odum and Heald 1975; Alongi et al. 1992; Robertson et al. 1992; Wafar et al. 1997) and thus, vital to both mangrove and adjacent coastal ecosystems. For these reasons, litter production is considered as an indicator of net primary production (NPP) and biogeochemical cycles in mangrove ecosystems.

Litter produced in mangrove forests normally contains leaves, flowers, fruits, bark, needles and twigs (Odum and Heald 1975; Slim et al. 1996; Betoulle et al. 2001). Of them, leaves constitute the major fraction accounting to more than 50% of the total litter production (Wafar et al. 1997). The high productivity of mangrove ecosystem is, thus, attributed to higher rate of leaf production and their rapid decomposition (Park and Kang-Hyun 2003; Mahmood et al. 2011). Leaf litter also plays a major role in enhancement of soil fertility and supply of nutrient materials to animal life in mangrove ecosystems (Srivastava 1980). Hence, leaf litter production as well as decomposition is the key to the nutrient cycling in mangrove sediments (Sherman 2002; Ngoran et al. 2006; Mahmood and Hoque 2008; Mahmood et al. 2009; Triadiati et al. 2011). An assessment of leaf litter production and decomposition indicates the productivity of a mangrove ecosystem (Pool et al. 1975; Snedaker and Snedaker 1984). It also signals the nutrient status of the surrounding coastal ecosystems (Pool et al. 1986). Comparatively, a higher rate of litter production has been observed in tropical mangroves (Mahmood and Hoque 2008).

Production of leaf litter exhibits seasonal variation (Duke 1999; May 1999; Mfilinge et al. 2005a) and is influenced by several factors, viz.: geographical location (Saenger and Snedaker 1993), nutrient concentrations, topography (Cox and Allen 1999), rainfall (Wangondu et al. 2014), temperature (Mchenga and Ali 2017), salinity (Day et al. 1996), solar radiation (Saenger and Snedaker 1993), winds (Ghosh and Banerjee 2013), forest types (Pool et al. 1975), stand structure (Saberi 1989), freshwater drainage (Feller et al. 1999), species composition (Coupland et al. 2005) and anthropogenic influence (Silva et al. 1998). More frequent tides and river inundations induce riverine mangroves to produce higher amount of leaf litter (Pool et al. 1975; Twilley 1995). Decomposition rate of mangrove leaves is influenced by the abundance and diversity of macro benthic fauna and microorganisms (Meentemeyer 1978; Chapin et al. 2002), quality of decomposing substrate (Bosire et al. 2005), physico-chemical nature of the site (Wafar et al. 1997), climatic conditions, particularly the temperature and salinity (Mackey and Smail 1996; Lorıa-Naranjo et al. 2019), tides (Middleton and McKee 2001; Bosire et al. 2005; Imgraben and Dittmann 2008) and hydrocarbon pollution (Frick et al. 1999).

No works exist on litter production and decomposition on Indian mangroves except that of Wafar et al. (1997) on litter decomposition in the Madovi–Zuari Estuaries on the Central West Coast of India, Ghosh and Banerjee (2013) on inter relationship between physico-chemical variables and litter production on mangroves in Sundarbans, Rani et al. (2016) on assessment of NPP of Cochin mangroves and Suresh et al. (2017) on estimation of total carbon stock in Indian mangroves. They correlated litter production positively with temperature, wind velocity, salinity and species diversity but negatively with rainfall. However, the significance of leaf litter decomposition in carbon sequestration and thereby its contribution to NPP was not investigated in Indian context despite the fact that the country harbors 4,952 km2 of mangrove forests (India State of Forest Report 2021) with 57% of true mangrove species in the world (Ragavan et al. 2016).

Materials and methods

Study area

Kunhimangalam mangrove forest, located between 12° 03′−12° 06′ N and 75° 12′ − 75° 14′ E, was our study area (Fig. 1). It extends to 5 km2 along the banks of Perumba and Pullankodu Rivers. It is the largest and least disturbed mangrove forest in Kerala. It occurs in narrow strips with a width of 50–100 m along the water line and is micro-tidal with a tidal amplitude of one meter. The forest has 10 true mangrove species and 11 mangrove associates. Acanthus ilicifolius L., Aegiceras corniculatum (L.) Blanco., Avicennia officinalis L., Excoecaria agallocha L. and Rhizophora mucronata Lam. are dominant with representations of Avicennia marina (Forssk.) Vierh., Bruguiera cylindrica (L.) Blume, Kandelia candel (L.) Druce, Rhizophora apiculata Blume and Sonneratia caseolaris (L.) Engl. There are three types of plant associations, viz: (i) dominated by A. officinalis, A. marina, E. agallocha and A. corniculatum in the upper intertidal area, and R. mucronata, R. apiculata and K. candel in the lower intertidal area, (ii) dominated by A. officinalis and A. corniculatum in the upper intertidal area, and R. mucronata in the lower intertidal area, and (iii) dominated by A. officinalis and E. agallocha in the upper intertidal area, and A. ilicifolius in the lower intertidal area. A well defined mid-intertidal area cannot be discerned in Kunhimangalam mangrove forest. It harbors 60 species of birds, 42 species of fishes, 14 species of mollusks and 15 species of crabs (Praveen et al. 2015). Kunhimangalam experiences tropical warm and humid climate with pronounced monsoon (June to October), winter (November to February) and summer (March to May).

Fig. 1
figure 1

Location map of Kunhimangalam mangrove

Full size image

We focused our study to (i) quantify the leaf litter production in Kunhimangalam mangrove forest and analyze the environmental factors influencing litter production, (ii) measure the rate of leaf litter decomposition, (iii) calculate the carbon (C) content in leaf litter and (iv) assess the relative share of carbon content in leaf litter to NPP.

Study design

We conducted this study in three stages: (i) assessment of the rate of leaf litter production for a period of two years by litter trap method and the influence of environmental factors on leaf litter production, (ii) estimation of organic carbon in leaf litter to understand its share in NPP of the forest and (iii) assessment of the rate of microbial decomposition on the leaf litter of Aegiceras corniculatum, Avicennia officinalis, Excoecaria agallocha and Rhizophora mucronata, the four major mangrove species in Kunhimangalam, by litter bag method. These four species together constituted 92.49% of abundance in the forest (Praveen 2014). The decomposition rate is expressed as percentage loss of initial dry weight per day. The dry weight corresponding to the fresh weight of the litter weighed was estimated using a predetermined fresh weight–dry weight ratio to express all measurements in dry weight.

Quantification of leaf litter production

Quantification of litter production was done using litter trap method (Brown 1984). Twenty litter traps, each made of nylon net (mesh size 1 mm2) and spread over 1 m2 PVC frames (Fig. 2a), were randomly placed in the forest. They were fixed on poles at one meter height above the high tide mark. Hanging traps were also used (Fig. 2b). Trap surface was made in conical shape to facilitate accumulation of litter between collection intervals of one week. Collected litter was sorted as buds, flowers, fruits, twigs and leaves. It was quantified in dry weight. Leaves of each species were segregated and weighed separately. This quantification provided total leaf litter production against total litter production in Kunhimangalam forest day−1 m−2 in dry weight. Litter was collected for a period of two years from August 2008 to July 2010 ( Shanij et al. 2016).

Environmental parameters

Water salinity, water pH, water temperature, soil temperature, soil pH and air temperature were recorded monthly using Erma refractometer Dual scale 0–100 ppt of salinity, Safeseed PHMTR001, mercury thermometer (for water, soil and air), Takemura soil pH meter and moisture tester DM-13 respectively. Data on humidity and rain fall were obtained from Krishi Vigyan Kendra of Kerala Agricultural University at Thaliparamba, about 15 km from Kunhimangalam.

Estimation of organic carbon

We collected freshly fallen senescent leaves from the mangrove forest floor and used for organic carbon estimation. Three samples of 100 g senescent leaves from each of the four species were collected in each season. They were washed with distilled water, oven dried at 60 °C to constant weight, grounded in a mill to pass a 1 mm sieve and analyzed for organic carbon using Walkley and Black method (Jackson 1973). The average value of carbon content in three samples of each species in each season was calculated. The average carbon content of three seasons with respect to the leaf litter of these four species was considered the mean carbon content of leaf litter in the mangrove forest. These data were converted to NPP following Twilley et al. (1992).

Measurement of leaf litter decomposition

We assessed the in situ decomposition rate of the leaf litter employing litter bag method (Harmon et al. 1999). Nylon net of mesh size 1 mm2 was used. The decomposition rate of leaf litter in four mangrove species separately and that in four species mixed were assessed. Thus, altogether five sets of litter bags, each set with 10 replicates were used. Each litter bag carried 50 g freshly fallen senescent leaves. Litter bags were placed randomly in the intertidal zone tied to nearby mangrove trees (Fig. 2c) and allowed to decompose without the influence of any macrobenthos (Fig. 2d, e and f). We inspected the litter bags and weighed the contents weekly, after gently washing off the sediments deposited on them in tidal water. The dry weight corresponding to the wet weight of leaf litter was estimated through predetermined fresh weight - dry weight ratio. This process was continued until the litter was fully decomposed. The loss of weight was plotted against time and decomposition rate of leaf litter was expressed as weight loss in grams day−1.

Fig. 2
figure 2

Arrangements made for litter collection and leaf litter decomposition experiments in Kunhimangalam A A litter trap established on the mangrove forest floor B A hanging litter trap, C A litter bag deployed in the intertidal region, exposed during low tide D The condition of leaf litter at the beginning of decomposition experiment E The condition of leaf litter after 70-80% decomposition during the experiment F Leaf litter at the end of the decomposition experiment

Full size image

Calculation of fresh weight-dry weight ratio

Freshly fallen leaf litter weighing 5 g, 10 g, 15 g, 20 g, 25 and 30 g was taken from the forest floor and oven dried to constant weight at 70 °C. Their dry weight was plotted on a graph against corresponding fresh weight. The graph provided the dry weight corresponding to the fresh weight of leaf litter.

Statistical analysis

One-way ANOVA was done to find out the variation in monthly leaf litter production, rates of leaf litter production and leaf litter decomposition. Results were considered significant at p ≤ 0.05. We fitted both linear and exponential decay models to the species wise decomposition data to assess the trend of leaf litter decomposition. Model parameters were estimated using least square method and presented in a table format. Monthly average values recorded for rates in leaf litter production, and environmental parameters were analyzed separately using simple linear regression models to identify the dependence behavior between the variables. The fitted model performances were analyzed using R2 values.

Results

Leaf litter production

Of the total 3.12 ± 0.18 g m−2 day−1 (11.39 ± 0.66 t ha−1 yr−1) of litter produced (Table 1), leaf litter contributed to 78% (8.83 ± 0.95t ha−1 yr−1). Twigs, fruits, flowers and buds contributed the remaining 22% (Fig. 3). The mean daily production of leaf litter ranged from 1.36 ± 0.02 to 3.93 ± 0.65 g m−2 day−1 (Table 1), the highest was in November followed by April and December. August recorded the lowest followed by July. This variation was significant in ANOVA (F = 7.641, p < 0.01). The pattern of leaf litter production was the same in both the years, 2008–2009 and 2009–2010 (Fig. 4). Season wise, the highest leaf litter production was recorded in winter and the lowest in monsoon (Fig. 5). Major contributions came from Aegiceras corniculatum (17%), Avicennia officinalis (27%), Excoecaria agallocha (18%) and Rhizophora mucronata (22%) (Fig. 6). The rest 16% was contributed by Avicennia marina (4%), Bruguiera cylindrica (3%), Rhizophora apiculata (3%), Kandelia candel (3%) and others (3%).

Fig. 3
figure 3

Percentage composition of different components towards total litter production in Kunhimangalam mangrove forest

Full size image
Fig. 4
figure 4

Pattern of monthly leaf litter production rates in Kunhimangalam mangrove forest from 2008 to 2010

Full size image
Fig. 5
figure 5

Source: Shanij et al. 2016

Seasonal variation in the leaf litter production rates in Kunhimangalam mangrove forest from 2008 to 2010.

Full size image
Fig. 6
figure 6

Source: Shanij et al. 2016

Percentage composition of different mangrove species towards total leaf litter production in Kunhimangalam mangrove forest.

Full size image
Table 1 Average monthly litter and leaf litter production rates in dry weight (g m−2 day−1) in Kunhimangalam mangrove forest from 2008 to 2010
Full size table

The salinity ranged from 0 ppt (monsoon) to 28 ppt (summer), soil pH from 5.17 (summer) to 5.69 (monsoon), water pH from 5.85 (monsoon) to 7.21 (summer), soil temperature from 23.7 °C (winter) to 27.7 °C (monsoon), water temperature from 25 °C (monsoon) to 28.2 in (summer), air temperature from 25 °C (winter) to 30.4 °C (summer), rainfall from 0 (winter) to 27 cm (monsoon) and Relative Humidity from 84% (summer) to 91% (monsoon).

Organic carbon content in leaf litter

We estimated the average carbon content in leaf litter at 40.14 ± 1.38%. There were no significant seasonal differences in carbon content in leaf litters collected from the four species, though they exhibited significant variation in C content. Aegiceras corniculatum possessed the highest C content (44.62 ± 1.23%) followed by Avicennia officinalis (43.52 ± 0.59%) and Rhizophora mucronata (40.79 ± 1.64%). The least C content was in Excoecaria agallocha (31.63 ± 3.08%). The relative share of leaf litter from these four species to the NPP of Kunhimangalam mangrove forest was calculated at 3.56 ± 0.01 t C ha−1 y−1.

Leaf litter decomposition

Average decomposition rates of leaf litter expressed as percentage of initial dry weight remaining are presented in Figs. 7, 8, 9, 10 and 11. The average rate of decomposition was 0.07 ± 0.01 g d−1 for Aegiceras corniculatum, 0.10 ± 0.01 g d−1 for Avicennia officinalis, 0.11 ± 0.02 g d−1 for Excoecaria agallocha and 0.13 ± 0.02 g d−1 for Rhizophora mucronata. For mixed leaf litter category, it was 0.14 ± 0.04 g d −1 which was 0.62% of 22.5 g dry weight of leaf litter (50 g in fresh weight) deployed in the litterbag (Table 2). Decomposition rates varied significantly (F = 2.536, p < 0.05) among the five categories. The leaf litter of A. corniculatum took 46 weeks, A. officinalis 31 weeks, E. agallocha 26 weeks, R. mucronata 31 weeks and the mixed one 25 weeks to attain 100% decomposition.

Fig. 7
figure 7

Graph showing the fitted models of the decomposition data of mixed leaf litter

Full size image
Fig. 8
figure 8

Graph showing the fitted models of the decomposition data of R. mucronata leaf litter

Full size image
Fig. 9
figure 9

Graph showing the fitted models of the decomposition data of E. agallocha leaf litter

Full size image
Fig. 10
figure 10

Graph showing the fitted models of the decomposition data of A. corniculatum leaf litter

Full size image
Fig. 11
figure 11

Graph showing the fitted models of the decomposition data of A. officinalis leaf litter

Full size image
Table 2 Leaf litter decomposition rates obtained from the litter bag decomposition study
Full size table

Decomposition data of leaf litter from mixed category and Rhizophora mucronata fitted best to the exponential decay model (Figs. 7 and 8; Table 3). This showed that the rate of decomposition was faster for the mixed category (0.27 ± 0.07 g d−1) and R. mucronata (0.26 ± 0.04 g d−1) at the early phase. Thereafter, it took a slower pace, 0.04 ± 0.01 g d−1 and 0.05 ± 0.01 g d−1 respectively. The pattern of decomposition in Aegiceras corniculatum and Avicennia officinalis best fitted to the linear model (Figs. 9 and 10; Table 3) as leaf litter from these two species lost their weight at a steady pace, 0.07 ± 0.01 g d−1and 0.10 ± 0.01 g d−1, respectively during the process of decomposition. Interestingly, leaf litter from Excoecaria agallocha was fitted almost equal to the linear (R2 = 0.923) as well as the exponential (R2 = 0.926) models. However, in all the five experimental treatments, the ‘y intercept’ of the exponential decay model was significantly higher (Table 3) than that of the linear model. This indicated a faster rate in the initial phase of decomposition.

Table 3 Fitted model parameter and R2 values
Full size table

Discussion

Leaf litter production

The annual global litter production in different mangrove ecosystems ranged between 7 and 15 t ha−1 yr−1 (Hossain and Hoque 2008), though an exceptionally higher litter production was recorded in mangrove forests in Australia (34.4 t ha−1 yr−1) (Alongi et al. 2005). The average annual litter production in Kunhimangalam mangrove forest was 11.39 ± 0.66 t ha−1 (Table 1). Rani et al. (2016) reported a mean annual litter production rate in Cochin mangrove forest as 16.57 ± 6.58 t ha−1, about 300 km south of Kunhimangalam.

While leaves contributed 78% (8.83 ± 0.95 t ha−1 yr−1) of the average annual litter production (11.39 ± 0.66 t ha−1 yr−1) in Kunhimangalam mangrove forest, their share in Cochin mangrove forest was only 53.9% (8.93 ± 3.49 t ha−1 yr−1) (Rani et al. 2016). Both forests are riverine. Studies of Cintro´n and Schaeffer-Novelli (1983), Lacerda et al. (2001) and Zaldivar Jiménezet al. (2004) have shown that leaves contribute a higher fraction to total litter production than twigs, buds and fruits in young mangrove forests. Wind speed enhances the percentage fraction of twigs, buds and fruits in total litter production. There is a significant positive correlation between litter production and wind speed (Mfilinge et al. 2005a; Mchenga and Ali 2017). Cochin mangrove forest experiences a higher wind speed (11–27 km h−1) than Kunhimangalam mangrove forest (3–7 km h−1) as the former occurs very close to the coast, while the latter is an upstream mangrove area about 15 km off the coast.

The quantity of leaf litter production at global level is estimated between 7 and 15 t ha−1 yr−1. The range increases from higher to lower latitudes (Saenger and Snedaker 1993; Komiyama et al. 2008; Bernini and Rezende 2010). Latitude influences leaf litter production in mangrove forests. But it has not influenced leaf litter production in Kunhimangalam (8.88 ± 0.51 t ha−1 yr−1), though the area is at lower latitude. Soil analysis shows the forest is nutrient poor (Shanij 2017).

Production of leaf litter showed a unimodal pattern (Fig. 4). It was the highest in winter (November to February) and the lowest in monsoon (June to October). Aegiceras corniculatum, Avicennia officinalis and Rhizophora mucronata produced maximum leaf litter in winter. They are evergreen species. Evergreen trees normally shed more leaves before they enter into the reproductive phase. Cumulative effect of stress due to high water salinity, low precipitation, high evapo-transpiration, low contents of nutrients because of reduced fresh water runoff and emergence of new leaves also tend to produce more leaf litter in winter (Ochieng and Erftemeijer 2002; Kamruzzaman et al. 2019). Excoecaria agallocha, a semi-deciduous species, contributed maximum leaf litter in summer (March to May).

Leaf litter production in Kunhimangalam is negatively correlated with soil pH and rainfall in linear regression analysis (Figs. 12 and 13). Temperature, salinity and humidity did not show any remarkable influence on leaf litter production. Rani et al. (2016) reported negative correlation between rainfall and leaf litter production. Clough (1992), Twilley et al. (1997), Wafar et al. (1997), Sherman et al. (2003), Arreola-Lizarraga et al. (2004) and Aké-Castilho et al. (2006) have reported salinity, temperature and rainfall showing positive correlation with leaf litter production in mangrove forests. However, Lopez-Portilho and Ezcurra (1985) and Silva et al. (2006) did not find any such correlation. The above studies indicate that geomorphic and physico-chemical characteristics do not show a uniform pattern of influence on total leaf litter production in mangrove forests.

Fig. 12
figure 12

Influence of soil pH on litter fall in Kunhimangalam mangrove forest

Full size image
Fig. 13
figure 13

Influence of rainfall on litter fall in Kunhimangalam mangrove forest

Full size image

Leaf litter decomposition

The activity of decomposers, physico-chemical traits of litter and environmental factors influence leaf litter decomposition (Anderson and Swift 1983; Heal et al. 1997; Kavvadias et al. 2001), but the nature of species has more influence on the decomposition rate (Salinas et al. 2011). The leaf litter of Sonneratia alba decomposed faster than that of Rhizophora apiculata, R. mucronata and Bruguiera parviflora in a Malasian mangrove forest (Ashton et al.1999). In mangrove forests of Maputo Bay, Mozambique, the leaf litter of Avicennia marina decomposed faster than that of R. mucronata. The rates of leaf litter decomposition varied from species to species in our experiments. Variations in the rate of decomposition are ascribed to differences in morphology, texture and chemical composition of mangrove leaves (Ashton et al. 1999; Fernando and Bandeira 2009).

Leaf litter decomposition in mangrove forests normally involves two phases: (i) a rapid weight loss phase through leaching and (ii) a breakdown phase by the activities of decomposers (Polunin 1982). This pattern of decomposition occurred in treatments of mixed, Rhizophora mucronata and Excoecaria agallocha categories in our experiments where the decomposition data fitted best to the exponential decay model (Figs. 7, 8 and 9). The initial rapid weight loss is attributed to the high quantity of water soluble phenolic compounds, flavonoids and non-lignified carbohydrates present in the leaves. These compounds leach faster (Mason 1977; Cundell et al. 1979; Mahmood et al. 2007; Ibrahima et al. 2008, 2010; Simali and Roy 2012). The low level of non-leachable structural compounds like lignin, tannin, cellulose etc. in E. agallocha and R. mucronata might have equally contributed to the weight loss. (see Van der Valk and Attiwill 1984; Mfilinge et al. 2005b). The high quantity of non-leachable structural compounds like lignin, tannin, cellulose etc. present in the leaf litter of Avicennia officinalis and Aegiceras corniculatum made the rate of decomposition slower (see Ardon and Pringle 2008; Alvim et al. 2015) and so, their decomposition rate fitted best to the linear model. The decomposition in litter bags occurs in a micro-climatic condition different from that in a natural environment (Boulton and Boon 1991).

Low C/N ratio favored decomposition of mangrove leaf litter (Twilley et al. 1997; Edu et al. 2014) during the break down phase. Low C/N ratio in the leaf litter of Excoecaria agallocha marked high nitrogen concentration and resultant high nutritional value. Both of them promoted faster microbial colonization and decomposition. However, presence of tannin slowed down the decomposition rate, although the C/N ratio in this species is lower than that of Avicennia officinalis and Rhizophora mucronata (Rani et al. 2023). Leaf blades of Aegiceras corniculatum are thinner than those of A. officinalis and R. mucronata but presence of tannin in this species inhibited faster microbial activity and delayed the decomposition process (Cundell et al. 1979; Coen 1988; Steinke et al. 1993; Edu et al. 2014).

Assorted leaf litters accumulated from different species in mangrove forests decomposed faster, when there was remarkable leaching towards the first phase or leaf litters contained leaves of species with low C/N ratio (Mall et al. 1991; Fyles and Fyles 1993; McArthur et al. 1994; Briones and Ineson 1996; McTiernan et al. 1997; Wardle et al. 2003). In our experiments, the rate of leaf litter decomposition was the highest for mixed category. Leaf litters of Excoecaria agallocha and Rhizophora mucronata leached fast during their first phase of decomposition. E. agallocha possessed low C/N ratio. These combinations created a conducive chemical atmosphere for triggering early microbial activities in mixed leaf litter category and enhanced the rate of decomposition. The high nitrogen content in E. agallocha promoted the rate of decomposition. Nutrients released from the fast decomposing E. agallocha, Avicennia officinalis and R. mucronata accelerated action of microbes on the leaves of Aegiceras corniculatum, subduing the effect of high tannin content. The litter of E. agallocha, the fastest decomposing material among the four, quickened microbial actions in the mixed leaf litter category much faster and earlier than the other categories constituted separately by A. officinalis, R. mucronata and A. corniculatum.

Net primary productivity

Primary productivity of Avicennia officinalis, Aegiceras corniculatum. Excoecaria agallocha and Rhizophora mucronata by way of leaf litter in Kunhimangalam mangrove forest was estimated at 3.56 ± 0.01 t C ha−1 y−1. In Cochin mangrove forest, Rani et al. (2016) estimated the litter share of NPP at 7.12 ± 2.81 t C ha−1 y−1. This difference, though the two mangrove forests are only about 300 km apart, could be due to the fact that they estimated the C content of the litter which included flowers, propagules and twigs apart from leaves. Carbon contents are more in mangrove propagules than in leaf litter. Alongi (2009), who estimated the mean global aboveground mangrove NPP rate at 11.1 t C ha−1 y−1, noted that there was considerable scatter in data owing to variations in tree age, structure and species composition of mangrove forests, besides variations in climatic and other environmental conditions. Estimates have shown that 32% of the Gross Primary Productivity (GPP) in mangrove plants is shunted to NPP of the canopy (Alongi 2014). In Kunhimangalam almost 30% of the NPP was constituted by leaves. So, it is logical to think that our estimates of NPP from leaf litter have rightly reflected the total NPP of Kunhimangalam mangrove forest. This shows its higher potential for carbon sequestration.

The mangrove ecosystem has high potential for carbon assimilation. It is an established fact. However, it is still not clear whether they function as a significant carbon sink in the coastal ocean (Alongi 2014). Less than 1% (0.62%) of the leaf litter only was decomposed in Kunhimangalam mangrove forest when macrobenthos were excluded from the system. The slow rate of leaf litter decomposition indicates that the release of sequestered carbon is very slow. What did happen to the major portion of leaf litter (99%)? The possibilities are: (i) it might have been washed off to coastal waters supplying organic matter to ‘near shore’ consumers (Teal 1962; Odum and Heald 1975; Odum 1980), (ii) entered into higher consumer trophic levels through the detritus food chain (Lee 1997; Nicholson 2009) or (iii) remained on the forest floor, later to be converted to the extensive peat deposit (Middleton and McKee 2001). Quantification of the amount of leaf litter exported to coastal waters, detritus food chain or peat formation is critical to know the entire pathways of carbon. Precisely, not the decomposed leaf litter constituting an insignificant 0.62% but the balance 99.38% determines the potential of Kunhimangalam mangrove forest to act as a source of or sink for carbon. Thus, without taking these factors into account, equating the litter production directly to the carbon sequestration potential of a mangrove forest may lead to erroneous conclusion.

Conclusions

Leaf litter formed the major fraction of total litter produced in Kunhimangalam mangrove forest. Geomorphological and physico-chemical characters of mangrove forests influence leaf litter production. Higher productions of leaf litter in Kunhimangalam mangrove forest shows its higher potential for carbon sequestration. The results of our decomposition study show that it is the destiny of 99% of the leaf litter that determines the capacity of this forest to act as a source or sink for carbon. The role of detritivores like crabs is highly significant in converting leaf litter directly to particulate organic carbon or soil organic carbon by feeding on leaves or shredding them into small pieces suitable for microorganisms to act upon. The efficiency of a mangrove forest as a principal carbon sequestering ecosystem does not depend merely on its primary productivity, but on the delicate balance existing among different ecosystem functions operating directly or indirectly on the litter produced.